05 Sep '18
The package rpms/perdition.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/perdition.git/commit/?id=b5b949f38c….
Change:
+%ifarch s390x armv7hl
Thanks.
Full change:
============
commit b5b949f38c98f651e9db2e5b38aa9dbcb933e043
Author: Pavel Alexeev <pahan(a)hubbitus.info>
Date: Thu Sep 6 00:15:15 2018 +0300
Update to version 2.2
- Update to version 2.2 (request also for Epel: bz#1585846).
- Remov tetex-dvips requirementw to fix FBFS as it is not present anymore in rawhide (bz#1556104).
- Change distribuition tarball to xz archive.
- Switch libidn to libidn2 (bz#1449141).
- Change mysql-devel to mariadb-connector-c-devel (bz#1494071).
- Own /etc/openldap/schema directory (bz#1553431).
- Drop uotdated patch0 perdition-1.19-rc5-const-version.patch.
- Add libnsl2-devel BR.
- Fill bz#1625780 on dwz (<-https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/PB6DOPJSPJLDNUVZ3HUSNLMBORWLCXJK/#KORDFM5B3FXCY3RJ2MYBH5MT4YBNPJBJ)
o And unset _find_debuginfo_dwz_opts for s390x and armv7hl architectures.
diff --git a/.gitignore b/.gitignore
index 5073356..bf6ec12 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,3 @@
/perdition-2.0.tar.bz2
/perdition-2.1.tar.bz2
+/perdition-2.2.tar.xz
diff --git a/perdition.spec b/perdition.spec
index 44550a3..d15a532 100644
--- a/perdition.spec
+++ b/perdition.spec
@@ -1,26 +1,30 @@
Summary: Mail Retrieval Proxy
Name: perdition
-Version: 2.1
-Release: 12%{?dist}
+Version: 2.2
+Release: 1%{?dist}
License: GPLv2+
Group: Applications/Internet
URL: http://horms.net/projects/perdition/
-Source: http://horms.net/projects/%{name}/download/%{version}/perdition-%{version}.…
+Source: http://horms.net/projects/%{name}/download/%{version}/%{name}-%{version}.ta…
# My systemd service template
Source1: %{name}-template.service
BuildRequires: automake autoconf libtool vanessa_logger-devel >= 0.0.10
BuildRequires: vanessa_adt-devel >= 0.0.6 vanessa_socket-devel >= 0.0.7
-BuildRequires: gdbm-devel mysql-devel postgresql-devel openldap-devel
-BuildRequires: popt-devel pam-devel zlib-devel openssl-devel ghostscript
-BuildRequires: tetex-latex tetex-dvips libtiff unixODBC-devel tinycdb-devel
-BuildRequires: libidn-devel libdb-devel
+BuildRequires: mariadb-connector-c-devel postgresql-devel openldap-devel
+BuildRequires: gdbm-devel popt-devel pam-devel zlib-devel openssl-devel
+BuildRequires: ghostscript tetex-latex libtiff unixODBC-devel
+BuildRequires: libidn2-devel libdb-devel tinycdb-devel libnsl2-devel
Requires: vanessa_logger >= 0.0.10 vanessa_adt >= 0.0.6 vanessa_socket >= 0.0.7
-#???
-Patch0: perdition-1.19-rc5-const-version.patch
Requires(post): systemd
Requires(preun): systemd
Requires(postun): systemd
+# By suggestion https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org…
+# See bz#https://bugzilla.redhat.com/show_bug.cgi?id=1625780
+%ifarch s390x armv7hl
+%undefine _find_debuginfo_dwz_opts
+%endif
+
%description
Perdition is a fully featured POP3 and IMAP4 proxy server. It is able to
handle both SSL and non-SSL connections and redirect users to a
@@ -126,8 +130,6 @@ which do not support the systemd unit file format.
%prep
%setup -q -n %{name}-%{version}
-%patch0 -p1 -b .const
-
%build
CFLAGS="${CFLAGS:-%optflags} -I/usr/kerberos/include" ; export CFLAGS
if [ -f configure.in ]; then
@@ -165,7 +167,8 @@ for service in pop3 pop3s imap4 imap4s managesieve; do
install -Dm 644 %{name}-$service.service %{buildroot}/%{_unitdir}/%{name}-$service.service
done
-rm -f %{buildroot}%{_libdir}/*.{a,la,so}
+# debug.#dwz# files strange and lead to faile build on arm and s390 platform. See https://koji.fedoraproject.org/koji/taskinfo?taskID=27400316
+rm -f %{buildroot}%{_libdir}/*.{a,la,so} %{buildroot}%{_libdir}/debug/usr/sbin/%{name}*.debug.#dwz#.*
%clean
@@ -291,7 +294,7 @@ fi
%{_libdir}/lib%{name}db_ldap.so.0.0.0
%{_sbindir}/%{name}db_ldap_makedb
%{_mandir}/man8/%{name}db_ldap_makedb.*
-%config(noreplace) %{_sysconfdir}/openldap/schema/%{name}.schema
+%config(noreplace) %{_sysconfdir}/openldap/schema
%files mysql
%{_libdir}/lib%{name}db_mysql.so.0
@@ -312,6 +315,18 @@ fi
%{_mandir}/man8/%{name}db_odbc_makedb.*
%changelog
+* Wed Sep 05 2018 Pavel Alexeev <Pahan(a)Hubbitus.info> - 2.2-1
+- Update to version 2.2 (request also for Epel: bz#1585846).
+- Remov tetex-dvips requirementw to fix FBFS as it is not present anymore in rawhide (bz#1556104).
+- Change distribuition tarball to xz archive.
+- Switch libidn to libidn2 (bz#1449141).
+- Change mysql-devel to mariadb-connector-c-devel (bz#1494071).
+- Own /etc/openldap/schema directory (bz#1553431).
+- Drop uotdated patch0 perdition-1.19-rc5-const-version.patch.
+- Add libnsl2-devel BR.
+- Fill https://bugzilla.redhat.com/show_bug.cgi?id=1625780 (<-https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/PB6DOPJSPJLDNUVZ3HUSNLMBORWLCXJK/#KORDFM5B3FXCY3RJ2MYBH5MT4YBNPJBJ)
+ o And unset _find_debuginfo_dwz_opts for s390x and armv7hl architectures.
+
* Fri Jul 13 2018 Fedora Release Engineering <releng(a)fedoraproject.org> - 2.1-12
- Rebuilt for https://fedoraproject.org/wiki/Fedora_29_Mass_Rebuild
diff --git a/sources b/sources
index 9a13003..0df1182 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-1e2c5ecfdcca0c02dde831186a077084 perdition-2.1.tar.bz2
+SHA512 (perdition-2.2.tar.xz) = 1ee64505ffe5ad11893fe61a37343fd004d5c15ea88b3dd18e790482248d4b68adc57f2b9c4a1fc12b06087fa7fa01f913db6d1e44bd7e82fd549fd85937a866
1
0
05 Sep '18
The package rpms/perdition.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/perdition.git/commit/?id=b5b949f38c….
Change:
+%ifarch s390x armv7hl
Thanks.
Full change:
============
commit b5b949f38c98f651e9db2e5b38aa9dbcb933e043
Author: Pavel Alexeev <pahan(a)hubbitus.info>
Date: Thu Sep 6 00:15:15 2018 +0300
Update to version 2.2
- Update to version 2.2 (request also for Epel: bz#1585846).
- Remov tetex-dvips requirementw to fix FBFS as it is not present anymore in rawhide (bz#1556104).
- Change distribuition tarball to xz archive.
- Switch libidn to libidn2 (bz#1449141).
- Change mysql-devel to mariadb-connector-c-devel (bz#1494071).
- Own /etc/openldap/schema directory (bz#1553431).
- Drop uotdated patch0 perdition-1.19-rc5-const-version.patch.
- Add libnsl2-devel BR.
- Fill bz#1625780 on dwz (<-https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/PB6DOPJSPJLDNUVZ3HUSNLMBORWLCXJK/#KORDFM5B3FXCY3RJ2MYBH5MT4YBNPJBJ)
o And unset _find_debuginfo_dwz_opts for s390x and armv7hl architectures.
diff --git a/.gitignore b/.gitignore
index 5073356..bf6ec12 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,2 +1,3 @@
/perdition-2.0.tar.bz2
/perdition-2.1.tar.bz2
+/perdition-2.2.tar.xz
diff --git a/perdition.spec b/perdition.spec
index 44550a3..d15a532 100644
--- a/perdition.spec
+++ b/perdition.spec
@@ -1,26 +1,30 @@
Summary: Mail Retrieval Proxy
Name: perdition
-Version: 2.1
-Release: 12%{?dist}
+Version: 2.2
+Release: 1%{?dist}
License: GPLv2+
Group: Applications/Internet
URL: http://horms.net/projects/perdition/
-Source: http://horms.net/projects/%{name}/download/%{version}/perdition-%{version}.…
+Source: http://horms.net/projects/%{name}/download/%{version}/%{name}-%{version}.ta…
# My systemd service template
Source1: %{name}-template.service
BuildRequires: automake autoconf libtool vanessa_logger-devel >= 0.0.10
BuildRequires: vanessa_adt-devel >= 0.0.6 vanessa_socket-devel >= 0.0.7
-BuildRequires: gdbm-devel mysql-devel postgresql-devel openldap-devel
-BuildRequires: popt-devel pam-devel zlib-devel openssl-devel ghostscript
-BuildRequires: tetex-latex tetex-dvips libtiff unixODBC-devel tinycdb-devel
-BuildRequires: libidn-devel libdb-devel
+BuildRequires: mariadb-connector-c-devel postgresql-devel openldap-devel
+BuildRequires: gdbm-devel popt-devel pam-devel zlib-devel openssl-devel
+BuildRequires: ghostscript tetex-latex libtiff unixODBC-devel
+BuildRequires: libidn2-devel libdb-devel tinycdb-devel libnsl2-devel
Requires: vanessa_logger >= 0.0.10 vanessa_adt >= 0.0.6 vanessa_socket >= 0.0.7
-#???
-Patch0: perdition-1.19-rc5-const-version.patch
Requires(post): systemd
Requires(preun): systemd
Requires(postun): systemd
+# By suggestion https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org…
+# See bz#https://bugzilla.redhat.com/show_bug.cgi?id=1625780
+%ifarch s390x armv7hl
+%undefine _find_debuginfo_dwz_opts
+%endif
+
%description
Perdition is a fully featured POP3 and IMAP4 proxy server. It is able to
handle both SSL and non-SSL connections and redirect users to a
@@ -126,8 +130,6 @@ which do not support the systemd unit file format.
%prep
%setup -q -n %{name}-%{version}
-%patch0 -p1 -b .const
-
%build
CFLAGS="${CFLAGS:-%optflags} -I/usr/kerberos/include" ; export CFLAGS
if [ -f configure.in ]; then
@@ -165,7 +167,8 @@ for service in pop3 pop3s imap4 imap4s managesieve; do
install -Dm 644 %{name}-$service.service %{buildroot}/%{_unitdir}/%{name}-$service.service
done
-rm -f %{buildroot}%{_libdir}/*.{a,la,so}
+# debug.#dwz# files strange and lead to faile build on arm and s390 platform. See https://koji.fedoraproject.org/koji/taskinfo?taskID=27400316
+rm -f %{buildroot}%{_libdir}/*.{a,la,so} %{buildroot}%{_libdir}/debug/usr/sbin/%{name}*.debug.#dwz#.*
%clean
@@ -291,7 +294,7 @@ fi
%{_libdir}/lib%{name}db_ldap.so.0.0.0
%{_sbindir}/%{name}db_ldap_makedb
%{_mandir}/man8/%{name}db_ldap_makedb.*
-%config(noreplace) %{_sysconfdir}/openldap/schema/%{name}.schema
+%config(noreplace) %{_sysconfdir}/openldap/schema
%files mysql
%{_libdir}/lib%{name}db_mysql.so.0
@@ -312,6 +315,18 @@ fi
%{_mandir}/man8/%{name}db_odbc_makedb.*
%changelog
+* Wed Sep 05 2018 Pavel Alexeev <Pahan(a)Hubbitus.info> - 2.2-1
+- Update to version 2.2 (request also for Epel: bz#1585846).
+- Remov tetex-dvips requirementw to fix FBFS as it is not present anymore in rawhide (bz#1556104).
+- Change distribuition tarball to xz archive.
+- Switch libidn to libidn2 (bz#1449141).
+- Change mysql-devel to mariadb-connector-c-devel (bz#1494071).
+- Own /etc/openldap/schema directory (bz#1553431).
+- Drop uotdated patch0 perdition-1.19-rc5-const-version.patch.
+- Add libnsl2-devel BR.
+- Fill https://bugzilla.redhat.com/show_bug.cgi?id=1625780 (<-https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/PB6DOPJSPJLDNUVZ3HUSNLMBORWLCXJK/#KORDFM5B3FXCY3RJ2MYBH5MT4YBNPJBJ)
+ o And unset _find_debuginfo_dwz_opts for s390x and armv7hl architectures.
+
* Fri Jul 13 2018 Fedora Release Engineering <releng(a)fedoraproject.org> - 2.1-12
- Rebuilt for https://fedoraproject.org/wiki/Fedora_29_Mass_Rebuild
diff --git a/sources b/sources
index 9a13003..0df1182 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-1e2c5ecfdcca0c02dde831186a077084 perdition-2.1.tar.bz2
+SHA512 (perdition-2.2.tar.xz) = 1ee64505ffe5ad11893fe61a37343fd004d5c15ea88b3dd18e790482248d4b68adc57f2b9c4a1fc12b06087fa7fa01f913db6d1e44bd7e82fd549fd85937a866
1
0
05 Sep '18
The package rpms/racket.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/racket.git/commit/?id=f241dafce9084…
https://src.fedoraproject.org/cgit/rpms/racket.git/commit/?id=b8569a44d31ac….
Change:
+ExcludeArch: s390x armv7hl
+%ifarch %{ix86}
Thanks.
Full change:
============
commit b8569a44d31acfcec836302dff9ae178d3957c0b
Author: dbenoit <dbenoit(a)redhat.com>
Date: Wed Sep 5 15:37:42 2018 -0400
fix i686 and remove ppc64
diff --git a/racket.spec b/racket.spec
index 5fb06c9..bb31acd 100644
--- a/racket.spec
+++ b/racket.spec
@@ -1,6 +1,6 @@
Name: racket
Version: 7.0
-Release: 1%{?dist}
+Release: 2%{?dist}
Summary: General purpose programming language
License: GPLv3 and LGPLv3 and MIT
@@ -21,7 +21,8 @@ Patch0: racket-7.0-remove-nonfree.patch
Patch1: racket-7.0-fix-doc-open-url.patch
# Issue Building for s390x and armv7hl in koji
-ExcludeArch: s390x armv7hl
+#ExcludeArch: s390x armv7hl
+ExcludeArch: s390x armv7hl ppc64
# To compile the program
BuildRequires: gcc
@@ -70,11 +71,20 @@ rm -r src/foreign/libffi
%build
cd src
+
+# Disable SSE on i686 until fixed upstream
+# https://github.com/racket/racket/issues/2245
+%ifarch %{ix86}
+ %set_build_flags
+ export CFLAGS=$(echo $CFLAGS | sed -e "s/-mfpmath=sse *//")
+%endif
+
%configure \
--enable-pthread \
--enable-shared \
--enable-libffi \
--disable-strip
+
%make_build
%install
@@ -231,7 +241,13 @@ A local installation of the Racket documentation system.
%{_datadir}/doc/racket
%changelog
-* Fri Aug 3 2018 David Benoit <dbenoit(a)redhat.com> 7.0-1
+* Wed Sep 5 2018 David Benoit <dbenoit(a)redhat.com> 7.0-2
+- Disable SSE math on i686 until issue is fixed upstream
+- Exclude ppc due to issue building Racket v7.0 and
+ arch being deprecated in next release
+
+* Fri Aug 17 2018 David Benoit <dbenoit(a)redhat.com> 7.0-1
+- Update sources to Racket v7.0
- Remove 6.12 patches and add update remove nonfree
srfi patch to 7.0
commit ee22cb0e92daba659fd9d9fd4855d952e21cc53f
Author: dbenoit <dbenoit(a)redhat.com>
Date: Fri Aug 31 15:39:51 2018 -0400
update package to v7.0
diff --git a/.gitignore b/.gitignore
index d013960..4fd1ad5 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1 +1 @@
-/racket-6.12-src.tgz
+/racket-7.0-src.tgz
diff --git a/racket-6.12-fix-doc-open-url.patch b/racket-6.12-fix-doc-open-url.patch
deleted file mode 100644
index 7c497f5..0000000
--- a/racket-6.12-fix-doc-open-url.patch
+++ /dev/null
@@ -1,69 +0,0 @@
-diff -urN racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt racket-6.12/share/pkgs/scribble-lib/help/search.rkt
---- racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:14:56.017273013 -0400
-+++ racket-6.12/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:15:37.839113929 -0400
-@@ -17,33 +17,38 @@
- #:notify [notify void])
- (define open-url (get-doc-open-url))
- (cond
-- [open-url
-- (define dest-url (let ([u (string->url open-url)])
-- (combine-url/relative
-- u
-- (string-join
-- (for/list ([s (explode-path sub)])
-- (if (path? s)
-- (path-element->string s)
-- (format "~a" s)))
-- "/"))))
-- (notify (url->string dest-url))
-- (send-url (url->string
-- (struct-copy url dest-url
-- [fragment (or fragment
-- (url-fragment dest-url))]
-- [query (append
-- (url-query dest-url)
-- (if query
-- (url-query
-- (string->url
-- (format "q?~a" query)))
-- null))])))]
-- [else
-- (let* ([path (build-path (find-user-doc-dir) sub)]
-- [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
-- (notify path)
-- (send-url/file path #:fragment fragment #:query query))]))
-+ [open-url
-+ (define dest-url (let ([u (string->url open-url)])
-+ (combine-url/relative
-+ u
-+ (string-join
-+ (for/list ([s (explode-path sub)])
-+ (if (path? s)
-+ (path-element->string s)
-+ (format "~a" s)))
-+ "/"))))
-+ (notify (url->string dest-url))
-+ (send-url (url->string
-+ (struct-copy url dest-url
-+ [fragment (or fragment
-+ (url-fragment dest-url))]
-+ [query (append
-+ (url-query dest-url)
-+ (if query
-+ (url-query
-+ (string->url
-+ (format "q?~a" query)))
-+ null))])))]
-+ [else
-+ (let* ([path (build-path (find-user-doc-dir) sub)]
-+ [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
-+ (notify path)
-+ (if (file-exists? path)
-+ (send-url/file path #:fragment fragment #:query query)
-+ (let ([part (lambda (pfx x) (if x (string-append pfx x) ""))])
-+ (send-url (string-append
-+ "https://docs.racket-lang.org/"
-+ sub (part "#" fragment) (part "?" query))))))]))
-
- ;; This is an example of changing this code to use the online manuals.
- ;; Normally, it's better to set `doc-open-url` in "etc/config.rktd",
diff --git a/racket-6.12-fix-rpaths.patch b/racket-6.12-fix-rpaths.patch
deleted file mode 100644
index f76730e..0000000
--- a/racket-6.12-fix-rpaths.patch
+++ /dev/null
@@ -1,13 +0,0 @@
-diff -urN racket-6.12-f/collects/racket/runtime-path.rkt racket-6.12/collects/racket/runtime-path.rkt
---- racket-6.12-f/collects/racket/runtime-path.rkt 2018-02-06 16:04:29.969878308 -0500
-+++ racket-6.12/collects/racket/runtime-path.rkt 2018-02-06 16:07:10.549176282 -0500
-@@ -186,7 +186,8 @@
- (path-of
- #,(datum->syntax
- #'orig-stx
-- `(,#'this-expression-source-file))))
-+ `(,#'this-expression-source-file)
-+ #'orig-stx)))
- #'void)])
- (apply to-values (resolve-paths (#%variable-reference)
- get-dir
diff --git a/racket-6.12-fix-webserver-rpaths.patch b/racket-6.12-fix-webserver-rpaths.patch
deleted file mode 100644
index 78614cf..0000000
--- a/racket-6.12-fix-webserver-rpaths.patch
+++ /dev/null
@@ -1,35 +0,0 @@
-diff -urN racket-6.12-rp/share/pkgs/web-server-lib/web-server/formlets/lib.rkt racket-6.12-wbsvr/share/pkgs/web-server-lib/web-server/formlets/lib.rkt
---- racket-6.12-rp/share/pkgs/web-server-lib/web-server/formlets/lib.rkt 2018-02-22 18:42:06.131455756 -0500
-+++ racket-6.12-wbsvr/share/pkgs/web-server-lib/web-server/formlets/lib.rkt 2018-03-20 15:07:09.666222028 -0400
-@@ -7,6 +7,7 @@
- racket/function
- racket/serialize
- syntax/location
-+ setup/collects
- (for-syntax racket/base
- syntax/parse))
-
-@@ -70,20 +71,20 @@
- #:range-contracts (map (curry coerce-contract 'formlet/c)
- contracts)))]))
- (define quote-this-module-path
-- (quote-module-path))
-+ (path->collects-relative (quote-module-path)))
- (define-syntax formlet/c
- (syntax-parser
- [(_ range ...)
- #:declare range (expr/c #'contract?
- #:name "range contract argument")
- #'(formlet/c** (-> listof-binding
-- (values (coerce-contract 'formlet/c range.c) ...)))]
-+ (values (coerce-contract 'formlet/c range #;.c) ...)))]
- [name:id
- #`(contract
- (-> contract? (... ...) contract?)
- dynamic-formlet/c
- quote-this-module-path
-- (quote-module-path)
-+ (path->collects-relative (quote-module-path))
- "formlet/c"
- #'name)]))
-
diff --git a/racket-6.12-remove-nonfree.patch b/racket-6.12-remove-nonfree.patch
deleted file mode 100644
index 4e72a3e..0000000
--- a/racket-6.12-remove-nonfree.patch
+++ /dev/null
@@ -1,1124 +0,0 @@
-diff -urN racket-6.12-nf/share/links.rktd racket-6.12/share/links.rktd
---- racket-6.12-nf/share/links.rktd 2018-01-31 13:04:06.757393819 -0500
-+++ racket-6.12/share/links.rktd 2018-01-31 13:04:25.419309681 -0500
-@@ -133,7 +133,6 @@
- (root "pkgs/r5rs-doc")
- (root "pkgs/srfi-lite-lib")
- (root "pkgs/srfi-lib")
-- (root "pkgs/srfi-lib-nonfree")
- (root "pkgs/html-doc")
- (root "pkgs/images-gui-lib")
- (root "pkgs/images-doc")
-@@ -177,7 +176,6 @@
- (root "pkgs/slideshow-exe")
- (root "pkgs/slideshow-plugin")
- (root "pkgs/srfi-doc")
-- (root "pkgs/srfi-doc-nonfree")
- (root "pkgs/syntax-color-doc")
- (root "pkgs/web-server-lib")
- (root "pkgs/unix-socket-lib")
-diff -urN racket-6.12-nf/share/pkgs/srfi/info.rkt racket-6.12/share/pkgs/srfi/info.rkt
---- racket-6.12-nf/share/pkgs/srfi/info.rkt 2018-01-31 13:04:07.216391749 -0500
-+++ racket-6.12/share/pkgs/srfi/info.rkt 2018-01-31 13:05:32.601025227 -0500
-@@ -3,13 +3,9 @@
- (define collection 'multi)
-
- (define deps '("srfi-lib"
-- "srfi-lib-nonfree"
-- "srfi-doc"
-- "srfi-doc-nonfree"))
-+ "srfi-doc"))
- (define implies '("srfi-lib"
-- "srfi-lib-nonfree"
-- "srfi-doc"
-- "srfi-foc-nonfree"))
-+ "srfi-doc"))
-
- (define pkg-desc "Legacy SRFI (Scheme) libraries")
-
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/info.rkt 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,19 +0,0 @@
--#lang info
--
--(define collection 'multi)
--
--(define build-deps '("mzscheme-doc"
-- "scheme-lib"
-- "base"
-- "scribble-lib"
-- "srfi-doc"
-- "srfi-lib-nonfree"
-- "racket-doc"
-- "r5rs-doc"
-- "r6rs-doc"
-- "compatibility-lib"))
--(define update-implies '("srfi-lib-nonfree"))
--
--(define pkg-desc "documentation part of \"srfi nonfree\"")
--
--(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,3 +0,0 @@
--#lang info
--
--(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
-@@ -1,66 +0,0 @@
--#lang scribble/doc
--@(require srfi/scribblings/util
-- scribble/manual
-- scribble/eval
-- scriblib/render-cond
-- scribble/core
-- scribble/html-properties
-- (for-syntax scheme/base)
-- (for-label scheme/base
-- racket/stream))
--
--@; ----------------------------------------------------------------------
--
--@title{SRFI Nonfree Libraries and Documentation}
--
--The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
--Implementation} (a.k.a. @deftech{SRFI}) process allows individual
--members of the Scheme community to propose libraries and extensions to
--be supported by multiple Scheme implementations.
--
--Racket is distributed with implementations of many SRFIs, most of
--which can be implemented as libraries. To import the bindings of SRFI
--@math{n}, use
--
--@racketblock[
--(require @#,elem{@racketidfont{srfi/}@math{n}})
--]
--
--This document lists the SRFIs that are supported by Racket and
--provides a link to the original SRFI specification (which is also
--distributed as part of Racket's documentation).
--
--The following SRFI specification documents are licensed restrictively.
--
--@table-of-contents[]
--
--
--@; ----------------------------------------
--
--@srfi[5]{A compatible let form with signatures and rest arguments}
--
--@redirect[5 '(
-- (let #t "unnamed")
--)]
--
--Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
--
--@; ----------------------------------------
--
--@srfi[29]{Localization}
--
--@redirect[29 '(
-- (current-language #f "current-language")
-- (current-country #f "current-country")
-- (current-locale-details #f "current-locale-details")
-- (declare-bundle! #f "declare-bundle!")
-- (store-bundle #f "store-bundle")
-- (load-bundle! #f "load-bundle!")
-- (localized-template #f "localized-template")
--)]
--
--Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
--
--@; ----------------------------------------
--
--@index-section[]
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
-@@ -1,507 +0,0 @@
--<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
--<html>
-- <head>
-- <meta name="generator" content="HTML Tidy, see www.w3.org">
-- <title>SRFI 29: Localization</title>
-- <meta name="author" content="Scott G. Miller">
-- <meta name="description" content="Localization">
-- </head>
-- <body>
-- <H1>Title</H1>
--
-- SRFI 29: Localization
--
-- <H1>Author</H1>
--
-- Scott G. Miller
--
-- <H1>Abstract</H1>
--
-- This document specifies an interface to retrieving and
-- displaying locale sensitive messages. A Scheme program can
-- register one or more translations of templated messages, and
-- then write Scheme code that can transparently retrieve the
-- appropriate message for the locale under which the Scheme
-- system is running. <br>
--
--
-- <H1>Rationale</H1>
--
-- <p>As any programmer that has ever had to deal with making his
-- or her code readable in more than one locale, the process of
-- sufficiently abstracting program messages from their
-- presentation to the user is non-trivial without help from the
-- programming language. Most modern programming language
-- libraries do provide some mechanism for performing this
-- separation.</p>
--
-- <p>A portable API that allows a piece of code to run without
-- modification in different countries and under different
-- languages is a must for any non-trivial software project.
-- The interface should separate the logic of a program from
-- the myriad of translations that may be necessary.</p>
--
-- <p>The interface described in this document provides such
-- functionality. The underlying implementation is also allowed to
-- use whatever datastructures it likes to provide access to the
-- translations in the most efficient manner possible. In
-- addition, the implementation is provided with standardized
-- functions that programs will use for accessing an external,
-- unspecified repository of translations.</p>
--
-- <p>This interface <i>does not</i> cover all aspects of
-- localization, including support for non-latin characters,
-- number and date formatting, etc. Such functionality is the
-- scope of a future SRFI that may extend this one.</p>
--
-- <H1>Dependencies</H1>
--
-- An SRFI-29 conformant implementation must also implement
-- SRFI-28, Basic Format Strings. Message templates are strings
-- that must be processed by the <tt>format</tt> function
-- specified in that SRFI.
--
-- <H1>Specification</H1>
--
-- <h3>Message Bundles</h3>
--
-- <p>A Message Bundle is a set of message templates and their
-- identifying keys. Each bundle contains one or more such
-- key/value pairs. The bundle itself is associated with a
-- <i>bundle specifier</i> which uniquely identifies the
-- bundle.</p>
--
-- <h3>Bundle Specifiers</h3>
--
-- <p>A Bundle Specifier is a Scheme list that describes, in order
-- of importance, the package and locale of a message bundle.
-- In most cases, a locale specifier will have between one
-- and three elements. The first element is a symbol denoting the
-- package for which this bundle applies. The second and third
-- elements denote a <i>locale</i>. The second element (first
-- element of the locale) if present, is the two letter, ISO 639-1
-- language code for the bundle. The third element, if present, is
-- a two letter ISO 3166-1 country code. In some cases, a
-- fourth element may be present, specifying the encoding used for
-- the bundle. All bundle specifier elements are Scheme
-- symbols.</p>
--
-- <p>If only one translation is provided, it should be designated
-- only by a package name, for example <tt>(mathlib)</tt>. This
-- translation is called the <i>default</i> translation.</p>
--
-- <h3>Bundle Searching</h3>
--
-- <p>When a message template is retrieved from a bundle, the
-- Scheme implementation will provide the locale under which the
-- system is currently running. When the template is retrieved,
-- the package name will be specified. The Scheme system should
-- construct a Bundle Specifier from the provided package name and
-- the active locale. For example, when retrieving a message
-- template for French Canadian, in the <tt>mathlib</tt> package,
-- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
-- program may also retrieve the elements of the current locale
-- using the no-argument procedures:</p>
--
-- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
-- <i>symbol</i></tt><br>
-- <tt><b>current-language</b> <i>symbol</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- When given no arguments, returns the current ISO 639-1
-- language code as a symbol. If provided with an
-- argument, the current language is set to that named by the
-- symbol for the currently executing Scheme thread (or for the
-- entire Scheme system if such a distinction is not possible).
--
-- </blockquote>
--
-- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
-- <i>symbol</i></tt><br>
-- <tt><b>current-country</b> <i>symbol</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- returns the current ISO 3166-1 country code as a symbol.
-- If provided with an argument, the current country is
-- set to that named by the symbol for the currently executing
-- Scheme thread (or for the entire Scheme system if such a
-- distinction is not possible).
-- </blockquote>
--
-- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
-- symbol</i></tt>s<br>
-- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- Returns a list of additional locale details as a list of
-- symbols. This list may contain information about
-- encodings or other more specific information. If
-- provided with an argument, the current locale details are set
-- to those given in the currently executing Scheme thread (or
-- for the entire Scheme system if such a distinction is not
-- possible).
-- </blockquote>
--
-- <p>The Scheme System should first check for a bundle with the
-- exact name provided. If no such bundle is found, the last
-- element from the list is removed and a search is tried for a
-- bundle with that name. If no bundle is then found, the list is
-- shortened by removing the last element again. If no message is
-- found and the bundle specifier is now the empty list, an error
-- should be raised.</p>
--
-- <p>The reason for this search order is to provide the most
-- locale sensitive template possible, but to fall back on more
-- general templates if a translation has not yet been provided
-- for the given locale.</p>
--
-- <h3>Message Templates</h3>
--
-- <p>A message template is a localized message that may or may
-- not contain one of a number of formatting codes. A message
-- template is a Scheme string. The string is of a form that can
-- be processed by the <tt>format</tt> procedure found in many
-- Scheme systems and formally specified in SRFI-28 (Basic Format
-- Strings).</p>
--
-- <p>This SRFI also extends SRFI-28 to provide an additional
-- <tt>format</tt> escape code:</p>
--
-- <blockquote>
-- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
-- follows this code immediately to reference the [N]'th
-- optional value absolutely, rather than the next unconsumed
-- value. The referenced value is <i>not</i> consumed.
-- </blockquote>
-- This extension allows optional values to be positionally
-- referenced, so that message templates can be constructed that
-- can produce the proper word ordering for a language.
--
-- <h3>Preparing Bundles</h3>
-- Before a bundle may be used by the Scheme system to retrieve
-- localized template messages, they must be made available to the
-- Scheme system. This SRFI specifies a way to portably
-- define the bundles, as well as store them in and retrieve them
-- from an unspecified system which may be provided by resources
-- outside the Scheme system.<br>
--
--
-- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
-- association-list</i> -> undefined<br>
-- </tt></p>
--
-- <blockquote>
-- Declares a new bundle named by the given bundle-specifier.
-- The contents of the bundle are defined by the provided
-- association list. The list contains associations
-- between Scheme symbols and the message templates (Scheme
-- strings) they name. If a bundle already exists with the
-- given name, it is overwritten with the newly declared
-- bundle.<br>
-- </blockquote>
-- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
-- boolean</tt><br>
--
--
-- <blockquote>
-- Attempts to store a bundle named by the given bundle
-- specifier, and previously made available using
-- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
-- unspecified mechanism that may be persistent across Scheme
-- system restarts. If successful, a non-false value is
-- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
-- </blockquote>
-- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
-- boolean</tt><br>
--
--
-- <blockquote>
-- Attempts to retrieve a bundle from an unspecified mechanism
-- which stores bundles outside the Scheme system. If the
-- bundle was retrieved successfully, the function returns a
-- non-false value, and the bundle is immediately available to
-- the Scheme system. If the bundle could not be found or loaded
-- successfully, the function returns <tt>#f</tt>, and the
-- Scheme system's bundle registry remains unaffected.<br>
-- </blockquote>
-- A compliant Scheme system may choose not to provide any
-- external mechanism to store localized bundles. If it does
-- not, it must still provide implementations for
-- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
-- case, both functions must return <tt>#f</tt> regardless of the
-- arguments given. Users of this SRFI should recognize that the
-- inability to load or store a localized bundle in an external
-- repository is <i>not</i> a fatal error.<br>
--
--
-- <h3>Retrieving Localized Message Templates</h3>
--
-- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
-- message-template-name</tt></i> <tt>-> <i>string or #f<br>
-- </i></tt></p>
--
-- <blockquote>
-- Retrieves a localized message template for the given package
-- name and the given message template name (both symbols).
-- If no such message could be found, false (#f) is
-- returned.<br>
-- <br>
-- </blockquote>
-- After retrieving a template, the calling program can use
-- <tt>format</tt> to produce a string that can be displayed to
-- the user.<br>
--
--
-- <h2>Examples</h2>
-- The below example makes use of SRFI-29 to display simple,
-- localized messages. It also defines its bundles in such a
-- way that the Scheme system may store and retrieve the bundles
-- from a more efficient system catalog, if available.<br>
--
--<pre>
--(let ((translations
-- '(((en) . ((time . "Its ~a, ~a.")
-- (goodbye . "Goodbye, ~a.")))
-- ((fr) . ((time . "~1@*~a, c'est ~a.")
-- (goodbye . "Au revoir, ~a."))))))
-- (for-each (lambda (translation)
-- (let ((bundle-name (cons 'hello-program (car translation))))
-- (if (not (load-bundle! bundle-name))
-- (begin
-- (declare-bundle! bundle-name (cdr translation))
-- (store-bundle! bundle-name)))))
-- translations))
--
--(define localized-message
-- (lambda (message-name . args)
-- (apply format (cons (localized-template 'hello-program
-- message-name)
-- args))))
--
--(let ((myname "Fred"))
-- (display (localized-message 'time "12:00" myname))
-- (display #\newline)
--
-- (display (localized-message 'goodbye myname))
-- (display #\newline))
--
--;; Displays (English):
--;; Its 12:00, Fred.
--;; Goodbye, Fred.
--;;
--;; French:
--;; Fred, c'est 12:00.
--;; Au revoir, Fred.
--</pre>
--
-- <H1>Implementation</H1>
--
-- <p>The implementation requires that the Scheme system provide a
-- definition for <tt>current-language</tt> and
-- <tt>current-country</tt> capable of distinguishing the correct
-- locale present during a Scheme session. The definitions of
-- those functions in the reference implementation are not capable
-- of that distinction. Their implementation is provided only so
-- that the following code can run in any R4RS scheme system.
-- <br>
-- </p>
--
-- <p>In addition, the below implementation of a compliant
-- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
-- SRFI-23 (Error reporting)</p>
--<pre>
--;; The association list in which bundles will be stored
--(define *localization-bundles* '())
--
--;; The current-language and current-country functions provided
--;; here must be rewritten for each Scheme system to default to the
--;; actual locale of the session
--(define current-language
-- (let ((current-language-value 'en))
-- (lambda args
-- (if (null? args)
-- current-language-value
-- (set! current-language-value (car args))))))
--
--(define current-country
-- (let ((current-country-value 'us))
-- (lambda args
-- (if (null? args)
-- current-country-value
-- (set! current-country-value (car args))))))
--
--;; The load-bundle! and store-bundle! both return #f in this
--;; reference implementation. A compliant implementation need
--;; not rewrite these procedures.
--(define load-bundle!
-- (lambda (bundle-specifier)
-- #f))
--
--(define store-bundle!
-- (lambda (bundle-specifier)
-- #f))
--
--;; Declare a bundle of templates with a given bundle specifier
--(define declare-bundle!
-- (letrec ((remove-old-bundle
-- (lambda (specifier bundle)
-- (cond ((null? bundle) '())
-- ((equal? (caar bundle) specifier)
-- (cdr bundle))
-- (else (cons (car bundle)
-- (remove-old-bundle specifier
-- (cdr bundle))))))))
-- (lambda (bundle-specifier bundle-assoc-list)
-- (set! *localization-bundles*
-- (cons (cons bundle-specifier bundle-assoc-list)
-- (remove-old-bundle bundle-specifier
-- *localization-bundles*))))))
--
--;;Retrieve a localized template given its package name and a template name
--(define localized-template
-- (letrec ((rdc
-- (lambda (ls)
-- (if (null? (cdr ls))
-- '()
-- (cons (car ls) (rdc (cdr ls))))))
-- (find-bundle
-- (lambda (specifier template-name)
-- (cond ((assoc specifier *localization-bundles*) =>
-- (lambda (bundle) bundle))
-- ((null? specifier) #f)
-- (else (find-bundle (rdc specifier)
-- template-name))))))
-- (lambda (package-name template-name)
-- (let loop ((specifier (cons package-name
-- (list (current-language)
-- (current-country)))))
-- (and (not (null? specifier))
-- (let ((bundle (find-bundle specifier template-name)))
-- (and bundle
-- (cond ((assq template-name bundle) => cdr)
-- ((null? (cdr specifier)) #f)
-- (else (loop (rdc specifier)))))))))))
--
--;;An SRFI-28 and SRFI-29 compliant version of format. It requires
--;;SRFI-23 for error reporting.
--(define format
-- (lambda (format-string . objects)
-- (let ((buffer (open-output-string)))
-- (let loop ((format-list (string->list format-string))
-- (objects objects)
-- (object-override #f))
-- (cond ((null? format-list) (get-output-string buffer))
-- ((char=? (car format-list) #\~)
-- (cond ((null? (cdr format-list))
-- (error 'format "Incomplete escape sequence"))
-- ((char-numeric? (cadr format-list))
-- (let posloop ((fl (cddr format-list))
-- (pos (string->number
-- (string (cadr format-list)))))
-- (cond ((null? fl)
-- (error 'format "Incomplete escape sequence"))
-- ((and (eq? (car fl) '#\@)
-- (null? (cdr fl)))
-- (error 'format "Incomplete escape sequence"))
-- ((and (eq? (car fl) '#\@)
-- (eq? (cadr fl) '#\*))
-- (loop (cddr fl) objects (list-ref objects pos)))
-- (else
-- (posloop (cdr fl)
-- (+ (* 10 pos)
-- (string->number
-- (string (car fl)))))))))
-- (else
-- (case (cadr format-list)
-- ((#\a)
-- (cond (object-override
-- (begin
-- (display object-override buffer)
-- (loop (cddr format-list) objects #f)))
-- ((null? objects)
-- (error 'format "No value for escape sequence"))
-- (else
-- (begin
-- (display (car objects) buffer)
-- (loop (cddr format-list)
-- (cdr objects) #f)))))
-- ((#\s)
-- (cond (object-override
-- (begin
-- (display object-override buffer)
-- (loop (cddr format-list) objects #f)))
-- ((null? objects)
-- (error 'format "No value for escape sequence"))
-- (else
-- (begin
-- (write (car objects) buffer)
-- (loop (cddr format-list)
-- (cdr objects) #f)))))
-- ((#\%)
-- (if object-override
-- (error 'format "Escape sequence following positional override does not require a value"))
-- (display #\newline buffer)
-- (loop (cddr format-list) objects #f))
-- ((#\~)
-- (if object-override
-- (error 'format "Escape sequence following positional override does not require a value"))
-- (display #\~ buffer)
-- (loop (cddr format-list) objects #f))
-- (else
-- (error 'format "Unrecognized escape sequence"))))))
-- (else (display (car format-list) buffer)
-- (loop (cdr format-list) objects #f)))))))
--
--</pre>
--
-- <H1>Copyright</H1>
--
-- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
--
-- <p>This document and translations of it may be copied and
-- furnished to others, and derivative works that comment on or
-- otherwise explain it or assist in its implementation may be
-- prepared, copied, published and distributed, in whole or in
-- part, without restriction of any kind, provided that the above
-- copyright notice and this paragraph are included on all such
-- copies and derivative works. However, this document itself may
-- not be modified in any way, such as by removing the copyright
-- notice or references to the Scheme Request For Implementation
-- process or editors, except as needed for the purpose of
-- developing SRFIs in which case the procedures for copyrights
-- defined in the SRFI process must be followed, or as required to
-- translate it into languages other than English.</p>
--
-- <p>The limited permissions granted above are perpetual and will
-- not be revoked by the authors or their successors or
-- assigns.</p>
--
-- <p>This document and the information contained herein is
-- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
-- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
-- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
-- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
-- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
-- <hr>
--
-- <address>
-- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
-- Rush</a>
-- </address>
--
-- <address>
-- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
-- Miller</a>
-- </address>
-- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
-- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
-- Daylight Time 2002 <!-- hhmts end --> <br>
-- </body>
--</html>
--
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
-@@ -1,345 +0,0 @@
--<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
--<html>
--<head>
-- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
--</head>
--<body>
--
--<H1>Title</H1>
--
--SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
--
--<H1>Author</H1>
--
--Andy Gaynor
--
--<H1>Status</H1>
--
--This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
--You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
--<P><UL>
--<LI>Received: 1999/2/2
--<LI>Draft: 1999/2/10-1999/04/12
--<LI>Final: 1999/4/26
--<LI>Revised reference implementation: 2003/01/27
--</UL>
--
--<H1>Abstract</H1>
--
--The <i>named-let</i> incarnation of the <code>let</code> form has two slight
--inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
--form makes no accommodation for rest arguments, an issue of functionality
--and consistency. As defined, the <code>let</code> form does not accommodate
--signature-style syntax, an issue of aesthetics and consistency. Both
--issues are addressed here in a manner which is compatible with the traditional
--<code>let</code> form but for minor extensions.
--
--<H1>Rationale</H1>
--
--<H2>Signature-style Syntax</H2>
--
--Consider the following two equivalent definitions:
--
--<p><pre>
--(define fibonacci
-- (lambda (n i f0 f1)
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
--
--(define (fibonacci n i f0 f1)
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--</pre>
--
--Although there is a named-let analog for the former form, there is none
--for the latter. To wit, suppose one wished to compute the 10th element
--of the Fibonacci sequence using a named let:
--
--<p>
--<pre>
--(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--Values: 55
--</pre>
--
--As it stands, one cannot equivalently write
--
--<p>
--<pre>
--(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--</pre>
--
--which is consistent with <code>define</code>'s signature-style form.
--<p>Those that favor the signature style may prefer this extension.
--In any case, it may be more appropriate to include all bound names within
--the binding section. As presented, this straightforward extension
--introduces no ambiguity or incompatibility with the existing definition
--of let.
--
--<H2>Rest Arguments</H2>
--
--As it stands, one cannot write a named let with rest arguments, as in
--
--<p>
--<pre>
--(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
-- (if (null? x)
-- 'just-a-silly-contrived-example
-- (begin
-- (write (car x) port)
-- (apply blast port (cdr x)))))
--</pre>
--
--otherwise equivalent to
--
--<p>
--<pre>
--(letrec ((blast (lambda (port . x)
-- (if (null? x)
-- 'just-a-silly-contrived-example
-- (begin
-- (write (car x) port)
-- (apply blast port (cdr x)))))))
-- (blast (current-output-port) (+ 1 2) 4 5))
--</pre>
--
--While this example is rather contrived, the functionality is not.
--There are several times when the author has used this construct in practice.
--Regardless, there is little reason to deny the <code>let</code> form access to
--all the features of lambda functionality.
--
--<H2>Symbols in Binding Sections</H2>
--
--Both the features above rely upon the placement of symbols in <code>let</code>
--binding lists (this statement is intentially simplistic). The only
--other apparent use of such symbol placement is to tersely bind variables
--to unspecified values. For example, one might desire to use
--<code>(let (foo bar baz) ...)</code>
--to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
--unspecified values.
--
--<p>This usage is considered less important in light of the rationales
--presented above, and an alternate syntax is immediately apparent, as
--in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
--be preferable, consistently parenthesizing normal binding clauses.
--
--<H1>Specification</H1>
--
--<H2>Syntax</H2>
--
--<p>
--A formal specification of the syntax follows. Below, body, expression,
--and identifier are free. Each instantiation of binding-name must be
--unique.
--</p>
--
--<p>
--<pre>
-- let = "(" "let" let-bindings body ")"
-- expressions = nothing | expression expressions
-- let-bindings = let-name bindings
-- |"(" let-name "." bindings ")"
-- let-name = identifier
-- bindings = "(" ")"
-- | rest-binding
-- | "(" normal-bindings ["." rest-binding] ")"
--normal-bindings = nothing
-- | normal-binding normal-bindings
--normal-binding = "(" binding-name expression ")"
-- binding-name = identifier
-- rest-binding = "(" binding-name expressions ")"
--</pre>
--
--<p>
--For clarity and convenience, an informal specification follows.
--</p>
--
--<ol>
--<li><a name="unnamed">Unnamed</a>
--
--<p><pre>
--(let ((<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-non-sig">
--Named, non-signature-style, no rest argument</a>
--
--<p><pre>
--(let <name> ((<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-sig">Named, signature-style, no rest argument</a>
--
--<p><pre>
--(let (<name> (<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
--
--<p><pre>
--(let <name> ((<parameter> <argument>)...
--
--. (<rest-parameter> <rest-argument>...))
-- <body>...)
--</pre>
--
--<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
--
--<p><pre>
--(let (<name> (<parameter> <argument>)...
--
--. (<rest-parameter> <rest-argument>...))
-- <body>...)
--</pre>
--</li>
--</ol>
--
--<H2>Semantics</H2>
--
--Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
--and <code>letrec</code> forms, respectively.
--
--<ul>
--<li>For informal syntax 1:
--
--<p><pre>
--(($lambda (<parameter>...) <body>...) <argument>...)
--</pre>
--</li>
--
--<li>For informal syntaxes 2 and 3:
--
--<p>
--<pre>
--($letrec ((<name> ($lambda (<parameter>...) <body>...)))
-- (<name> <argument>...))
--</pre>
--</li>
--
--<li>For informal syntaxes 4 and 5:
--
--<p>
--<pre>
--($letrec ((<name> ($lambda (<parameter>...
--
--. <rest-parameter>) <body>...)))
-- (<name> <argument>... <rest-argument>...))
--</pre>
--</li>
--</ul>
--
--<H1>Implementation</H1>
--
--Here is an implementation using <code>SYNTAX-RULES</code>.
--
--<p>
--<pre>
--;; Use your own standard let.
--;; Or call a lambda.
--;; (define-syntax standard-let
--;;
--;; (syntax-rules ()
--;;
--;; ((let ((var val) ...) body ...)
--;; ((lambda (var ...) body ...) val ...))))
--
--(define-syntax let
--
-- (syntax-rules ()
--
-- ;; No bindings: use standard-let.
-- ((let () body ...)
-- (standard-let () body ...))
-- ;; Or call a lambda.
-- ;; ((lambda () body ...))
--
-- ;; All standard bindings: use standard-let.
-- ((let ((var val) ...) body ...)
-- (standard-let ((var val) ...) body ...))
-- ;; Or call a lambda.
-- ;; ((lambda (var ...) body ...) val ...)
--
-- ;; One standard binding: loop.
-- ;; The all-standard-bindings clause didn't match,
-- ;; so there must be a rest binding.
-- ((let ((var val) . bindings) body ...)
-- (let-loop #f bindings (var) (val) (body ...)))
--
-- ;; Signature-style name: loop.
-- ((let (name binding ...) body ...)
-- (let-loop name (binding ...) () () (body ...)))
--
-- ;; defun-style name: loop.
-- ((let name bindings body ...)
-- (let-loop name bindings () () (body ...)))))
--
--(define-syntax let-loop
--
-- (syntax-rules ()
--
-- ;; Standard binding: destructure and loop.
-- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
-- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
--
-- ;; Rest binding, no name: use standard-let, listing the rest values.
-- ;; Because of let's first clause, there is no "no bindings, no name" clause.
-- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
-- ;; Or call a lambda with a rest parameter on all values.
-- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
-- ;; Or use one of several other reasonable alternatives.
--
-- ;; No bindings, name: call a letrec'ed lambda.
-- ((let-loop name () (var ...) (val ...) body)
-- ((letrec ((name (lambda (var ...) . body)))
-- name)
-- val ...))
--
-- ;; Rest binding, name: call a letrec'ed lambda.
-- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
-- ((letrec ((name (lambda (var ... . rest-var) . body)))
-- name)
-- val ... rest-val ...))))
--</pre>
--
--
--<H1>Copyright</H1>
--
--Copyright (C) Andy Gaynor (1999). All Rights Reserved.
--<p>This document and translations of it may be copied and furnished to
--others, and derivative works that comment on or otherwise explain it or
--assist in its implementation may be prepared, copied, published and distributed,
--in whole or in part, without restriction of any kind, provided that the
--above copyright notice and this paragraph are included on all such copies
--and derivative works. However, this document itself may not be modified
--in any way, such as by removing the copyright notice or references to the
--Scheme Request For Implementation process or editors, except as needed
--for the purpose of developing SRFIs in which case the procedures for copyrights
--defined in the SRFI process must be followed, or as required to translate
--it into languages other than English.
--<p>The limited permissions granted above are perpetual and will not be
--revoked by the authors or their successors or assigns.
--<p>This document and the information contained herein is provided on an
--"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
--EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
--USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
--WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
--
-- <hr>
-- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
--
--</body>
--</html>
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/info.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/info.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,14 +0,0 @@
--#lang info
--
--(define collection 'multi)
--
--(define deps '("scheme-lib"
-- "base"
-- "r6rs-lib"
-- "srfi-lib"
-- "compatibility-lib"))
--
--
--(define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"")
--
--(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1 +0,0 @@
--#lang s-exp srfi/provider srfi/%3a5
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,3 +0,0 @@
--#lang s-exp srfi/provider srfi/5
--
--;; FIXME: "rest arguments" need to generate mutable lists
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,86 +0,0 @@
--;;;
--;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
--;;; Time-stamp: <03/04/08 09:56:06 solsona>
--;;;
--;;; Usually, I would add a copyright notice, and the announce that
--;;; this code is under the LGPL licence. Nevertheless, I only did the
--;;; port to PLT Scheme v200, and here is the copyright notice,
--;;; comments, and licence from the original source:
--;;;
--;;; Copyright (C) Andy Gaynor (1999-2003)
--;;;
--;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
--;;; Renamed to s:let by Eli Barzilay
--
--#lang scheme/base
--(provide s:let)
--
--(define-syntax s:let
-- (syntax-rules ()
-- ;; standard
-- ((s:let () body ...)
-- (let () body ...))
-- ((s:let ((var val) ...) body ...)
-- (let ((var val) ...) body ...))
--
-- ;; rest style
-- ((s:let ((var val) . bindings) body ...)
-- (let-loop #f bindings (var) (val) (body ...)))
--
-- ;; signature style
-- ((s:let (name bindings ...) body ...)
-- (let-loop name (bindings ...) () () (body ...)))
--
-- ;; standard named style
-- ((s:let name (bindings ...) body ...)
-- (let-loop name (bindings ...) () () (body ...)))
--
-- ))
--
--;; A loop to walk down the list of bindings.
--
--(define-syntax let-loop
-- (syntax-rules ()
--
-- ;; No more bindings - make a LETREC.
-- ((let-loop name () (vars ...) (vals ...) body)
-- ((letrec ((name (lambda (vars ...) . body)))
-- name)
-- vals ...))
--
-- ;; Rest binding, no name
-- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-- (let ((var val) ... (rest-var (list rest-val ...))) . body))
--
-- ;; Process a (var val) pair.
-- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
-- (let-loop name (more ...) (vars ... var) (vals ... val) body))
--
-- ;; End with a rest variable - make a LETREC.
-- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
-- ((letrec ((name (lambda (vars ... . rest-var) . body)))
-- name)
-- vals ... rest-vals ...))))
--
--;; Four loops - normal and `signature-style', each with and without a rest
--;; binding.
--;;
--;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
--;; (if (= i n)
--;; f0
--;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
--;;
--;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
--;; (if (= i n)
--;; f0
--;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
--;;
--;;(let fibonacci ((n 10) (i 0) . (f 0 1))
--;; (if (= i n)
--;; (car f)
--;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
--;;
--;;(let (fibonacci (n 10) (i 0) . (f 0 1))
--;; (if (= i n)
--;; (car f)
--;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,2 +0,0 @@
--;; module loader for SRFI-5
--#lang s-exp srfi/provider srfi/5/let #:unprefix s:
diff --git a/racket-6.12-update-srfi.patch b/racket-6.12-update-srfi.patch
deleted file mode 100644
index 1894e32..0000000
--- a/racket-6.12-update-srfi.patch
+++ /dev/null
@@ -1,4369 +0,0 @@
-diff -urN racket-6.12-orig/share/links.rktd racket-6.12/share/links.rktd
---- racket-6.12-orig/share/links.rktd 2018-01-26 16:12:13.000000000 -0500
-+++ racket-6.12/share/links.rktd 2018-01-31 12:22:57.600504036 -0500
-@@ -133,6 +133,7 @@
- (root "pkgs/r5rs-doc")
- (root "pkgs/srfi-lite-lib")
- (root "pkgs/srfi-lib")
-+ (root "pkgs/srfi-lib-nonfree")
- (root "pkgs/html-doc")
- (root "pkgs/images-gui-lib")
- (root "pkgs/images-doc")
-@@ -176,6 +177,7 @@
- (root "pkgs/slideshow-exe")
- (root "pkgs/slideshow-plugin")
- (root "pkgs/srfi-doc")
-+ (root "pkgs/srfi-doc-nonfree")
- (root "pkgs/syntax-color-doc")
- (root "pkgs/web-server-lib")
- (root "pkgs/unix-socket-lib")
-diff -urN racket-6.12-orig/share/pkgs/srfi/info.rkt racket-6.12/share/pkgs/srfi/info.rkt
---- racket-6.12-orig/share/pkgs/srfi/info.rkt 2018-01-26 16:10:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi/info.rkt 2018-01-31 12:22:58.139502019 -0500
-@@ -1 +1,16 @@
--(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("srfi-lib" "srfi-doc"))) (define implies (quote ("srfi-lib" "srfi-doc"))) (define pkg-desc "Legacy SRFI (Scheme) libraries") (define pkg-authors (quote (mflatt noel chongkai jay)))))
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define deps '("srfi-lib"
-+ "srfi-lib-nonfree"
-+ "srfi-doc"
-+ "srfi-doc-nonfree"))
-+(define implies '("srfi-lib"
-+ "srfi-lib-nonfree"
-+ "srfi-doc"
-+ "srfi-foc-nonfree"))
-+
-+(define pkg-desc "Legacy SRFI (Scheme) libraries")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/info.rkt racket-6.12/share/pkgs/srfi-doc/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-doc/info.rkt 2018-01-26 16:10:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/info.rkt 2018-01-31 12:22:58.139502019 -0500
-@@ -1 +1,24 @@
--(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define build-deps (quote ("mzscheme-doc" "scheme-lib" "base" "scribble-lib" "srfi-lib" "racket-doc" "r5rs-doc" "r6rs-doc" "compatibility-lib"))) (define update-implies (quote ("srfi-lib"))) (define pkg-desc "documentation part of \"srfi\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define build-deps '("mzscheme-doc"
-+ "scheme-lib"
-+ "base"
-+ "scribble-lib"
-+ "srfi-lib"
-+ "racket-doc"
-+ "r5rs-doc"
-+ "r6rs-doc"
-+ "compatibility-lib"))
-+
-+(define deps '("scheme-lib"
-+ "base"
-+ "scribble-lib"
-+ "compatibility-lib"))
-+
-+(define update-implies '("srfi-lib"))
-+
-+(define pkg-desc "documentation part of \"srfi\"")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl
---- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl 2018-01-31 12:22:58.144502000 -0500
-@@ -1,5 +1,6 @@
- #lang scribble/doc
--@(require scribble/manual
-+@(require srfi/scribblings/util
-+ scribble/manual
- scribble/eval
- scriblib/render-cond
- scribble/core
-@@ -8,54 +9,6 @@
- (for-label scheme/base
- racket/stream))
-
--@(define-syntax (srfi stx)
-- (syntax-case stx ()
-- [(_ num #:subdir subdir? . title)
-- (with-syntax ([srfi/n (string->symbol (format "srfi/~a" (syntax-e #'num)))])
-- #'(begin
-- (section #:tag (format "srfi-~a" num)
-- #:style 'unnumbered
-- (format "SRFI ~a: " num)
-- . title)
-- (defmodule srfi/n)
-- "Original specification: "
-- (let* ([label (format "SRFI ~a" num)]
-- [sub (if subdir? (format "srfi-~a/" num) "")]
-- [url ( (b) (format "~a/srfi-std/~asrfi-~a.html" b sub num))])
-- (cond-element
-- [(or latex text) @link[(url "http://docs.racket-lang.org") label]]
-- [else @link[(url ".") label]]))))]
-- [(_ num . title) #'(srfi num #:subdir #f . title)]))
--
--@;{ The `lst' argument is a list of
-- (list sym syntactic-form? html-anchor) }
--@(define (redirect n lst #:subdir [subdir? #f])
-- (let ([file (if subdir?
-- (format "srfi-~a/srfi-~a.html" n n)
-- (format "srfi-~a.html" n))]
-- [mod-path (string->symbol (format "srfi/~a" n))])
-- (make-binding-redirect-elements mod-path
-- (map (lambda (b)
-- (list (car b) (cadr b)
-- (build-path "srfi-std" file)
-- (caddr b)))
-- lst))))
--
--@(define in-core
-- (case-lambda
-- [() (in-core ".")]
-- [(k) @elem{This SRFI's bindings are also available in
-- @racketmodname[racket/base]@|k|}]))
--
--@(begin
-- (define-syntax-rule (def-mz mz-if)
-- (begin
-- (require (for-label mzscheme))
-- (define mz-if (racket if))))
-- (def-mz mz-if))
--
--@(define srfi-std (style #f (list (install-resource "srfi-std"))))
--
- @; ----------------------------------------------------------------------
-
- @title{SRFIs: Libraries}
-@@ -245,14 +198,6 @@
-
- @; ----------------------------------------
-
--@srfi[5]{A compatible let form with signatures and rest arguments}
--
--@redirect[5 '(
-- (let #t "unnamed")
--)]
--
--@; ----------------------------------------
--
- @srfi[6]{Basic String Ports}
-
- @redirect[6 '(
-@@ -671,20 +616,6 @@
-
- @; ----------------------------------------
-
--@srfi[29]{Localization}
--
--@redirect[29 '(
-- (current-language #f "current-language")
-- (current-country #f "current-country")
-- (current-locale-details #f "current-locale-details")
-- (declare-bundle! #f "declare-bundle!")
-- (store-bundle #f "store-bundle")
-- (load-bundle! #f "load-bundle!")
-- (localized-template #f "localized-template")
--)]
--
--@; ----------------------------------------
--
- @srfi[30]{Nested Multi-line Comments}
-
- This SRFI's syntax is part of Racket's default reader.
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html
---- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
-@@ -1,507 +0,0 @@
--<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
--<html>
-- <head>
-- <meta name="generator" content="HTML Tidy, see www.w3.org">
-- <title>SRFI 29: Localization</title>
-- <meta name="author" content="Scott G. Miller">
-- <meta name="description" content="Localization">
-- </head>
-- <body>
-- <H1>Title</H1>
--
-- SRFI 29: Localization
--
-- <H1>Author</H1>
--
-- Scott G. Miller
--
-- <H1>Abstract</H1>
--
-- This document specifies an interface to retrieving and
-- displaying locale sensitive messages. A Scheme program can
-- register one or more translations of templated messages, and
-- then write Scheme code that can transparently retrieve the
-- appropriate message for the locale under which the Scheme
-- system is running. <br>
--
--
-- <H1>Rationale</H1>
--
-- <p>As any programmer that has ever had to deal with making his
-- or her code readable in more than one locale, the process of
-- sufficiently abstracting program messages from their
-- presentation to the user is non-trivial without help from the
-- programming language. Most modern programming language
-- libraries do provide some mechanism for performing this
-- separation.</p>
--
-- <p>A portable API that allows a piece of code to run without
-- modification in different countries and under different
-- languages is a must for any non-trivial software project.
-- The interface should separate the logic of a program from
-- the myriad of translations that may be necessary.</p>
--
-- <p>The interface described in this document provides such
-- functionality. The underlying implementation is also allowed to
-- use whatever datastructures it likes to provide access to the
-- translations in the most efficient manner possible. In
-- addition, the implementation is provided with standardized
-- functions that programs will use for accessing an external,
-- unspecified repository of translations.</p>
--
-- <p>This interface <i>does not</i> cover all aspects of
-- localization, including support for non-latin characters,
-- number and date formatting, etc. Such functionality is the
-- scope of a future SRFI that may extend this one.</p>
--
-- <H1>Dependencies</H1>
--
-- An SRFI-29 conformant implementation must also implement
-- SRFI-28, Basic Format Strings. Message templates are strings
-- that must be processed by the <tt>format</tt> function
-- specified in that SRFI.
--
-- <H1>Specification</H1>
--
-- <h3>Message Bundles</h3>
--
-- <p>A Message Bundle is a set of message templates and their
-- identifying keys. Each bundle contains one or more such
-- key/value pairs. The bundle itself is associated with a
-- <i>bundle specifier</i> which uniquely identifies the
-- bundle.</p>
--
-- <h3>Bundle Specifiers</h3>
--
-- <p>A Bundle Specifier is a Scheme list that describes, in order
-- of importance, the package and locale of a message bundle.
-- In most cases, a locale specifier will have between one
-- and three elements. The first element is a symbol denoting the
-- package for which this bundle applies. The second and third
-- elements denote a <i>locale</i>. The second element (first
-- element of the locale) if present, is the two letter, ISO 639-1
-- language code for the bundle. The third element, if present, is
-- a two letter ISO 3166-1 country code. In some cases, a
-- fourth element may be present, specifying the encoding used for
-- the bundle. All bundle specifier elements are Scheme
-- symbols.</p>
--
-- <p>If only one translation is provided, it should be designated
-- only by a package name, for example <tt>(mathlib)</tt>. This
-- translation is called the <i>default</i> translation.</p>
--
-- <h3>Bundle Searching</h3>
--
-- <p>When a message template is retrieved from a bundle, the
-- Scheme implementation will provide the locale under which the
-- system is currently running. When the template is retrieved,
-- the package name will be specified. The Scheme system should
-- construct a Bundle Specifier from the provided package name and
-- the active locale. For example, when retrieving a message
-- template for French Canadian, in the <tt>mathlib</tt> package,
-- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
-- program may also retrieve the elements of the current locale
-- using the no-argument procedures:</p>
--
-- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
-- <i>symbol</i></tt><br>
-- <tt><b>current-language</b> <i>symbol</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- When given no arguments, returns the current ISO 639-1
-- language code as a symbol. If provided with an
-- argument, the current language is set to that named by the
-- symbol for the currently executing Scheme thread (or for the
-- entire Scheme system if such a distinction is not possible).
--
-- </blockquote>
--
-- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
-- <i>symbol</i></tt><br>
-- <tt><b>current-country</b> <i>symbol</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- returns the current ISO 3166-1 country code as a symbol.
-- If provided with an argument, the current country is
-- set to that named by the symbol for the currently executing
-- Scheme thread (or for the entire Scheme system if such a
-- distinction is not possible).
-- </blockquote>
--
-- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
-- symbol</i></tt>s<br>
-- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- Returns a list of additional locale details as a list of
-- symbols. This list may contain information about
-- encodings or other more specific information. If
-- provided with an argument, the current locale details are set
-- to those given in the currently executing Scheme thread (or
-- for the entire Scheme system if such a distinction is not
-- possible).
-- </blockquote>
--
-- <p>The Scheme System should first check for a bundle with the
-- exact name provided. If no such bundle is found, the last
-- element from the list is removed and a search is tried for a
-- bundle with that name. If no bundle is then found, the list is
-- shortened by removing the last element again. If no message is
-- found and the bundle specifier is now the empty list, an error
-- should be raised.</p>
--
-- <p>The reason for this search order is to provide the most
-- locale sensitive template possible, but to fall back on more
-- general templates if a translation has not yet been provided
-- for the given locale.</p>
--
-- <h3>Message Templates</h3>
--
-- <p>A message template is a localized message that may or may
-- not contain one of a number of formatting codes. A message
-- template is a Scheme string. The string is of a form that can
-- be processed by the <tt>format</tt> procedure found in many
-- Scheme systems and formally specified in SRFI-28 (Basic Format
-- Strings).</p>
--
-- <p>This SRFI also extends SRFI-28 to provide an additional
-- <tt>format</tt> escape code:</p>
--
-- <blockquote>
-- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
-- follows this code immediately to reference the [N]'th
-- optional value absolutely, rather than the next unconsumed
-- value. The referenced value is <i>not</i> consumed.
-- </blockquote>
-- This extension allows optional values to be positionally
-- referenced, so that message templates can be constructed that
-- can produce the proper word ordering for a language.
--
-- <h3>Preparing Bundles</h3>
-- Before a bundle may be used by the Scheme system to retrieve
-- localized template messages, they must be made available to the
-- Scheme system. This SRFI specifies a way to portably
-- define the bundles, as well as store them in and retrieve them
-- from an unspecified system which may be provided by resources
-- outside the Scheme system.<br>
--
--
-- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
-- association-list</i> -> undefined<br>
-- </tt></p>
--
-- <blockquote>
-- Declares a new bundle named by the given bundle-specifier.
-- The contents of the bundle are defined by the provided
-- association list. The list contains associations
-- between Scheme symbols and the message templates (Scheme
-- strings) they name. If a bundle already exists with the
-- given name, it is overwritten with the newly declared
-- bundle.<br>
-- </blockquote>
-- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
-- boolean</tt><br>
--
--
-- <blockquote>
-- Attempts to store a bundle named by the given bundle
-- specifier, and previously made available using
-- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
-- unspecified mechanism that may be persistent across Scheme
-- system restarts. If successful, a non-false value is
-- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
-- </blockquote>
-- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
-- boolean</tt><br>
--
--
-- <blockquote>
-- Attempts to retrieve a bundle from an unspecified mechanism
-- which stores bundles outside the Scheme system. If the
-- bundle was retrieved successfully, the function returns a
-- non-false value, and the bundle is immediately available to
-- the Scheme system. If the bundle could not be found or loaded
-- successfully, the function returns <tt>#f</tt>, and the
-- Scheme system's bundle registry remains unaffected.<br>
-- </blockquote>
-- A compliant Scheme system may choose not to provide any
-- external mechanism to store localized bundles. If it does
-- not, it must still provide implementations for
-- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
-- case, both functions must return <tt>#f</tt> regardless of the
-- arguments given. Users of this SRFI should recognize that the
-- inability to load or store a localized bundle in an external
-- repository is <i>not</i> a fatal error.<br>
--
--
-- <h3>Retrieving Localized Message Templates</h3>
--
-- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
-- message-template-name</tt></i> <tt>-> <i>string or #f<br>
-- </i></tt></p>
--
-- <blockquote>
-- Retrieves a localized message template for the given package
-- name and the given message template name (both symbols).
-- If no such message could be found, false (#f) is
-- returned.<br>
-- <br>
-- </blockquote>
-- After retrieving a template, the calling program can use
-- <tt>format</tt> to produce a string that can be displayed to
-- the user.<br>
--
--
-- <h2>Examples</h2>
-- The below example makes use of SRFI-29 to display simple,
-- localized messages. It also defines its bundles in such a
-- way that the Scheme system may store and retrieve the bundles
-- from a more efficient system catalog, if available.<br>
--
--<pre>
--(let ((translations
-- '(((en) . ((time . "Its ~a, ~a.")
-- (goodbye . "Goodbye, ~a.")))
-- ((fr) . ((time . "~1@*~a, c'est ~a.")
-- (goodbye . "Au revoir, ~a."))))))
-- (for-each (lambda (translation)
-- (let ((bundle-name (cons 'hello-program (car translation))))
-- (if (not (load-bundle! bundle-name))
-- (begin
-- (declare-bundle! bundle-name (cdr translation))
-- (store-bundle! bundle-name)))))
-- translations))
--
--(define localized-message
-- (lambda (message-name . args)
-- (apply format (cons (localized-template 'hello-program
-- message-name)
-- args))))
--
--(let ((myname "Fred"))
-- (display (localized-message 'time "12:00" myname))
-- (display #\newline)
--
-- (display (localized-message 'goodbye myname))
-- (display #\newline))
--
--;; Displays (English):
--;; Its 12:00, Fred.
--;; Goodbye, Fred.
--;;
--;; French:
--;; Fred, c'est 12:00.
--;; Au revoir, Fred.
--</pre>
--
-- <H1>Implementation</H1>
--
-- <p>The implementation requires that the Scheme system provide a
-- definition for <tt>current-language</tt> and
-- <tt>current-country</tt> capable of distinguishing the correct
-- locale present during a Scheme session. The definitions of
-- those functions in the reference implementation are not capable
-- of that distinction. Their implementation is provided only so
-- that the following code can run in any R4RS scheme system.
-- <br>
-- </p>
--
-- <p>In addition, the below implementation of a compliant
-- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
-- SRFI-23 (Error reporting)</p>
--<pre>
--;; The association list in which bundles will be stored
--(define *localization-bundles* '())
--
--;; The current-language and current-country functions provided
--;; here must be rewritten for each Scheme system to default to the
--;; actual locale of the session
--(define current-language
-- (let ((current-language-value 'en))
-- (lambda args
-- (if (null? args)
-- current-language-value
-- (set! current-language-value (car args))))))
--
--(define current-country
-- (let ((current-country-value 'us))
-- (lambda args
-- (if (null? args)
-- current-country-value
-- (set! current-country-value (car args))))))
--
--;; The load-bundle! and store-bundle! both return #f in this
--;; reference implementation. A compliant implementation need
--;; not rewrite these procedures.
--(define load-bundle!
-- (lambda (bundle-specifier)
-- #f))
--
--(define store-bundle!
-- (lambda (bundle-specifier)
-- #f))
--
--;; Declare a bundle of templates with a given bundle specifier
--(define declare-bundle!
-- (letrec ((remove-old-bundle
-- (lambda (specifier bundle)
-- (cond ((null? bundle) '())
-- ((equal? (caar bundle) specifier)
-- (cdr bundle))
-- (else (cons (car bundle)
-- (remove-old-bundle specifier
-- (cdr bundle))))))))
-- (lambda (bundle-specifier bundle-assoc-list)
-- (set! *localization-bundles*
-- (cons (cons bundle-specifier bundle-assoc-list)
-- (remove-old-bundle bundle-specifier
-- *localization-bundles*))))))
--
--;;Retrieve a localized template given its package name and a template name
--(define localized-template
-- (letrec ((rdc
-- (lambda (ls)
-- (if (null? (cdr ls))
-- '()
-- (cons (car ls) (rdc (cdr ls))))))
-- (find-bundle
-- (lambda (specifier template-name)
-- (cond ((assoc specifier *localization-bundles*) =>
-- (lambda (bundle) bundle))
-- ((null? specifier) #f)
-- (else (find-bundle (rdc specifier)
-- template-name))))))
-- (lambda (package-name template-name)
-- (let loop ((specifier (cons package-name
-- (list (current-language)
-- (current-country)))))
-- (and (not (null? specifier))
-- (let ((bundle (find-bundle specifier template-name)))
-- (and bundle
-- (cond ((assq template-name bundle) => cdr)
-- ((null? (cdr specifier)) #f)
-- (else (loop (rdc specifier)))))))))))
--
--;;An SRFI-28 and SRFI-29 compliant version of format. It requires
--;;SRFI-23 for error reporting.
--(define format
-- (lambda (format-string . objects)
-- (let ((buffer (open-output-string)))
-- (let loop ((format-list (string->list format-string))
-- (objects objects)
-- (object-override #f))
-- (cond ((null? format-list) (get-output-string buffer))
-- ((char=? (car format-list) #\~)
-- (cond ((null? (cdr format-list))
-- (error 'format "Incomplete escape sequence"))
-- ((char-numeric? (cadr format-list))
-- (let posloop ((fl (cddr format-list))
-- (pos (string->number
-- (string (cadr format-list)))))
-- (cond ((null? fl)
-- (error 'format "Incomplete escape sequence"))
-- ((and (eq? (car fl) '#\@)
-- (null? (cdr fl)))
-- (error 'format "Incomplete escape sequence"))
-- ((and (eq? (car fl) '#\@)
-- (eq? (cadr fl) '#\*))
-- (loop (cddr fl) objects (list-ref objects pos)))
-- (else
-- (posloop (cdr fl)
-- (+ (* 10 pos)
-- (string->number
-- (string (car fl)))))))))
-- (else
-- (case (cadr format-list)
-- ((#\a)
-- (cond (object-override
-- (begin
-- (display object-override buffer)
-- (loop (cddr format-list) objects #f)))
-- ((null? objects)
-- (error 'format "No value for escape sequence"))
-- (else
-- (begin
-- (display (car objects) buffer)
-- (loop (cddr format-list)
-- (cdr objects) #f)))))
-- ((#\s)
-- (cond (object-override
-- (begin
-- (display object-override buffer)
-- (loop (cddr format-list) objects #f)))
-- ((null? objects)
-- (error 'format "No value for escape sequence"))
-- (else
-- (begin
-- (write (car objects) buffer)
-- (loop (cddr format-list)
-- (cdr objects) #f)))))
-- ((#\%)
-- (if object-override
-- (error 'format "Escape sequence following positional override does not require a value"))
-- (display #\newline buffer)
-- (loop (cddr format-list) objects #f))
-- ((#\~)
-- (if object-override
-- (error 'format "Escape sequence following positional override does not require a value"))
-- (display #\~ buffer)
-- (loop (cddr format-list) objects #f))
-- (else
-- (error 'format "Unrecognized escape sequence"))))))
-- (else (display (car format-list) buffer)
-- (loop (cdr format-list) objects #f)))))))
--
--</pre>
--
-- <H1>Copyright</H1>
--
-- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
--
-- <p>This document and translations of it may be copied and
-- furnished to others, and derivative works that comment on or
-- otherwise explain it or assist in its implementation may be
-- prepared, copied, published and distributed, in whole or in
-- part, without restriction of any kind, provided that the above
-- copyright notice and this paragraph are included on all such
-- copies and derivative works. However, this document itself may
-- not be modified in any way, such as by removing the copyright
-- notice or references to the Scheme Request For Implementation
-- process or editors, except as needed for the purpose of
-- developing SRFIs in which case the procedures for copyrights
-- defined in the SRFI process must be followed, or as required to
-- translate it into languages other than English.</p>
--
-- <p>The limited permissions granted above are perpetual and will
-- not be revoked by the authors or their successors or
-- assigns.</p>
--
-- <p>This document and the information contained herein is
-- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
-- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
-- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
-- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
-- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
-- <hr>
--
-- <address>
-- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
-- Rush</a>
-- </address>
--
-- <address>
-- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
-- Miller</a>
-- </address>
-- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
-- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
-- Daylight Time 2002 <!-- hhmts end --> <br>
-- </body>
--</html>
--
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html
---- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
-@@ -1,345 +0,0 @@
--<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
--<html>
--<head>
-- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
--</head>
--<body>
--
--<H1>Title</H1>
--
--SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
--
--<H1>Author</H1>
--
--Andy Gaynor
--
--<H1>Status</H1>
--
--This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
--You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
--<P><UL>
--<LI>Received: 1999/2/2
--<LI>Draft: 1999/2/10-1999/04/12
--<LI>Final: 1999/4/26
--<LI>Revised reference implementation: 2003/01/27
--</UL>
--
--<H1>Abstract</H1>
--
--The <i>named-let</i> incarnation of the <code>let</code> form has two slight
--inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
--form makes no accommodation for rest arguments, an issue of functionality
--and consistency. As defined, the <code>let</code> form does not accommodate
--signature-style syntax, an issue of aesthetics and consistency. Both
--issues are addressed here in a manner which is compatible with the traditional
--<code>let</code> form but for minor extensions.
--
--<H1>Rationale</H1>
--
--<H2>Signature-style Syntax</H2>
--
--Consider the following two equivalent definitions:
--
--<p><pre>
--(define fibonacci
-- (lambda (n i f0 f1)
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
--
--(define (fibonacci n i f0 f1)
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--</pre>
--
--Although there is a named-let analog for the former form, there is none
--for the latter. To wit, suppose one wished to compute the 10th element
--of the Fibonacci sequence using a named let:
--
--<p>
--<pre>
--(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--Values: 55
--</pre>
--
--As it stands, one cannot equivalently write
--
--<p>
--<pre>
--(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--</pre>
--
--which is consistent with <code>define</code>'s signature-style form.
--<p>Those that favor the signature style may prefer this extension.
--In any case, it may be more appropriate to include all bound names within
--the binding section. As presented, this straightforward extension
--introduces no ambiguity or incompatibility with the existing definition
--of let.
--
--<H2>Rest Arguments</H2>
--
--As it stands, one cannot write a named let with rest arguments, as in
--
--<p>
--<pre>
--(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
-- (if (null? x)
-- 'just-a-silly-contrived-example
-- (begin
-- (write (car x) port)
-- (apply blast port (cdr x)))))
--</pre>
--
--otherwise equivalent to
--
--<p>
--<pre>
--(letrec ((blast (lambda (port . x)
-- (if (null? x)
-- 'just-a-silly-contrived-example
-- (begin
-- (write (car x) port)
-- (apply blast port (cdr x)))))))
-- (blast (current-output-port) (+ 1 2) 4 5))
--</pre>
--
--While this example is rather contrived, the functionality is not.
--There are several times when the author has used this construct in practice.
--Regardless, there is little reason to deny the <code>let</code> form access to
--all the features of lambda functionality.
--
--<H2>Symbols in Binding Sections</H2>
--
--Both the features above rely upon the placement of symbols in <code>let</code>
--binding lists (this statement is intentially simplistic). The only
--other apparent use of such symbol placement is to tersely bind variables
--to unspecified values. For example, one might desire to use
--<code>(let (foo bar baz) ...)</code>
--to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
--unspecified values.
--
--<p>This usage is considered less important in light of the rationales
--presented above, and an alternate syntax is immediately apparent, as
--in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
--be preferable, consistently parenthesizing normal binding clauses.
--
--<H1>Specification</H1>
--
--<H2>Syntax</H2>
--
--<p>
--A formal specification of the syntax follows. Below, body, expression,
--and identifier are free. Each instantiation of binding-name must be
--unique.
--</p>
--
--<p>
--<pre>
-- let = "(" "let" let-bindings body ")"
-- expressions = nothing | expression expressions
-- let-bindings = let-name bindings
-- |"(" let-name "." bindings ")"
-- let-name = identifier
-- bindings = "(" ")"
-- | rest-binding
-- | "(" normal-bindings ["." rest-binding] ")"
--normal-bindings = nothing
-- | normal-binding normal-bindings
--normal-binding = "(" binding-name expression ")"
-- binding-name = identifier
-- rest-binding = "(" binding-name expressions ")"
--</pre>
--
--<p>
--For clarity and convenience, an informal specification follows.
--</p>
--
--<ol>
--<li><a name="unnamed">Unnamed</a>
--
--<p><pre>
--(let ((<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-non-sig">
--Named, non-signature-style, no rest argument</a>
--
--<p><pre>
--(let <name> ((<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-sig">Named, signature-style, no rest argument</a>
--
--<p><pre>
--(let (<name> (<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
--
--<p><pre>
--(let <name> ((<parameter> <argument>)...
--
--. (<rest-parameter> <rest-argument>...))
-- <body>...)
--</pre>
--
--<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
--
--<p><pre>
--(let (<name> (<parameter> <argument>)...
--
--. (<rest-parameter> <rest-argument>...))
-- <body>...)
--</pre>
--</li>
--</ol>
--
--<H2>Semantics</H2>
--
--Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
--and <code>letrec</code> forms, respectively.
--
--<ul>
--<li>For informal syntax 1:
--
--<p><pre>
--(($lambda (<parameter>...) <body>...) <argument>...)
--</pre>
--</li>
--
--<li>For informal syntaxes 2 and 3:
--
--<p>
--<pre>
--($letrec ((<name> ($lambda (<parameter>...) <body>...)))
-- (<name> <argument>...))
--</pre>
--</li>
--
--<li>For informal syntaxes 4 and 5:
--
--<p>
--<pre>
--($letrec ((<name> ($lambda (<parameter>...
--
--. <rest-parameter>) <body>...)))
-- (<name> <argument>... <rest-argument>...))
--</pre>
--</li>
--</ul>
--
--<H1>Implementation</H1>
--
--Here is an implementation using <code>SYNTAX-RULES</code>.
--
--<p>
--<pre>
--;; Use your own standard let.
--;; Or call a lambda.
--;; (define-syntax standard-let
--;;
--;; (syntax-rules ()
--;;
--;; ((let ((var val) ...) body ...)
--;; ((lambda (var ...) body ...) val ...))))
--
--(define-syntax let
--
-- (syntax-rules ()
--
-- ;; No bindings: use standard-let.
-- ((let () body ...)
-- (standard-let () body ...))
-- ;; Or call a lambda.
-- ;; ((lambda () body ...))
--
-- ;; All standard bindings: use standard-let.
-- ((let ((var val) ...) body ...)
-- (standard-let ((var val) ...) body ...))
-- ;; Or call a lambda.
-- ;; ((lambda (var ...) body ...) val ...)
--
-- ;; One standard binding: loop.
-- ;; The all-standard-bindings clause didn't match,
-- ;; so there must be a rest binding.
-- ((let ((var val) . bindings) body ...)
-- (let-loop #f bindings (var) (val) (body ...)))
--
-- ;; Signature-style name: loop.
-- ((let (name binding ...) body ...)
-- (let-loop name (binding ...) () () (body ...)))
--
-- ;; defun-style name: loop.
-- ((let name bindings body ...)
-- (let-loop name bindings () () (body ...)))))
--
--(define-syntax let-loop
--
-- (syntax-rules ()
--
-- ;; Standard binding: destructure and loop.
-- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
-- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
--
-- ;; Rest binding, no name: use standard-let, listing the rest values.
-- ;; Because of let's first clause, there is no "no bindings, no name" clause.
-- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
-- ;; Or call a lambda with a rest parameter on all values.
-- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
-- ;; Or use one of several other reasonable alternatives.
--
-- ;; No bindings, name: call a letrec'ed lambda.
-- ((let-loop name () (var ...) (val ...) body)
-- ((letrec ((name (lambda (var ...) . body)))
-- name)
-- val ...))
--
-- ;; Rest binding, name: call a letrec'ed lambda.
-- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
-- ((letrec ((name (lambda (var ... . rest-var) . body)))
-- name)
-- val ... rest-val ...))))
--</pre>
--
--
--<H1>Copyright</H1>
--
--Copyright (C) Andy Gaynor (1999). All Rights Reserved.
--<p>This document and translations of it may be copied and furnished to
--others, and derivative works that comment on or otherwise explain it or
--assist in its implementation may be prepared, copied, published and distributed,
--in whole or in part, without restriction of any kind, provided that the
--above copyright notice and this paragraph are included on all such copies
--and derivative works. However, this document itself may not be modified
--in any way, such as by removing the copyright notice or references to the
--Scheme Request For Implementation process or editors, except as needed
--for the purpose of developing SRFIs in which case the procedures for copyrights
--defined in the SRFI process must be followed, or as required to translate
--it into languages other than English.
--<p>The limited permissions granted above are perpetual and will not be
--revoked by the authors or their successors or assigns.
--<p>This document and the information contained herein is provided on an
--"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
--EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
--USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
--WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
--
-- <hr>
-- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
--
--</body>
--</html>
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/util.rkt racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/util.rkt
---- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/util.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/util.rkt 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,61 @@
-+#lang scribble/doc
-+
-+@(require scribble/manual
-+ scribble/eval
-+ scriblib/render-cond
-+ scribble/core
-+ scribble/html-properties
-+ (for-syntax scheme/base)
-+ (for-label scheme/base
-+ racket/stream))
-+
-+@(provide (all-defined-out))
-+
-+@(define-syntax (srfi stx)
-+ (syntax-case stx ()
-+ [(_ num #:subdir subdir? . title)
-+ (with-syntax ([srfi/n (string->symbol (format "srfi/~a" (syntax-e #'num)))])
-+ #'(begin
-+ (section #:tag (format "srfi-~a" num)
-+ #:style 'unnumbered
-+ (format "SRFI ~a: " num)
-+ . title)
-+ (defmodule srfi/n)
-+ "Original specification: "
-+ (let* ([label (format "SRFI ~a" num)]
-+ [sub (if subdir? (format "srfi-~a/" num) "")]
-+ [url ( (b) (format "~a/srfi-std/~asrfi-~a.html" b sub num))])
-+ (cond-element
-+ [(or latex text) @link[(url "http://docs.racket-lang.org") label]]
-+ [else @link[(url ".") label]]))))]
-+ [(_ num . title) #'(srfi num #:subdir #f . title)]))
-+
-+@;{ The `lst' argument is a list of
-+ (list sym syntactic-form? html-anchor) }
-+@(define (redirect n lst #:subdir [subdir? #f])
-+ (let ([file (if subdir?
-+ (format "srfi-~a/srfi-~a.html" n n)
-+ (format "srfi-~a.html" n))]
-+ [mod-path (string->symbol (format "srfi/~a" n))])
-+ (make-binding-redirect-elements mod-path
-+ (map (lambda (b)
-+ (list (car b) (cadr b)
-+ (build-path "srfi-std" file)
-+ (caddr b)))
-+ lst))))
-+
-+@(define in-core
-+ (case-lambda
-+ [() (in-core ".")]
-+ [(k) @elem{This SRFI's bindings are also available in
-+ @racketmodname[racket/base]@|k|}]))
-+
-+@(begin
-+ (define-syntax-rule (def-mz mz-if)
-+ (begin
-+ (require (for-label mzscheme))
-+ (define mz-if (racket if))))
-+ (def-mz mz-if))
-+
-+@(define srfi-std (style #f (list (install-resource "srfi-std"))))
-+
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,19 @@
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define build-deps '("mzscheme-doc"
-+ "scheme-lib"
-+ "base"
-+ "scribble-lib"
-+ "srfi-doc"
-+ "srfi-lib-nonfree"
-+ "racket-doc"
-+ "r5rs-doc"
-+ "r6rs-doc"
-+ "compatibility-lib"))
-+(define update-implies '("srfi-lib-nonfree"))
-+
-+(define pkg-desc "documentation part of \"srfi nonfree\"")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,3 @@
-+#lang info
-+
-+(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,66 @@
-+#lang scribble/doc
-+@(require srfi/scribblings/util
-+ scribble/manual
-+ scribble/eval
-+ scriblib/render-cond
-+ scribble/core
-+ scribble/html-properties
-+ (for-syntax scheme/base)
-+ (for-label scheme/base
-+ racket/stream))
-+
-+@; ----------------------------------------------------------------------
-+
-+@title{SRFI Nonfree Libraries and Documentation}
-+
-+The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
-+Implementation} (a.k.a. @deftech{SRFI}) process allows individual
-+members of the Scheme community to propose libraries and extensions to
-+be supported by multiple Scheme implementations.
-+
-+Racket is distributed with implementations of many SRFIs, most of
-+which can be implemented as libraries. To import the bindings of SRFI
-+@math{n}, use
-+
-+@racketblock[
-+(require @#,elem{@racketidfont{srfi/}@math{n}})
-+]
-+
-+This document lists the SRFIs that are supported by Racket and
-+provides a link to the original SRFI specification (which is also
-+distributed as part of Racket's documentation).
-+
-+The following SRFI specification documents are licensed restrictively.
-+
-+@table-of-contents[]
-+
-+
-+@; ----------------------------------------
-+
-+@srfi[5]{A compatible let form with signatures and rest arguments}
-+
-+@redirect[5 '(
-+ (let #t "unnamed")
-+)]
-+
-+Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
-+
-+@; ----------------------------------------
-+
-+@srfi[29]{Localization}
-+
-+@redirect[29 '(
-+ (current-language #f "current-language")
-+ (current-country #f "current-country")
-+ (current-locale-details #f "current-locale-details")
-+ (declare-bundle! #f "declare-bundle!")
-+ (store-bundle #f "store-bundle")
-+ (load-bundle! #f "load-bundle!")
-+ (localized-template #f "localized-template")
-+)]
-+
-+Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
-+
-+@; ----------------------------------------
-+
-+@index-section[]
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,507 @@
-+<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
-+<html>
-+ <head>
-+ <meta name="generator" content="HTML Tidy, see www.w3.org">
-+ <title>SRFI 29: Localization</title>
-+ <meta name="author" content="Scott G. Miller">
-+ <meta name="description" content="Localization">
-+ </head>
-+ <body>
-+ <H1>Title</H1>
-+
-+ SRFI 29: Localization
-+
-+ <H1>Author</H1>
-+
-+ Scott G. Miller
-+
-+ <H1>Abstract</H1>
-+
-+ This document specifies an interface to retrieving and
-+ displaying locale sensitive messages. A Scheme program can
-+ register one or more translations of templated messages, and
-+ then write Scheme code that can transparently retrieve the
-+ appropriate message for the locale under which the Scheme
-+ system is running. <br>
-+
-+
-+ <H1>Rationale</H1>
-+
-+ <p>As any programmer that has ever had to deal with making his
-+ or her code readable in more than one locale, the process of
-+ sufficiently abstracting program messages from their
-+ presentation to the user is non-trivial without help from the
-+ programming language. Most modern programming language
-+ libraries do provide some mechanism for performing this
-+ separation.</p>
-+
-+ <p>A portable API that allows a piece of code to run without
-+ modification in different countries and under different
-+ languages is a must for any non-trivial software project.
-+ The interface should separate the logic of a program from
-+ the myriad of translations that may be necessary.</p>
-+
-+ <p>The interface described in this document provides such
-+ functionality. The underlying implementation is also allowed to
-+ use whatever datastructures it likes to provide access to the
-+ translations in the most efficient manner possible. In
-+ addition, the implementation is provided with standardized
-+ functions that programs will use for accessing an external,
-+ unspecified repository of translations.</p>
-+
-+ <p>This interface <i>does not</i> cover all aspects of
-+ localization, including support for non-latin characters,
-+ number and date formatting, etc. Such functionality is the
-+ scope of a future SRFI that may extend this one.</p>
-+
-+ <H1>Dependencies</H1>
-+
-+ An SRFI-29 conformant implementation must also implement
-+ SRFI-28, Basic Format Strings. Message templates are strings
-+ that must be processed by the <tt>format</tt> function
-+ specified in that SRFI.
-+
-+ <H1>Specification</H1>
-+
-+ <h3>Message Bundles</h3>
-+
-+ <p>A Message Bundle is a set of message templates and their
-+ identifying keys. Each bundle contains one or more such
-+ key/value pairs. The bundle itself is associated with a
-+ <i>bundle specifier</i> which uniquely identifies the
-+ bundle.</p>
-+
-+ <h3>Bundle Specifiers</h3>
-+
-+ <p>A Bundle Specifier is a Scheme list that describes, in order
-+ of importance, the package and locale of a message bundle.
-+ In most cases, a locale specifier will have between one
-+ and three elements. The first element is a symbol denoting the
-+ package for which this bundle applies. The second and third
-+ elements denote a <i>locale</i>. The second element (first
-+ element of the locale) if present, is the two letter, ISO 639-1
-+ language code for the bundle. The third element, if present, is
-+ a two letter ISO 3166-1 country code. In some cases, a
-+ fourth element may be present, specifying the encoding used for
-+ the bundle. All bundle specifier elements are Scheme
-+ symbols.</p>
-+
-+ <p>If only one translation is provided, it should be designated
-+ only by a package name, for example <tt>(mathlib)</tt>. This
-+ translation is called the <i>default</i> translation.</p>
-+
-+ <h3>Bundle Searching</h3>
-+
-+ <p>When a message template is retrieved from a bundle, the
-+ Scheme implementation will provide the locale under which the
-+ system is currently running. When the template is retrieved,
-+ the package name will be specified. The Scheme system should
-+ construct a Bundle Specifier from the provided package name and
-+ the active locale. For example, when retrieving a message
-+ template for French Canadian, in the <tt>mathlib</tt> package,
-+ the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
-+ program may also retrieve the elements of the current locale
-+ using the no-argument procedures:</p>
-+
-+ <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
-+ <i>symbol</i></tt><br>
-+ <tt><b>current-language</b> <i>symbol</i> ->
-+ undefined</tt><br>
-+ </p>
-+
-+ <blockquote>
-+ When given no arguments, returns the current ISO 639-1
-+ language code as a symbol. If provided with an
-+ argument, the current language is set to that named by the
-+ symbol for the currently executing Scheme thread (or for the
-+ entire Scheme system if such a distinction is not possible).
-+
-+ </blockquote>
-+
-+ <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
-+ <i>symbol</i></tt><br>
-+ <tt><b>current-country</b> <i>symbol</i> ->
-+ undefined</tt><br>
-+ </p>
-+
-+ <blockquote>
-+ returns the current ISO 3166-1 country code as a symbol.
-+ If provided with an argument, the current country is
-+ set to that named by the symbol for the currently executing
-+ Scheme thread (or for the entire Scheme system if such a
-+ distinction is not possible).
-+ </blockquote>
-+
-+ <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
-+ symbol</i></tt>s<br>
-+ <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
-+ undefined</tt><br>
-+ </p>
-+
-+ <blockquote>
-+ Returns a list of additional locale details as a list of
-+ symbols. This list may contain information about
-+ encodings or other more specific information. If
-+ provided with an argument, the current locale details are set
-+ to those given in the currently executing Scheme thread (or
-+ for the entire Scheme system if such a distinction is not
-+ possible).
-+ </blockquote>
-+
-+ <p>The Scheme System should first check for a bundle with the
-+ exact name provided. If no such bundle is found, the last
-+ element from the list is removed and a search is tried for a
-+ bundle with that name. If no bundle is then found, the list is
-+ shortened by removing the last element again. If no message is
-+ found and the bundle specifier is now the empty list, an error
-+ should be raised.</p>
-+
-+ <p>The reason for this search order is to provide the most
-+ locale sensitive template possible, but to fall back on more
-+ general templates if a translation has not yet been provided
-+ for the given locale.</p>
-+
-+ <h3>Message Templates</h3>
-+
-+ <p>A message template is a localized message that may or may
-+ not contain one of a number of formatting codes. A message
-+ template is a Scheme string. The string is of a form that can
-+ be processed by the <tt>format</tt> procedure found in many
-+ Scheme systems and formally specified in SRFI-28 (Basic Format
-+ Strings).</p>
-+
-+ <p>This SRFI also extends SRFI-28 to provide an additional
-+ <tt>format</tt> escape code:</p>
-+
-+ <blockquote>
-+ <tt>~[n]@*</tt> - Causes a value-requiring escape code that
-+ follows this code immediately to reference the [N]'th
-+ optional value absolutely, rather than the next unconsumed
-+ value. The referenced value is <i>not</i> consumed.
-+ </blockquote>
-+ This extension allows optional values to be positionally
-+ referenced, so that message templates can be constructed that
-+ can produce the proper word ordering for a language.
-+
-+ <h3>Preparing Bundles</h3>
-+ Before a bundle may be used by the Scheme system to retrieve
-+ localized template messages, they must be made available to the
-+ Scheme system. This SRFI specifies a way to portably
-+ define the bundles, as well as store them in and retrieve them
-+ from an unspecified system which may be provided by resources
-+ outside the Scheme system.<br>
-+
-+
-+ <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
-+ association-list</i> -> undefined<br>
-+ </tt></p>
-+
-+ <blockquote>
-+ Declares a new bundle named by the given bundle-specifier.
-+ The contents of the bundle are defined by the provided
-+ association list. The list contains associations
-+ between Scheme symbols and the message templates (Scheme
-+ strings) they name. If a bundle already exists with the
-+ given name, it is overwritten with the newly declared
-+ bundle.<br>
-+ </blockquote>
-+ <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
-+ boolean</tt><br>
-+
-+
-+ <blockquote>
-+ Attempts to store a bundle named by the given bundle
-+ specifier, and previously made available using
-+ <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
-+ unspecified mechanism that may be persistent across Scheme
-+ system restarts. If successful, a non-false value is
-+ returned. If unsuccessful, <tt>#f</tt> is returned.<br>
-+ </blockquote>
-+ <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
-+ boolean</tt><br>
-+
-+
-+ <blockquote>
-+ Attempts to retrieve a bundle from an unspecified mechanism
-+ which stores bundles outside the Scheme system. If the
-+ bundle was retrieved successfully, the function returns a
-+ non-false value, and the bundle is immediately available to
-+ the Scheme system. If the bundle could not be found or loaded
-+ successfully, the function returns <tt>#f</tt>, and the
-+ Scheme system's bundle registry remains unaffected.<br>
-+ </blockquote>
-+ A compliant Scheme system may choose not to provide any
-+ external mechanism to store localized bundles. If it does
-+ not, it must still provide implementations for
-+ <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
-+ case, both functions must return <tt>#f</tt> regardless of the
-+ arguments given. Users of this SRFI should recognize that the
-+ inability to load or store a localized bundle in an external
-+ repository is <i>not</i> a fatal error.<br>
-+
-+
-+ <h3>Retrieving Localized Message Templates</h3>
-+
-+ <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
-+ message-template-name</tt></i> <tt>-> <i>string or #f<br>
-+ </i></tt></p>
-+
-+ <blockquote>
-+ Retrieves a localized message template for the given package
-+ name and the given message template name (both symbols).
-+ If no such message could be found, false (#f) is
-+ returned.<br>
-+ <br>
-+ </blockquote>
-+ After retrieving a template, the calling program can use
-+ <tt>format</tt> to produce a string that can be displayed to
-+ the user.<br>
-+
-+
-+ <h2>Examples</h2>
-+ The below example makes use of SRFI-29 to display simple,
-+ localized messages. It also defines its bundles in such a
-+ way that the Scheme system may store and retrieve the bundles
-+ from a more efficient system catalog, if available.<br>
-+
-+<pre>
-+(let ((translations
-+ '(((en) . ((time . "Its ~a, ~a.")
-+ (goodbye . "Goodbye, ~a.")))
-+ ((fr) . ((time . "~1@*~a, c'est ~a.")
-+ (goodbye . "Au revoir, ~a."))))))
-+ (for-each (lambda (translation)
-+ (let ((bundle-name (cons 'hello-program (car translation))))
-+ (if (not (load-bundle! bundle-name))
-+ (begin
-+ (declare-bundle! bundle-name (cdr translation))
-+ (store-bundle! bundle-name)))))
-+ translations))
-+
-+(define localized-message
-+ (lambda (message-name . args)
-+ (apply format (cons (localized-template 'hello-program
-+ message-name)
-+ args))))
-+
-+(let ((myname "Fred"))
-+ (display (localized-message 'time "12:00" myname))
-+ (display #\newline)
-+
-+ (display (localized-message 'goodbye myname))
-+ (display #\newline))
-+
-+;; Displays (English):
-+;; Its 12:00, Fred.
-+;; Goodbye, Fred.
-+;;
-+;; French:
-+;; Fred, c'est 12:00.
-+;; Au revoir, Fred.
-+</pre>
-+
-+ <H1>Implementation</H1>
-+
-+ <p>The implementation requires that the Scheme system provide a
-+ definition for <tt>current-language</tt> and
-+ <tt>current-country</tt> capable of distinguishing the correct
-+ locale present during a Scheme session. The definitions of
-+ those functions in the reference implementation are not capable
-+ of that distinction. Their implementation is provided only so
-+ that the following code can run in any R4RS scheme system.
-+ <br>
-+ </p>
-+
-+ <p>In addition, the below implementation of a compliant
-+ <tt>format</tt> requires SRFI-6 (Basic String Ports) and
-+ SRFI-23 (Error reporting)</p>
-+<pre>
-+;; The association list in which bundles will be stored
-+(define *localization-bundles* '())
-+
-+;; The current-language and current-country functions provided
-+;; here must be rewritten for each Scheme system to default to the
-+;; actual locale of the session
-+(define current-language
-+ (let ((current-language-value 'en))
-+ (lambda args
-+ (if (null? args)
-+ current-language-value
-+ (set! current-language-value (car args))))))
-+
-+(define current-country
-+ (let ((current-country-value 'us))
-+ (lambda args
-+ (if (null? args)
-+ current-country-value
-+ (set! current-country-value (car args))))))
-+
-+;; The load-bundle! and store-bundle! both return #f in this
-+;; reference implementation. A compliant implementation need
-+;; not rewrite these procedures.
-+(define load-bundle!
-+ (lambda (bundle-specifier)
-+ #f))
-+
-+(define store-bundle!
-+ (lambda (bundle-specifier)
-+ #f))
-+
-+;; Declare a bundle of templates with a given bundle specifier
-+(define declare-bundle!
-+ (letrec ((remove-old-bundle
-+ (lambda (specifier bundle)
-+ (cond ((null? bundle) '())
-+ ((equal? (caar bundle) specifier)
-+ (cdr bundle))
-+ (else (cons (car bundle)
-+ (remove-old-bundle specifier
-+ (cdr bundle))))))))
-+ (lambda (bundle-specifier bundle-assoc-list)
-+ (set! *localization-bundles*
-+ (cons (cons bundle-specifier bundle-assoc-list)
-+ (remove-old-bundle bundle-specifier
-+ *localization-bundles*))))))
-+
-+;;Retrieve a localized template given its package name and a template name
-+(define localized-template
-+ (letrec ((rdc
-+ (lambda (ls)
-+ (if (null? (cdr ls))
-+ '()
-+ (cons (car ls) (rdc (cdr ls))))))
-+ (find-bundle
-+ (lambda (specifier template-name)
-+ (cond ((assoc specifier *localization-bundles*) =>
-+ (lambda (bundle) bundle))
-+ ((null? specifier) #f)
-+ (else (find-bundle (rdc specifier)
-+ template-name))))))
-+ (lambda (package-name template-name)
-+ (let loop ((specifier (cons package-name
-+ (list (current-language)
-+ (current-country)))))
-+ (and (not (null? specifier))
-+ (let ((bundle (find-bundle specifier template-name)))
-+ (and bundle
-+ (cond ((assq template-name bundle) => cdr)
-+ ((null? (cdr specifier)) #f)
-+ (else (loop (rdc specifier)))))))))))
-+
-+;;An SRFI-28 and SRFI-29 compliant version of format. It requires
-+;;SRFI-23 for error reporting.
-+(define format
-+ (lambda (format-string . objects)
-+ (let ((buffer (open-output-string)))
-+ (let loop ((format-list (string->list format-string))
-+ (objects objects)
-+ (object-override #f))
-+ (cond ((null? format-list) (get-output-string buffer))
-+ ((char=? (car format-list) #\~)
-+ (cond ((null? (cdr format-list))
-+ (error 'format "Incomplete escape sequence"))
-+ ((char-numeric? (cadr format-list))
-+ (let posloop ((fl (cddr format-list))
-+ (pos (string->number
-+ (string (cadr format-list)))))
-+ (cond ((null? fl)
-+ (error 'format "Incomplete escape sequence"))
-+ ((and (eq? (car fl) '#\@)
-+ (null? (cdr fl)))
-+ (error 'format "Incomplete escape sequence"))
-+ ((and (eq? (car fl) '#\@)
-+ (eq? (cadr fl) '#\*))
-+ (loop (cddr fl) objects (list-ref objects pos)))
-+ (else
-+ (posloop (cdr fl)
-+ (+ (* 10 pos)
-+ (string->number
-+ (string (car fl)))))))))
-+ (else
-+ (case (cadr format-list)
-+ ((#\a)
-+ (cond (object-override
-+ (begin
-+ (display object-override buffer)
-+ (loop (cddr format-list) objects #f)))
-+ ((null? objects)
-+ (error 'format "No value for escape sequence"))
-+ (else
-+ (begin
-+ (display (car objects) buffer)
-+ (loop (cddr format-list)
-+ (cdr objects) #f)))))
-+ ((#\s)
-+ (cond (object-override
-+ (begin
-+ (display object-override buffer)
-+ (loop (cddr format-list) objects #f)))
-+ ((null? objects)
-+ (error 'format "No value for escape sequence"))
-+ (else
-+ (begin
-+ (write (car objects) buffer)
-+ (loop (cddr format-list)
-+ (cdr objects) #f)))))
-+ ((#\%)
-+ (if object-override
-+ (error 'format "Escape sequence following positional override does not require a value"))
-+ (display #\newline buffer)
-+ (loop (cddr format-list) objects #f))
-+ ((#\~)
-+ (if object-override
-+ (error 'format "Escape sequence following positional override does not require a value"))
-+ (display #\~ buffer)
-+ (loop (cddr format-list) objects #f))
-+ (else
-+ (error 'format "Unrecognized escape sequence"))))))
-+ (else (display (car format-list) buffer)
-+ (loop (cdr format-list) objects #f)))))))
-+
-+</pre>
-+
-+ <H1>Copyright</H1>
-+
-+ Copyright (C) Scott G. Miller (2002). All Rights Reserved.
-+
-+ <p>This document and translations of it may be copied and
-+ furnished to others, and derivative works that comment on or
-+ otherwise explain it or assist in its implementation may be
-+ prepared, copied, published and distributed, in whole or in
-+ part, without restriction of any kind, provided that the above
-+ copyright notice and this paragraph are included on all such
-+ copies and derivative works. However, this document itself may
-+ not be modified in any way, such as by removing the copyright
-+ notice or references to the Scheme Request For Implementation
-+ process or editors, except as needed for the purpose of
-+ developing SRFIs in which case the procedures for copyrights
-+ defined in the SRFI process must be followed, or as required to
-+ translate it into languages other than English.</p>
-+
-+ <p>The limited permissions granted above are perpetual and will
-+ not be revoked by the authors or their successors or
-+ assigns.</p>
-+
-+ <p>This document and the information contained herein is
-+ provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
-+ EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
-+ BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
-+ HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
-+ OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
-+ <hr>
-+
-+ <address>
-+ Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
-+ Rush</a>
-+ </address>
-+
-+ <address>
-+ Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
-+ Miller</a>
-+ </address>
-+ <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
-+ <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
-+ Daylight Time 2002 <!-- hhmts end --> <br>
-+ </body>
-+</html>
-+
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,345 @@
-+<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
-+<html>
-+<head>
-+ <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
-+</head>
-+<body>
-+
-+<H1>Title</H1>
-+
-+SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
-+
-+<H1>Author</H1>
-+
-+Andy Gaynor
-+
-+<H1>Status</H1>
-+
-+This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
-+You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
-+<P><UL>
-+<LI>Received: 1999/2/2
-+<LI>Draft: 1999/2/10-1999/04/12
-+<LI>Final: 1999/4/26
-+<LI>Revised reference implementation: 2003/01/27
-+</UL>
-+
-+<H1>Abstract</H1>
-+
-+The <i>named-let</i> incarnation of the <code>let</code> form has two slight
-+inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
-+form makes no accommodation for rest arguments, an issue of functionality
-+and consistency. As defined, the <code>let</code> form does not accommodate
-+signature-style syntax, an issue of aesthetics and consistency. Both
-+issues are addressed here in a manner which is compatible with the traditional
-+<code>let</code> form but for minor extensions.
-+
-+<H1>Rationale</H1>
-+
-+<H2>Signature-style Syntax</H2>
-+
-+Consider the following two equivalent definitions:
-+
-+<p><pre>
-+(define fibonacci
-+ (lambda (n i f0 f1)
-+ (if (= i n)
-+ f0
-+ (fibonacci n (+ i 1) f1 (+ f0 f1)))))
-+
-+(define (fibonacci n i f0 f1)
-+ (if (= i n)
-+ f0
-+ (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+</pre>
-+
-+Although there is a named-let analog for the former form, there is none
-+for the latter. To wit, suppose one wished to compute the 10th element
-+of the Fibonacci sequence using a named let:
-+
-+<p>
-+<pre>
-+(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
-+ (if (= i n)
-+ f0
-+ (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+Values: 55
-+</pre>
-+
-+As it stands, one cannot equivalently write
-+
-+<p>
-+<pre>
-+(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
-+ (if (= i n)
-+ f0
-+ (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+</pre>
-+
-+which is consistent with <code>define</code>'s signature-style form.
-+<p>Those that favor the signature style may prefer this extension.
-+In any case, it may be more appropriate to include all bound names within
-+the binding section. As presented, this straightforward extension
-+introduces no ambiguity or incompatibility with the existing definition
-+of let.
-+
-+<H2>Rest Arguments</H2>
-+
-+As it stands, one cannot write a named let with rest arguments, as in
-+
-+<p>
-+<pre>
-+(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
-+ (if (null? x)
-+ 'just-a-silly-contrived-example
-+ (begin
-+ (write (car x) port)
-+ (apply blast port (cdr x)))))
-+</pre>
-+
-+otherwise equivalent to
-+
-+<p>
-+<pre>
-+(letrec ((blast (lambda (port . x)
-+ (if (null? x)
-+ 'just-a-silly-contrived-example
-+ (begin
-+ (write (car x) port)
-+ (apply blast port (cdr x)))))))
-+ (blast (current-output-port) (+ 1 2) 4 5))
-+</pre>
-+
-+While this example is rather contrived, the functionality is not.
-+There are several times when the author has used this construct in practice.
-+Regardless, there is little reason to deny the <code>let</code> form access to
-+all the features of lambda functionality.
-+
-+<H2>Symbols in Binding Sections</H2>
-+
-+Both the features above rely upon the placement of symbols in <code>let</code>
-+binding lists (this statement is intentially simplistic). The only
-+other apparent use of such symbol placement is to tersely bind variables
-+to unspecified values. For example, one might desire to use
-+<code>(let (foo bar baz) ...)</code>
-+to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
-+unspecified values.
-+
-+<p>This usage is considered less important in light of the rationales
-+presented above, and an alternate syntax is immediately apparent, as
-+in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
-+be preferable, consistently parenthesizing normal binding clauses.
-+
-+<H1>Specification</H1>
-+
-+<H2>Syntax</H2>
-+
-+<p>
-+A formal specification of the syntax follows. Below, body, expression,
-+and identifier are free. Each instantiation of binding-name must be
-+unique.
-+</p>
-+
-+<p>
-+<pre>
-+ let = "(" "let" let-bindings body ")"
-+ expressions = nothing | expression expressions
-+ let-bindings = let-name bindings
-+ |"(" let-name "." bindings ")"
-+ let-name = identifier
-+ bindings = "(" ")"
-+ | rest-binding
-+ | "(" normal-bindings ["." rest-binding] ")"
-+normal-bindings = nothing
-+ | normal-binding normal-bindings
-+normal-binding = "(" binding-name expression ")"
-+ binding-name = identifier
-+ rest-binding = "(" binding-name expressions ")"
-+</pre>
-+
-+<p>
-+For clarity and convenience, an informal specification follows.
-+</p>
-+
-+<ol>
-+<li><a name="unnamed">Unnamed</a>
-+
-+<p><pre>
-+(let ((<parameter> <argument>)...)
-+ <body>...)
-+</pre>
-+</li>
-+
-+<li><a name="named-non-sig">
-+Named, non-signature-style, no rest argument</a>
-+
-+<p><pre>
-+(let <name> ((<parameter> <argument>)...)
-+ <body>...)
-+</pre>
-+</li>
-+
-+<li><a name="named-sig">Named, signature-style, no rest argument</a>
-+
-+<p><pre>
-+(let (<name> (<parameter> <argument>)...)
-+ <body>...)
-+</pre>
-+</li>
-+
-+<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
-+
-+<p><pre>
-+(let <name> ((<parameter> <argument>)...
-+
-+. (<rest-parameter> <rest-argument>...))
-+ <body>...)
-+</pre>
-+
-+<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
-+
-+<p><pre>
-+(let (<name> (<parameter> <argument>)...
-+
-+. (<rest-parameter> <rest-argument>...))
-+ <body>...)
-+</pre>
-+</li>
-+</ol>
-+
-+<H2>Semantics</H2>
-+
-+Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
-+and <code>letrec</code> forms, respectively.
-+
-+<ul>
-+<li>For informal syntax 1:
-+
-+<p><pre>
-+(($lambda (<parameter>...) <body>...) <argument>...)
-+</pre>
-+</li>
-+
-+<li>For informal syntaxes 2 and 3:
-+
-+<p>
-+<pre>
-+($letrec ((<name> ($lambda (<parameter>...) <body>...)))
-+ (<name> <argument>...))
-+</pre>
-+</li>
-+
-+<li>For informal syntaxes 4 and 5:
-+
-+<p>
-+<pre>
-+($letrec ((<name> ($lambda (<parameter>...
-+
-+. <rest-parameter>) <body>...)))
-+ (<name> <argument>... <rest-argument>...))
-+</pre>
-+</li>
-+</ul>
-+
-+<H1>Implementation</H1>
-+
-+Here is an implementation using <code>SYNTAX-RULES</code>.
-+
-+<p>
-+<pre>
-+;; Use your own standard let.
-+;; Or call a lambda.
-+;; (define-syntax standard-let
-+;;
-+;; (syntax-rules ()
-+;;
-+;; ((let ((var val) ...) body ...)
-+;; ((lambda (var ...) body ...) val ...))))
-+
-+(define-syntax let
-+
-+ (syntax-rules ()
-+
-+ ;; No bindings: use standard-let.
-+ ((let () body ...)
-+ (standard-let () body ...))
-+ ;; Or call a lambda.
-+ ;; ((lambda () body ...))
-+
-+ ;; All standard bindings: use standard-let.
-+ ((let ((var val) ...) body ...)
-+ (standard-let ((var val) ...) body ...))
-+ ;; Or call a lambda.
-+ ;; ((lambda (var ...) body ...) val ...)
-+
-+ ;; One standard binding: loop.
-+ ;; The all-standard-bindings clause didn't match,
-+ ;; so there must be a rest binding.
-+ ((let ((var val) . bindings) body ...)
-+ (let-loop #f bindings (var) (val) (body ...)))
-+
-+ ;; Signature-style name: loop.
-+ ((let (name binding ...) body ...)
-+ (let-loop name (binding ...) () () (body ...)))
-+
-+ ;; defun-style name: loop.
-+ ((let name bindings body ...)
-+ (let-loop name bindings () () (body ...)))))
-+
-+(define-syntax let-loop
-+
-+ (syntax-rules ()
-+
-+ ;; Standard binding: destructure and loop.
-+ ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
-+ (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
-+
-+ ;; Rest binding, no name: use standard-let, listing the rest values.
-+ ;; Because of let's first clause, there is no "no bindings, no name" clause.
-+ ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-+ (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
-+ ;; Or call a lambda with a rest parameter on all values.
-+ ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
-+ ;; Or use one of several other reasonable alternatives.
-+
-+ ;; No bindings, name: call a letrec'ed lambda.
-+ ((let-loop name () (var ...) (val ...) body)
-+ ((letrec ((name (lambda (var ...) . body)))
-+ name)
-+ val ...))
-+
-+ ;; Rest binding, name: call a letrec'ed lambda.
-+ ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
-+ ((letrec ((name (lambda (var ... . rest-var) . body)))
-+ name)
-+ val ... rest-val ...))))
-+</pre>
-+
-+
-+<H1>Copyright</H1>
-+
-+Copyright (C) Andy Gaynor (1999). All Rights Reserved.
-+<p>This document and translations of it may be copied and furnished to
-+others, and derivative works that comment on or otherwise explain it or
-+assist in its implementation may be prepared, copied, published and distributed,
-+in whole or in part, without restriction of any kind, provided that the
-+above copyright notice and this paragraph are included on all such copies
-+and derivative works. However, this document itself may not be modified
-+in any way, such as by removing the copyright notice or references to the
-+Scheme Request For Implementation process or editors, except as needed
-+for the purpose of developing SRFIs in which case the procedures for copyrights
-+defined in the SRFI process must be followed, or as required to translate
-+it into languages other than English.
-+<p>The limited permissions granted above are perpetual and will not be
-+revoked by the authors or their successors or assigns.
-+<p>This document and the information contained herein is provided on an
-+"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
-+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
-+USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
-+WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
-+
-+ <hr>
-+ <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
-+
-+</body>
-+</html>
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/info.rkt racket-6.12/share/pkgs/srfi-lib/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/info.rkt 2018-01-26 16:10:02.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/info.rkt 2018-01-31 12:22:58.144502000 -0500
-@@ -1 +1,15 @@
--(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("scheme-lib" "base" "srfi-lite-lib" "r6rs-lib" "compatibility-lib"))) (define implies (quote ("srfi-lite-lib"))) (define pkg-desc "implementation (no documentation) part of \"srfi\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define deps '("scheme-lib"
-+ "base"
-+ "srfi-lite-lib"
-+ "r6rs-lib"
-+ "compatibility-lib"))
-+
-+(define implies '("srfi-lite-lib"))
-+
-+(define pkg-desc "implementation (no documentation) part of \"srfi\"")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/sort.txt racket-6.12/share/pkgs/srfi-lib/srfi/32/sort.txt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/sort.txt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/32/sort.txt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,1069 +0,0 @@
--The SRFI-32 sort libraries -*- outline -*-
--Olin Shivers
--First draft: 1998/10/19
--Last update: 2002/7/21
--
--[Todo: del-list-neighbor-dups!
-- vector-copy -> subvector
-- use srfi-23 for reporting errors
-- use srfi-16 for n-aries?
--
--Emacs should display this document in outline mode. Say c-h m for
--instructions on how to move through it by sections (e.g., c-c c-n, c-c c-p).
--
--* Table of contents
---------------------
--Abstract
--Procedure index
--Introduction
--What's wrong with the current state of affairs?
--Design rules
-- What vs. how
-- Consistency across function signatures
-- Data parameter first, less-than parameter after
-- Ordering, comparison functions & stability
-- All vector operations accept optional subrange parameters
-- Required vs. allowed side-effects
--Procedure specification
-- Procedure naming and functionality
-- Types of parameters and return values
-- sort-lib - general sorting package
-- Algorithm-specific sorting packages
--Algorithmic properties
--Topics to be resolved during discussion phase
--Porting and optimisation
--References & Links
--Acknowledgements
--Copyright
--
--
--* Abstract
------------
--Current Scheme sorting packages are, every one of them, surprisingly bad. I've
--designed the API for a full-featured sort toolkit, which I propose as a SRFI.
--
--The spec comes with 1200 lines of high-quality reference code: tightly
--written, highly commented, portable code, available for free. Implementors
--want this code. It's better than what you have.
--
---------------------------------------------------------------------------------
--* Procedure index
-------------------
--list-sorted? vector-sorted?
--
--list-merge vector-merge
--list-sort vector-sort
--list-stable-sort vector-stable-sort
--list-delete-neighbor-dups vector-delete-neighbor-dups
--
--list-merge! vector-merge!
--list-sort! vector-sort!
--list-stable-sort! vector-stable-sort!
--list-delete-neighbor-dups! vector-delete-neighbor-dups!
--
--quick-sort heap-sort insert-sort list-merge-sort vector-merge-sort
--quick-sort! heap-sort! insert-sort! list-merge-sort! vector-merge-sort!
--quick-sort3!
--
--vector-binary-search
--vector-binary-search3
--
---------------------------------------------------------------------------------
--* Introduction
----------------
--As I'll detail below, I wasn't very happy with the state of the Scheme
--world for sorting and merging lists and vectors. So I have designed and
--written a fairly comprehensive sorting & merging toolkit. It is
--
-- - very portable,
--
-- - much better code than what is currently in Elk, Gambit, Bigloo,
-- Scheme->C, MzScheme, RScheme, Scheme48, MIT Scheme, or slib, and
--
-- - priced to move: free code.
--
--The package includes
-- - Vector insert sort (stable)
-- - Vector heap sort
-- - Vector quick sort (with median-of-3 pivot picking)
-- - Vector merge sort (stable)
-- - Pure and destructive list merge sort (stable)
-- - Stable vector and list merge
-- - Miscellaneous sort-related procedures: Vector and list merging,
-- sorted? predicates, vector binary search, vector and list
-- delete-equal-neighbor procedures.
-- - A general, non-algorithmic set of procedure names for general sorting
-- and merging.
--
--Scheme programmers may want to adopt this package. I'd like Scheme
--implementors to adopt this code and its API -- in fact, the code is a bribe to
--make it easy for implementors to converge on the suggested API. I mean, you'd
--really have to be a boor to take this free code I wrote and mutate its
--interface over to your incompatible, unportable API, wouldn't you? But you
--could, of course -- it's freely available. More in the spirit of the offering,
--you could make this API available, and then also write a little module
--providing your old interface that is defined in terms of this API. "Scheme
--implementors," in this context, includes slib, which is not a standalone
--implementation of Scheme, but rather an influential collection of API's and
--code.
--
--The code is tightly bummed. It is clearly written, and commented in my usual
--voluminous style. This includes notes on porting and implementation-specific
--optimisations.
--
--
---------------------------------------------------------------------------------
--* What's wrong with the current state of affairs?
---------------------------------------------------
--
--It's just amazing to me that in 2002, sorting and merging hasn't been
--completely put to bed. These are well-understood algorithms, each of them well
--under a page of code. The straightforward algorithms are basic, core stuff --
--sophomore-level. But if you tour the major Scheme implementations out there on
--the Net, you find badly written code that provides extremely spotty coverage
--of the algorithm space. One implementation even has a buggy implementation
--that has been in use for about 20 years. Another has an O(n^2) algorithm...
--implemented in C for speed.
--
--Open source-code is a wonderful thing. In a couple of hours, I was able to
--download and check the sources of 9 Scheme systems. Here are my notes from the
--systems I checked. You can skip to the next section if you aren't morbidly
--curious.
--
--slib
-- sorted? vector-or-list <
-- merge list1 list2 <
-- merge! list1 list2 <
-- sort vector-or-list <
-- sort! vector-or-list <
--
-- Richard O'Keefe's stable list merge sort is right idea, but implemented
-- using gratuitous variable side effects. It also does redundant SET-CDR!s.
-- The vector sort converts to list, merge sorts, then reconverts
-- to vector. This is a bad idea -- non-local pointer chasing bad; vector
-- shuffling good. If you must allocate temp storage, might as well allocate
-- a temp vector and use vector merge sort.
--
--MIT Scheme
-- sort! vector <
-- merge-sort! vector <
-- quick-sort! vector <
--
-- sort vector-or-list <
-- merge-sort vector-or-list <
-- quick-sort vector-or-list <
--
-- Naive vector quicksort: loser, for worst-case performance reasons.
-- List sort by "list->vector; quicksort; vector->list," hence also loser.
-- A clever stable vector merge sort, albeit not very bummed.
--
--Scheme 48 & T
-- sort-list list <
-- sort-list! list <
-- list-merge! list1 list2 <
--
-- Bob Nix's implementation of online merge-sort, written in the early 80's.
-- Conses unnecessary bookkeeping structure, which isn't necessary with a
-- proper recursive formulation. Also, does redundant SET-CDR!s. No vector
-- sort. Also, has a bug -- is claimed to be a stable sort, but isn't! To see
-- this, get the S48 code, and try
-- (define (my< x y) (< (abs x) (abs y)))
-- (list-merge! (list 0 2) (list -2) my<) ; -> (0 2 -2)
-- (list-merge! (list 2) (list 0 -2) my<) ; -> (0 -2 2)
-- This could be fixed very easily, but it isn't worth it given the
-- other problems with the algorithm.
--
--RScheme
-- vector-sort! vector <
-- sort collection <
--
-- Good basic implementation of vector heapsort, which has O(n lg n)
-- worst-case time. Code ugly, needs tuning. List sort by "list->vector;
-- sort; vector->list." Nothing for stable sorting.
--
--MzScheme
-- quicksort lis <
-- mergesort alox <
--
-- Sorts lists with (list->vector; quicksort; vector->list) -- but the core
-- quicksort is not available for vector sorting. Nothing for stable sorting.
-- Quicksort picks pivot naively, inducing O(n^2) worse-case behaviour on a
-- fairly common case: an already-sorted list.
--
--Bigloo, STK
-- sort vector-or-list <
-- Uses an O(n^2) algorithm... implemented in C for speed. Hmm.
-- (See runtime/Ieee/vector.scm and runtime/Clib/cvector.c)
--
--Gambit
-- sort-list list <
-- Nothing for vectors. Simple, slow, unstable merge sort for lists.
--
--Elk
-- Another naive quicksort. Lists handled by converting to vector.
-- sort vector-or-list <
-- sort! vector-or-list <
--
--Chez Scheme
-- merge < list1 list2
-- merge! < list1 list2
-- sort < list
-- sort! < list
--
-- These are stable. I have not seen the source code.
--
--Common Lisp
-- sort sequence < [key]
-- stable-sort sequence < [key]
-- merge result-type sequence1 sequence2 < [key]
--
-- The sort procedures are allowed, but not required, to be destructive.
--
--SML/NJ
-- sort: ('a*'a -> bool) -> 'a list -> 'a list
-- "Smooth applicative merge sort," which is stable.
-- There is also a highly bummed quicksort for vectors.
--
--The right solution: Implement a full toolbox of carefully written standard sort
--routines.
--
--Having the source of all these above-cited Schemes available for study made
--life a lot easier writing this code. I appreciate the authors making their
--source available under such open terms.
--
--
---------------------------------------------------------------------------------
--* Design rules
----------------
--
--** What vs. how
--===============
--There are two different interfaces: "what" (simple) & "how" (detailed).
--
-- - Simple: you specify semantics: datatype (list or vector),
-- mutability, and stability.
--
-- - Detailed: you specify the actual algorithm (quick, heap,
-- insert, merge). Different algorithms have different properties,
-- both semantic & pragmatic, so these exports are necessary.
--
-- It is necessarily the case that the specifications of these procedures
-- make statements about execution "pragmatics." For example, the sole
-- distinction between heap sort and quick sort -- both of which are
-- provided by this library -- is one of execution time, which is not a
-- "semantic" distinction. Similar resource-use statements are made about
-- "iterative" procedures, meaning that they can execute on input of
-- arbitrary size in a constant number of stack frames.
--
--** Consistency across function signatures
--=========================================
--The two interfaces share common function signatures wherever
--possible, to facilitate switching a given call from one procedure
--to another.
--
--** Less-than parameter first, data parameter after
--==================================================
--These procedures uniformly observe the following parameter order:
--the data to be sorted comes after the comparison function.
--That is, we write
-- (sort < lis)
--not
-- (sort lis <).
--
--With the sole exception of Chez Scheme, this is the exact opposite of
--every sort function out there in current use in the Scheme world. (See
--the summary of related APIs above.) However, it is consistent with common
--practice across Scheme libraries in general to put the ordering function
--first -- the "operation currying" convention. (E.g., consider FOR-EACH or
--MAP or FIND.)
--
--The original draft of this SRFI used the data-first/comparison-last convention
--for backwards compatibility -- a decision I made with internal misgivings.
--Happily, however, the overwhelming response from the discussion phase
--supported "cleaning up" this issue and re-converging the parameter order with
--the general Scheme "op currying" convention. So the original decision was
--inverted in favor of the comparison-first/data-last convention.
--
--** Ordering, comparison functions & stability
--=============================================
--These routines take a < comparison function, not a <= comparison
--function, and they sort into increasing order. The difference between
--a < spec and a <= spec comes up in three places:
-- - the definition of an ordered or sorted data set,
-- - the definition of a stable sorting algorithm, and
-- - correctness of quicksort.
--
--+ We say that a data set (a list or vector) is *sorted* or *ordered*
-- if it contains no adjacent pair of values ... X Y ... such that Y < X.
--
-- In other words, scanning across the data never takes a "downwards" step.
--
-- If you use a <= procedure where these algorithms expect a <
-- procedure, you may not get the answers you expect. For example,
-- the LIST-SORTED? function will return false if you pass it a <= comparison
-- function and an ordered list containing adjacent equal elements.
--
--+ A "stable" sort is one that preserves the pre-existing order of equal
-- elements. Suppose, for example, that we sort a list of numbers by
-- comparing their absolute values, i.e., using comparison function
-- (lambda (x y) (< (abs x) (abs y)))
-- If we sort a list that contains both 3 and -3:
-- ... 3 ... -3 ...
-- then a stable sort is an algorithm that will not swap the order
-- of these two elements, that is, the answer is guaranteed to to look like
-- ... 3 -3 ...
-- not
-- ... -3 3 ...
--
-- Choosing < for the comparison function instead of <= affects how stability
-- is coded. Given an adjacent pair X Y, (< y x) means "Y should be moved in
-- front of X" -- otherwise, leave things as they are. So using a <= function
-- where a < function is expected will *invert* stability.
--
-- This is due to the definition of equality, given a < comparator:
-- (and (not (< x y))
-- (not (< y x)))
-- The definition is rather different, given a <= comparator:
-- (and (<= x y)
-- (<= y x))
--
--+ A "stable" merge is one that reliably favors one of its data sets
-- when equal items appear in both data sets. *All merge operations in
-- this library are stable*, breaking ties between data sets in favor
-- of the first data set -- elements of the first list come before equal
-- elements in the second list.
--
-- So, if we are merging two lists of numbers ordered by absolute value,
-- the stable merge operation LIST-MERGE
-- (list-merge (lambda (x y) (< (abs x) (abs y)))
-- '(0 -2 4 8 -10) '(-1 3 -4 7))
-- reliably places the 4 of the first list before the equal-comparing -4
-- of the second list:
-- (0 -1 -2 4 -4 7 8 -10)
--
--+ Some sort algorithms will *not work correctly* if given a <= when they
-- expect a < comparison (or vice-versa). For example, violating quicksort's
-- spec may cause it to produce wrong answers, diverge, raise an error, or do
-- some fourth thing. To see why, consider the left-scan part of the standard
-- quicksort partition step:
-- (let ((i (let scan ((i i)) (if (elt< (vector-ref v i) pivot)
-- (scan (+ i 1))
-- i))))
-- ...)
-- Consider applying this loop to a vector of all zeroes (hence, PIVOT, as
-- well, is zero), but erroneously using <= for the ELT< function. The loop
-- will scan right off the end of the vector, producing a vector-index error.
-- The guarantee that the scan loop will terminate before running off the end
-- of the vector depends critically upon ELT< performing as a true, irreflexive
-- < relation. Running off the end of the vector is only one of a variety of
-- possibly ways to lose -- other, variant implementations of quicksort can,
-- instead, loop forever on some data sets if ELT< is a <= predicate.
--
--In short, if your comparison function F answers true to (F x x), then
-- - using a stable sorting or merging algorithm will not give you a
-- stable sort or merge,
-- - LIST-SORTED? may surprise you, and
-- - quicksort may fail in a variety of possible ways.
--Note that you can synthesize a < function from a <= function with
-- (lambda (x y) (not (<= y x)))
--if need be.
--
--Precise definitions give sharp edges to tools, but require care in use.
--"Measure twice, cut once."
--
--I have adopted the choice of < from Common Lisp. One would assume the definers
--of Common Lisp had a good reason for adopting < instead of <=, but canvassing
--several of the principal actors in the definition process has turned up no
--better reason than "an arbitrary but consistent choice." At minimum, then,
--this SRFI extends the coverage of that consistent choice.
--
--** All vector operations accept optional subrange parameters
--============================================================
--The vector operations specified below all take optional START/END arguments
--indicating a selected subrange of a vector's elements. If a START parameter or
--START/END parameter pair is given to such a procedure, they must be exact,
--non-negative integers, such that
-- 0 <= START <= END <= (VECTOR-LENGTH V)
--where V is the related vector parameter. If not specified, they default to 0
--and the length of the vector, respectively. They are interpreted to select the
--range [START,END), that is, all elements from index START (inclusive) up to,
--but not including, index END.
--
--** Required vs. allowed side-effects
--====================================
--LIST-SORT! and LIST-STABLE-SORT! are allowed, but not required,
--to alter their arguments' cons cells to construct the result list. This is
--consistent with the what-not-how character of the group of procedures
--to which they belong (the "sort-lib" package).
--
--The LIST-DELETE-NEIGHBOR-DUPS!, LIST-MERGE! and LIST-MERGE-SORT! procedures,
--on the other hand, provide specific algorithms, and, as such, explicitly
--commit to the use of side-effects on their input lists in order to guarantee
--their key algorithmic properties (e.g., linear-time operation, constant-space
--stack use).
--
---------------------------------------------------------------------------------
--* Procedure specification
---------------------------
--The procedures are split into several packages. In a Scheme system that has a
--module or package system, these procedures should be contained in modules
--named as follows:
-- Package name Functionality
-- ------------ -------------
-- sort-lib General sorting for lists & vectors
-- sorted?-lib Sorted predicates for lists & vectors
-- list-merge-sort-lib List merge sort
-- vector-merge-sort-lib Vector merge sort
-- vector-heap-sort-lib Vector heap sort
-- vector-quick-sort-lib Vector quick sort
-- vector-insert-sort-lib Vector insertion sort
-- delndup-lib List and vector delete neighbor duplicates
-- binsearch-lib Vector binary search
--
--A Scheme system without a module system should provide all of the bindings
--defined in all of these modules as components of the "SRFI-32" package.
--
--Note that there is no "list insert sort" package, as you might as well always
--use list merge sort. The reference implementation's destructive list merge
--sort will do fewer SET-CDR!s than a destructive insert sort.
--
--** Procedure naming and functionality
--=====================================
--Almost all of the procedures described below are variants of two basic
--operations: sorting and merging. These procedures are consistently named
--by composing a set of basic lexemes to indicate what they do.
--
--Lexeme Meaning
-------- -------
--"sort" The procedure sorts its input data set by some < comparison function.
--
--"merge" The procedure merges two ordered data sets into a single ordered
-- result.
--
--"stable" This lexeme indicates that the sort is a stable one.
--
--"vector" The procedure operates upon vectors.
--
--"list" The procedure operates upon lists.
--
--"!" Procedures that end in "!" are allowed, and sometimes required,
-- to reuse their input storage to construct their answer.
--
--** Types of parameters and return values
--========================================
--In the procedures specified below,
-- - A LIS parameter is a list;
--
-- - A V parameter is a vector;
--
-- - A < or = parameter is a procedure accepting two arguments taken from the
-- specified procedure's data set(s), and returning a boolean;
--
-- - START and END parameters are exact, non-negative integers that
-- serve as vector indices selecting a subrange of some associated vector.
-- When specified, they must satisfy the relation
-- 0 <= start <= end <= (vector-length v)
-- where V is the associated vector.
--
--Passing values to procedures with these parameters that do not satisfy these
--types is an error.
--
--If a procedure is said to return "unspecified," this means that nothing at all
--is said about what the procedure returns, not even the number of return
--values. Such a procedure is not even required to be consistent from call to
--call in the nature or number of its return values. It is simply required to
--return a value (or values) that may be passed to a command continuation, e.g.
--as the value of an expression appearing as a non-terminal subform of a BEGIN
--expression. Note that in R5RS, this restricts such a procedure to returning a
--single value; non-R5RS systems may not even provide this restriction.
--
--** sort-lib - general sorting package
--=====================================
--This library provides basic sorting and merging functionality suitable for
--general programming. The procedures are named by their semantic properties,
--i.e., what they do to the data (sort, stable sort, merge, and so forth).
--
-- Procedure Suggested algorithm
-- -------------------------------------------------------------------------
-- list-sorted? < lis -> boolean
-- list-merge < lis1 lis2 -> list
-- list-merge! < lis1 lis2 -> list
-- list-sort < lis -> list (vector heap or quick)
-- list-sort! < lis -> list (list merge sort)
-- list-stable-sort < lis -> list (vector merge sort)
-- list-stable-sort! < lis -> list (list merge sort)
-- list-delete-neighbor-dups = lis -> list
-- list-delete-neighbor-dups! = lis -> list
--
-- vector-sorted? < v [start end] -> boolean
-- vector-merge < v1 v2 [start1 end1 start2 end2] -> vector
-- vector-merge! < v v1 v2 [start start1 end1 start2 end2] -> unspecified
-- vector-sort < v [start end] -> vector (heap or quick sort)
-- vector-sort! < v [start end] -> unspecified (heap or quick sort)
-- vector-stable-sort < v [start end] -> vector (vector merge sort)
-- vector-stable-sort! < v [start end] -> unspecified (vector merge sort)
-- vector-delete-neighbor-dups = v [start end] -> vector
-- vector-delete-neighbor-dups! = target source [t-start s-start s-end] -> t-end
--
-- LIST-SORTED? and VECTOR-SORTED? return true if their input list or vector
-- is in sorted order, as determined by their < comparison parameter.
--
-- All four merge operations are stable: an element of the initial list LIS1
-- or vector V1 will come before an equal-comparing element in the second
-- list LIS2 or vector V2 in the result.
--
-- The procedures
-- LIST-MERGE
-- LIST-SORT
-- LIST-STABLE-SORT
-- LIST-DELETE-NEIGHBOR-DUPS
-- do not alter their inputs and are allowed to return a value that shares
-- a common tail with a list argument.
--
-- The procedures
-- LIST-SORT!
-- LIST-STABLE-SORT!
-- are "linear update" operators -- they are allowed, but not required, to
-- alter the cons cells of their arguments to produce their results.
--
-- On the other hand, the procedures
-- LIST-DELETE-NEIGHBOR-DUPS!
-- LIST-MERGE!
-- make only a single, iterative, linear-time pass over their argument lists,
-- using SET-CDR!s to rearrange the cells of the lists into the final result
-- -- they work "in place." Hence, any cons cell appearing in the result must
-- have originally appeared in an input. The intent of this
-- iterative-algorithm commitment is to allow the programmer to be sure that
-- if, for example, LIST-MERGE! is asked to merge two ten-million-element
-- lists, the operation will complete without performing some extremely
-- (possibly twenty-million) deep recursion.
--
-- The vector procedures
-- VECTOR-SORT
-- VECTOR-STABLE-SORT
-- VECTOR-DELETE-NEIGHBOR-DUPS
-- do not alter their inputs, but allocate a fresh vector for their result,
-- of length END - START.
--
-- The vector procedures
-- VECTOR-SORT!
-- VECTOR-STABLE-SORT!
-- sort their data in-place. (But note that VECTOR-STABLE-SORT! may
-- allocate temporary storage proportional to the size of the input --
-- I am not aware of O(n lg n) stable vector-sorting algorithms that
-- run in constant space.)
--
-- VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
--
-- VECTOR-MERGE! writes its result into vector V, beginning at index START,
-- for indices less than END = START + (END1-START1) + (END2-START2). The
-- target subvector
-- V[start,end)
-- may not overlap either source subvector
-- V1[start1,end1)
-- V2[start2,end2).
--
-- The ...-DELETE-NEIGHBOR-DUPS-... procedures:
-- These procedures delete adjacent duplicate elements from a list or a
-- vector, using a given element-equality procedure. The first/leftmost
-- element of a run of equal elements is the one that survives. The list or
-- vector is not otherwise disordered.
--
-- These procedures are linear time -- much faster than the O(n^2) general
-- duplicate-element deletors that do not assume any "bunching" of elements
-- (such as the ones provided by SRFI-1). If you want to delete duplicate
-- elements from a large list or vector, you can sort the elements to bring
-- equal items together, then use one of these procedures, for a total time
-- of O(n lg n).
--
-- The comparison function = passed to these procedures is always applied
-- (= x y)
-- where X comes before Y in the containing list or vector.
--
-- - LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
-- may share storage with the input list.
--
-- - VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
-- rather allocates a fresh vector to hold the result.
--
-- - LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
-- mutate its input list in order to construct its answer.
--
-- - VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
-- answer, packing its answer into the index range [start,end'), where
-- END' is the non-negative exact integer returned as its value. It
-- returns END' as its result. The vector is not altered outside the range
-- [start,end').
--
-- - VECTOR-DELETE-NEIGHBOR-DUPS! scans vector SOURCE in range
-- [S-START,S-END), writing its result to vector TARGET beginning at index
-- T-START. It returns exact, non-negative integer T-END, which indicates
-- that the results of the operation are found in index range
-- [T-START,T-END) of TARGET; elements of TARGET outside this range
-- are unaltered.
--
-- It is an error for memory cell TARGET[T-START] to be a memory cell in
-- the region SOURCE[1 + S-START, S-END). In a Scheme implementation
-- that does not allow distinct vectors to share storage, this means
-- that one of the following must be true:
-- 1. (not (eq? source target))
-- 2. t-start not-in [s-start + 1, s-end)
--
-- - Examples:
-- (list-delete-neighbor-dups = '(1 1 2 7 7 7 0 -2 -2))
-- => (1 2 7 0 -2)
--
-- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2))
-- => #(1 2 7 0 -2)
--
-- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2) 3 7)
-- => #(7 0 -2)
--
-- ;; Result left in v[3,9):
-- (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
-- (cons (vector-delete-neighbor-dups! = v 3)
-- v))
-- => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
--
--
--** Algorithm-specific sorting packages
--======================================
--These packages provide more specific sorting functionality, that is,
--specific commitment to particular algorithms that have particular
--pragmatic consequences (such as memory locality, asymptotic running time)
--beyond their semantic behaviour (sorting, stable sorting, merging, etc.).
--Programmers that need a particular algorithm can use one of these packages.
--
--sorted?-lib - sorted predicates
-- list-sorted? < lis -> boolean
-- vector-sorted? < v [start end] -> boolean
--
-- Return #f iff there is an adjacent pair ... X Y ... in the input
-- list or vector such that Y < X. The optional START/END range
-- arguments restrict VECTOR-SORTED? to the indicated subvector.
--
--list-merge-sort-lib - list merge sort
-- list-merge-sort < lis -> list
-- list-merge-sort! < lis -> list
-- list-merge lis1 < lis2 -> list
-- list-merge! lis1 < lis2 -> list
--
-- The sort procedures sort their data using a list merge sort, which is
-- stable. (The reference implementation is, additionally, a "natural" sort.
-- See below for the properties of this algorithm.)
--
-- The ! procedures are destructive -- they use SET-CDR!s to rearrange the
-- cells of the lists into the proper order. As such, they do not allocate
-- any extra cons cells -- they are "in place" sorts. Additionally,
-- LIST-MERGE! is iterative -- it can operate on arguments of arbitrary size
-- with a constant number of stack frames.
--
-- The merge operations are stable: an element of LIS1 will come before an
-- equal-comparing element in LIS2 in the result list.
--
--vector-merge-sort-lib - vector merge sort
-- vector-merge-sort < v [start end temp] -> vector
-- vector-merge-sort! < v [start end temp] -> unspecified
-- vector-merge < v1 v2 [start1 end1 start2 end2] -> vector
-- vector-merge! < v v1 v2 [start start1 end1 start2 end2] -> unspecified
--
-- The sort procedures sort their data using vector merge sort, which is
-- stable. (The reference implementation is, additionally, a "natural" sort.
-- See below for the properties of this algorithm.)
--
-- The optional START/END arguments provide for sorting of subranges, and
-- default to 0 and the length of the corresponding vector.
--
-- Merge-sorting a vector requires the allocation of a temporary "scratch"
-- work vector for the duration of the sort. This scratch vector can be
-- passed in by the client as the optional TEMP argument; if so, the supplied
-- vector must be of size >= END, and will not be altered outside the range
-- [start,end). If not supplied, the sort routines allocate one themselves.
--
-- The merge operations are stable: an element of V1 will come before an
-- equal-comparing element in V2 in the result vector.
--
-- VECTOR-MERGE-SORT! leaves its result in V[start,end).
--
-- VECTOR-MERGE-SORT returns a vector of length END-START.
--
-- VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
--
-- VECTOR-MERGE! writes its result into vector V, beginning at index START,
-- for indices less than END = START + (END1-START1) + (END2-START2). The
-- target subvector
-- V[start,end)
-- may not overlap either source subvector
-- V1[start1,end1)
-- V2[start2,end2).
--
--vector-heap-sort-lib - vector heap sort
-- heap-sort < v [start end] -> vector
-- heap-sort! < v [start end] -> unspecified
--
-- These procedures sort their data using heap sort,
-- which is not a stable sorting algorithm.
--
-- HEAP-SORT returns a vector of length END-START.
-- HEAP-SORT! is in-place, leaving its result in V[start,end).
--
--vector-quick-sort-lib - vector quick sort
-- quick-sort < v [start end] -> vector
-- quick-sort! < v [start end] -> unspecified
-- quick-sort3! c v [start end] -> unspecified
--
-- These procedures sort their data using quick sort,
-- which is not a stable sorting algorithm.
--
-- QUICK-SORT returns a vector of length END-START.
-- QUICK-SORT! is in-place, leaving its result in V[start,end).
--
-- QUICK-SORT3! is a variant of quick-sort that takes a three-way
-- comparison function C. C compares a pair of elements and returns
-- an exact integer whose sign indicates their relationship:
-- (c x y) < 0 => x<y
-- (c x y) = 0 => x=y
-- (c x y) > 0 => x>y
-- To help remember the relationship between the sign of the result and
-- the relation, use the function - as the model for C: (- x y) < 0
-- means that x < y; (- x y) > 0 means that x > y.
--
-- The extra discrimination provided by the three-way comparison can
-- provide significant speedups when sorting data sets with many duplicates,
-- especially when the comparison function is relatively expensive (e.g.,
-- comparing long strings).
--
-- WARNING: Some sort algorithms, such as insertion sort or heap sort,
-- can tolerate being passed a <= comparison function when they expect a <
-- function -- insertion and merge sort may simply invert stability; and
-- heap sort will run a bit slower, but otherwise produce a correct answer.
--
-- Quicksort, however, is much more critically sensitive to the distinction
-- between a < and a <= comparison. If QUICK-SORT or QUICK-SORT! expect a <
-- comparison function, and are erroneously given a <= function, they may,
-- depending on implementation, produce an unsorted result, go into an
-- infinite loop, cause a run-time error, occasionally produce a correct
-- result, or do some fifth thing.
--
-- Implementors may wish to write QUICKSORT3! so that it (a) tests the
-- comparison function (by checking that (c v[start] v[start]) produces
-- false), or (b) is tolerant of an erroneous <= function, or (c) both.
-- Clients of this function, however, should not count on this.
--
--vector-insert-sort-lib - vector insertion sort
-- insert-sort < v [start end] -> vector
-- insert-sort! < v [start end] -> unspecified
--
-- These procedures stably sort their data using insertion sort.
--
-- INSERT-SORT returns a vector of length END-START.
-- INSERT-SORT! is in-place, leaving its result in V[start,end).
--
--delndup-lib - list and vector delete neighbor duplicates
-- list-delete-neighbor-dups = lis -> list
-- list-delete-neighbor-dups! = lis -> list
--
-- vector-delete-neighbor-dups = v [start end] -> vector
-- vector-delete-neighbor-dups! = v [start end] -> end'
--
-- These procedures delete adjacent duplicate elements from a list or
-- a vector, using a given element-equality procedure =. The first/leftmost
-- element of a run of equal elements is the one that survives. The list
-- or vector is not otherwise disordered.
--
-- These procedures are linear time -- much faster than the O(n^2) general
-- duplicate-element deletors that do not assume any "bunching" of elements
-- (such as the ones provided by SRFI-1). If you want to delete duplicate
-- elements from a large list or vector, you can sort the elements to bring
-- equal items together, then use one of these procedures, for a total time
-- of O(n lg n).
--
-- The comparison function = passed to these procedures is always applied
-- (= x y)
-- where X comes before Y in the containing list or vector.
--
-- LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
-- may share storage with the input list.
--
-- VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
-- rather allocates a fresh vector to hold the result.
--
-- LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
-- mutate its input list in order to construct its answer.
--
-- VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
-- answer, packing its answer into the index range [start,end'), where
-- END' is the non-negative exact integer returned as its value. It
-- returns END' as its result. The vector is not altered outside the range
-- [start,end').
--
-- Examples:
-- (list-delete-neighbor-dups = '(1 1 2 7 7 7 0 -2 -2))
-- => (1 2 7 0 -2)
--
-- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2))
-- => #(1 2 7 0 -2)
--
-- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2) 3 7)
-- => #(7 0 -2)
--
-- ;; Result left in v[3,9):
-- (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
-- (cons (vector-delete-neighbor-dups! = v 3)
-- v))
-- => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
--
--binsearch-lib - vector binary search lib
-- vector-binary-search elt< elt->key key v [start end] -> integer-or-false
-- vector-binary-search3 c v [start end] -> integer-or-false
--
-- VECTOR-BINARY-SEARCH searches vector V in range [START,END) (which
-- default to 0 and the length of V, respectively) for an element whose
-- associated key is equal to KEY. The procedure ELT->KEY is used to map
-- an element to its associated key. The elements of the vector are assumed
-- to be ordered by the ELT< relation on these keys. That is,
-- (vector-sorted? (lambda (x y) (elt< (elt->key x) (elt->key y)))
-- v start end) => true
-- An element E of V is a match for KEY if it's neither less nor greater
-- than the key:
-- (and (not (elt< (elt->key e) key))
-- (not (elt< key (elt->key e))))
-- If there is such an element, the procedure returns its index in the
-- vector as an exact integer. If there is no such element in the searched
-- range, the procedure returns false.
--
-- (vector-binary-search < car 4 '#((1 . one) (3 . three)
-- (4 . four) (25 . twenty-five)))
-- => 2
--
-- (vector-binary-search < car 7 '#((1 . one) (3 . three)
-- (4 . four) (25 . twenty-five)))
-- => #f
--
-- VECTOR-BINARY-SEARCH3 is a variant that uses a three-way comparison
-- function C. C compares its parameter to the search key, and returns an
-- exact integer whose sign indicates its relationship to the search key.
-- (c x) < 0 => x < search-key
-- (c x) = 0 => x = search-key
-- (c x) > 0 => x > search-key
--
-- (vector-binary-search3 (lambda (elt) (- (car elt) 4))
-- '#((1 . one) (3 . three)
-- (4 . four) (25 . twenty-five)))
-- => 2
--
-- Rationale:
-- - Why isn't VECTOR-BINARY-SEARCH's ELT->KEY computation simply absorbed
-- into the < function? It is separated out because the < function is
-- applied twice inside the binary-search inner loop, once with the search
-- key for the first argument and the element key for the second argument,
-- and once, with the reverse argument order. This is not necessary for
-- VECTOR-BINARY-SEARCH3.
--
-- - When a comparison operation is able to produce a three-way
-- discrimination, the inner loop of the binary search can trim the number
-- of per-iteration comparisons from an average of 1.5 to a guaranteed
-- single comparison per iteration. This can be a significant savings when
-- searching with an expensive comparison operation (e.g., one that
-- uses string compare, sends email, references a database, or queries
-- a network service such as a web server).
--
-- - Failure is signaled by false (rather than, say, -1) so that searches
-- can be used in conditional forms such as
-- (or (vector-binary-search ...) ...)
-- or
-- (cond ((vector-binary-search ...) => index-consumer)
-- ...)
--
---------------------------------------------------------------------------------
--* Algorithmic properties
--------------------------
--Different sort and merge algorithms have different properties.
--Choose the algorithm that matches your needs:
--
--Vector insert sort
-- Stable, but only suitable for small vectors -- O(n^2).
--
--Vector quick sort
-- Not stable. Is fast on average -- O(n lg n) -- but has bad worst-case
-- behaviour. Has good memory locality for big vectors (unlike heap sort).
-- A clever pivot-picking trick (median of three samples) helps avoid
-- worst-case behaviour, but pathological cases can still blow up.
--
--Vector heap sort
-- Not stable. Guaranteed fast -- O(n lg n) *worst* case. Poor locality
-- on large vectors. A very reliable workhorse.
--
--Vector merge sort
-- Stable. Not in-place -- requires a temporary buffer of equal size.
-- Fast -- O(n lg n) -- and has good memory locality for large vectors.
--
-- The implementation of vector merge sort provided by this SRFI's reference
-- implementation is, additionally, a "natural" sort, meaning that it
-- exploits existing order in the input data, providing O(n) best case.
--
--Destructive list merge sort
-- Stable, fast and in-place (i.e., allocates no new cons cells). "Fast"
-- means O(n lg n) worse-case, and substantially better if the data
-- is already mostly ordered, all the way down to linear time for
-- a completely-ordered input list (i.e., it is a "natural" sort).
--
-- Note that sorting lists involves chasing pointers through memory, which
-- can be a loser on modern machine architectures because of poor cache &
-- page locality. Pointer *writing*, which is what the SET-CDR!s of a
-- destructive list-sort algorithm do, is even worse, especially if your
-- Scheme has a generational GC -- the writes will thrash the write-barrier.
-- Sorting vectors has inherently better locality.
--
-- This SRFI's destructive list merge and merge sort implementations are
-- opportunistic -- they avoid redundant SET-CDR!s, and try to take long
-- already-ordered runs of list structure as-is when doing the merges.
--
--Pure list merge sort
-- Stable and fast -- O(n lg n) worst-case, and possibly O(n), depending
-- upon the input list (see discussion above).
--
--
--Algorithm Stable? Worst case Average case In-place
--------------------------------------------------------
--Vector insert Yes O(n^2) O(n^2) Yes
--Vector quick No O(n^2) O(n lg n) Yes
--Vector heap No O(n lg n) O(n lg n) Yes
--Vector merge Yes O(n lg n) O(n lg n) No
--List merge Yes O(n lg n) O(n lg n) Either
--
--
---------------------------------------------------------------------------------
--* Porting and optimisation
----------------------------
--This package should be trivial to port. There are only four non-R4RS bits
--in the code:
--- Use of multiple-value return, with the R5RS VALUES procedure, and the
-- simple (RECEIVE (var ...) mv-exp body ...) multiple-value binding macro
-- of SRFI-8.
--
--- A VECTOR-COPY procedure. This is a tiny little procedure:
-- (vector-copy v [start end])
--
--- Use of the LET-OPTIONALS macro from scsh to parse and default optional
-- arguments to three routines. Again, easy to port the macro or rewrite
-- the code to parse, default, and error check the args by hand.
--
--- Calls to an ERROR function for complaining about bad arguments.
--
--This code is tightly bummed, as far as I can go in portable Scheme.
--
--You could speed up the vector code a lot by error-checking the procedure
--parameters and then shifting over to fixnum-specific arithmetic and dangerous
--vector-indexing and vector-setting primitives. The comments in the code
--indicate where the initial error checks would have to be added. There are
--several (QUOTIENT N 2)'s that could be changed to a fixnum right-shift, as
--well, in both the list and vector code (SRFI 33 provides such an operator).
--The code is designed to enable this -- each file usually exports one or two
--"safe" procedures that end up calling an internal "dangerous" primitive. The
--little exported cover procedures are where you move the error checks.
--
--This should provide *big* speedups. In fact, all the code bumming I've done
--pretty much disappears in the noise unless you have a good compiler and also
--can dump the vector-index checks and generic arithmetic -- so I've really just
--set things up for you to exploit.
--
--The optional-arg parsing, defaulting, and error checking is done with a
--portable R4RS macro. But if your Scheme has a faster mechanism (e.g., Chez),
--you should definitely port over to it. Note that argument defaulting and
--error-checking are interleaved -- you don't have to error-check defaulted
--START/END args to see if they are fixnums that are legal vector indices for
--the corresponding vector, etc.
--
--
---------------------------------------------------------------------------------
--* References & Links
----------------------
--
--This document, in HTML:
-- http://srfi.schemers.org/srfi-32/srfi-32.html
-- [This link may not be valid while the SRFI is in draft form.]
--
--This document, in simple text format:
-- http://srfi.schemers.org/srfi-32/srfi-32.txt
--
--Archive of SRFI-32 discussion-list email:
-- http://srfi.schemers.org/srfi-32/mail-archive/maillist.html
--
--SRFI web site:
-- http://srfi.schemers.org/
--
--[CommonLisp]
-- Common Lisp: the Language
-- Guy L. Steele Jr. (editor).
-- Digital Press, Maynard, Mass., second edition 1990.
-- Available at http://www.elwood.com/alu/table/references.htm#cltl2
--
-- The Common Lisp "HyperSpec," produced by Kent Pitman, is essentially
-- the ANSI spec for Common Lisp:
-- http://www.xanalys.com/software_tools/reference/HyperSpec/
--
--[R5RS]
-- Revised^5 Report on the Algorithmic Language Scheme,
-- R. Kelsey, W. Clinger, J. Rees (editors).
-- Higher-Order and Symbolic Computation, Vol. 11, No. 1, September, 1998.
-- and ACM SIGPLAN Notices, Vol. 33, No. 9, October, 1998.
--
-- Available at http://www.schemers.org/Documents/Standards/
--
--
---------------------------------------------------------------------------------
--* Acknowledgements
--------------------
--
--I thank the authors of the open source I consulted when designing this
--library, particularly Richard O'Keefe, Donovan Kolby and the MIT Scheme Team.
--
--
---------------------------------------------------------------------------------
--* Copyright
-------------
--
--** SRFI text
--============
--This document is copyright (C) Olin Shivers (1998, 1999).
--All Rights Reserved.
--
--This document and translations of it may be copied and furnished to others,
--and derivative works that comment on or otherwise explain it or assist in its
--implementation may be prepared, copied, published and distributed, in whole or
--in part, without restriction of any kind, provided that the above copyright
--notice and this paragraph are included on all such copies and derivative
--works. However, this document itself may not be modified in any way, such as
--by removing the copyright notice or references to the Scheme Request For
--Implementation process or editors, except as needed for the purpose of
--developing SRFIs in which case the procedures for copyrights defined in the
--SRFI process must be followed, or as required to translate it into languages
--other than English.
--
--The limited permissions granted above are perpetual and will not be revoked by
--the authors or their successors or assigns.
--
--This document and the information contained herein is provided on an "AS IS"
--basis and THE AUTHORS AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR
--IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
--INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
--MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
--
--** Reference implementation
--===========================
--Short summary: no restrictions.
--
--While I wrote all of this code myself, I read a lot of code before I began
--writing. However, all such code is, itself, either open source or public
--domain, rendering irrelevant any issue of "copyright taint."
--
--The natural merge sorts (pure list, destructive list, and vector) are not only
--my own code, but are implementations of an algorithm of my own devising. They
--run in O(n lg n) worst case, O(n) best case, and require only a logarithmic
--number of stack frames. And they are stable. And the destructive-list variant
--allocates zero cons cells; it simply rearranges the cells of the input list.
--
--Hence the reference implementation is
-- Copyright (c) 1998 by Olin Shivers.
--and made available under the same copyright as the SRFI text (see above).
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/srfi-32.txt racket-6.12/share/pkgs/srfi-lib/srfi/32/srfi-32.txt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/srfi-32.txt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/32/srfi-32.txt 2018-01-31 12:22:58.152501970 -0500
-@@ -0,0 +1,938 @@
-+The SRFI-32 sort libraries -*- outline -*-
-+Olin Shivers
-+First draft: 1998/10/19
-+Last update: 2002/7/21
-+
-+Emacs should display this document in outline mode. Say c-h m for
-+instructions on how to move through it by sections (e.g., c-c c-n, c-c c-p).
-+
-+* Table of contents
-+-------------------
-+Abstract
-+Procedure index
-+Introduction
-+What's wrong with the current state of affairs?
-+Design rules
-+ What vs. how
-+ Consistency across function signatures
-+ Data parameter first, less-than parameter after
-+ Ordering, comparison functions & stability
-+ All vector operations accept optional subrange parameters
-+ Required vs. allowed side-effects
-+Procedure specification
-+ Procedure naming and functionality
-+ Types of parameters and return values
-+ sort-lib - general sorting package
-+ Algorithm-specific sorting packages
-+Algorithmic properties
-+Topics to be resolved during discussion phase
-+Porting and optimisation
-+References & Links
-+Acknowledgements
-+Copyright
-+
-+
-+* Abstract
-+----------
-+Current Scheme sorting packages are, every one of them, surprisingly bad. I've
-+designed the API for a full-featured sort toolkit, which I propose as an SRFI.
-+The spec comes with 1200 lines of high-quality reference code: tightly
-+written, highly commented, portable code, available for free. Implementors
-+want this code. It's better than what you have.
-+
-+-------------------------------------------------------------------------------
-+* Procedure index
-+-----------------
-+list-sorted? vector-sorted?
-+
-+list-merge vector-merge
-+list-sort vector-sort
-+list-stable-sort vector-stable-sort
-+list-delete-neighbor-dups vector-delete-neighbor-dups
-+
-+list-merge! vector-merge!
-+list-sort! vector-sort!
-+list-stable-sort! vector-stable-sort!
-+list-delete-neighbor-dups! vector-delete-neighbor-dups!
-+
-+heap-sort quick-sort insert-sort list-merge-sort vector-merge-sort
-+heap-sort! quick-sort! insert-sort! list-merge-sort! vector-merge-sort!
-+
-+-------------------------------------------------------------------------------
-+* Introduction
-+--------------
-+As I'll detail bewlow, I wasn't very happy with the state of the Scheme
-+world for sorting and merging lists and vectors. So I have designed and
-+written a fairly comprehensive sorting & merging toolkit. It is
-+
-+ - very portable,
-+
-+ - much better code than what is currently in Elk, Gambit, Bigloo,
-+ Scheme->C, MzScheme, RScheme, Scheme48, MIT Scheme, or slib, and
-+
-+ - priced to move: free code.
-+
-+The package includes
-+ - Vector insert sort (stable)
-+ - Vector heap sort
-+ - Vector quick sort (with median-of-3 pivot picking)
-+ - Vector merge sort (stable)
-+ - Pure and destructive list merge sort (stable)
-+ - Stable vector and list merge
-+ - Miscellaneous sort-related procedures: Vector and list merging,
-+ sorted? predicates, vector binary search, vector and list
-+ delete-equal-neighbor procedures.
-+ - A general, non-algorithmic set of procedure names for general sorting
-+ and merging.
-+
-+Scheme programmers may want to adopt this package. I'd like Scheme
-+implementors to adopt this code and its API -- in fact, the code is a bribe to
-+make it easy for implementors to converge on the suggested API. I mean, you'd
-+really have to be a boor to take this free code I wrote and mutate its
-+interface over to your incompatible, unportable API, wouldn't you? But you
-+could, of course -- it's freely available. More in the spirit of the offering,
-+you could make this API available, and then also write a little module
-+providing your old interface that is defined in terms of this API. "Scheme
-+implementors," in this context, includes slib, which isn't really a standalone
-+implementation of Scheme, but is an influential collection of API's and code.
-+
-+The code is tightly bummed. It is clearly written, and commented in my usual
-+voluminous style. This includes notes on porting and implementation-specific
-+optimisations.
-+
-+
-+-------------------------------------------------------------------------------
-+* What's wrong with the current state of affairs?
-+-------------------------------------------------
-+
-+It's just amazing to me that in 2002, sorting and merging hasn't been
-+completely put to bed. These are well-understood algorithms, each of them well
-+under a page of code. The straightforward algorithms are basic, core stuff --
-+sophomore-level. But if you tour the major Scheme implementations out there on
-+the Net, you find badly written code that provides extremely spotty coverage
-+of the algorithm space. One implementation even has a buggy implementation
-+that has been in use for about 20 years!
-+
-+Open source-code is a wonderful thing. In a couple of hours, I was able to
-+download and check the sources of 9 Scheme systems. Here are my notes from the
-+systems I checked. You can skip to the next section if you aren't morbidly
-+curious.
-+
-+slib
-+ sorted? vector-or-list <
-+ merge list1 list2 <
-+ merge! list1 list2 <
-+ sort vector-or-list <
-+ sort! vector-or-list <
-+
-+ Richard O'Keefe's stable list merge sort is right idea, but implemented
-+ using gratuitous variable side effects. It also does redundant SET-CDR!s.
-+ The vector sort converts to list, merge sorts, then reconverts
-+ to vector. This is a bad idea -- non-local pointer chasing bad; vector
-+ shuffling good.
-+
-+MIT Scheme
-+ sort! vector <
-+ merge-sort! vector <
-+ quick-sort! vector <
-+
-+ sort vector-or-list <
-+ merge-sort vector-or-list <
-+ quick-sort vector-or-list <
-+
-+ Naive vector quicksort: loser, for worst-case performance reasons.
-+ List sort by "list->vector; quicksort; vector->list," hence also loser.
-+ A clever stable vector merge sort, albeit not very bummed.
-+
-+Scheme 48 & T
-+ sort-list list <
-+ sort-list! list <
-+ list-merge! list1 list2 <
-+
-+ Bob Nix's implementation of online merge-sort, written in the early 80's.
-+ Conses unnecessary bookkeeping structure, which isn't necessary with a
-+ proper recursive formulation. Also, does redundant SET-CDR!s. No vector
-+ sort. Also, has a bug -- is claimed to be a stable sort, but isn't! To see
-+ this, get the S48 code, and try
-+ (define (my< x y) (< (quotient x 2) (quotient y 2)))
-+ (list-merge! (list 0 2) (list 3) my<) ; -> (0 2 3)
-+ (list-merge! (list 2) (list 0 3) my<) ; -> (0 3 2)
-+ This could be fixed very easily, but it isn't worth it given the
-+ other problems with the algorithm.
-+
-+RScheme
-+ vector-sort! vector <
-+ sort collection <
-+
-+ Good basic implementation of vector heapsort, which has O(n lg n)
-+ worst-case time. Code ugly, needs tuning. List sort by "list->vector;
-+ sort; vector->list", which allocates unneeded temp storage. Nothing
-+ for stable sorting.
-+
-+MzScheme
-+ Naive quicksort -- but not available for vector sorting, even
-+ though it internally uses a vector. Nothing for stable sorting,
-+ and naive quicksort has bad worst-case behaviour.
-+
-+Bigloo, Scheme->C
-+ Couldn't find anything -- but maybe I didn't search for the right
-+ thing, since the Bigloo names are French. (I invite correction from
-+ the Bigloo implementors.)
-+
-+Gambit
-+ sort-list list <
-+ Nothing for vectors. Simple, slow, unstable merge sort for lists.
-+
-+Elk
-+ Another naive quicksort. Lists handled by converting to vector.
-+ sort vector-or-list <
-+ sort! vector-or-list <
-+
-+Chez Scheme
-+ merge < list1 list2
-+ merge! < list1 list2
-+ sort < list
-+ sort! < list
-+
-+ These are stable. I have not seen the source code.
-+
-+Common Lisp
-+ sort sequence < [key]
-+ stable-sort sequence < [key]
-+ merge result-type sequence1 sequence2 < [key]
-+
-+ The sort procedures are allowed, but not required, to be destructive.
-+
-+SML/NJ
-+ sort: ('a*'a -> bool) -> 'a list -> 'a list
-+ "Smooth applicative merge sort," which is stable.
-+ There is also a highly bummed quicksort for vectors.
-+
-+The right solution: Implement a full toolbox of carefully written standard sort
-+routines.
-+
-+Having the source available for all of these above-cited Schemes made
-+life a lot easier writing this code. I appreciate the authors making their
-+source available under such open terms.
-+
-+
-+-------------------------------------------------------------------------------
-+* Design rules
-+--------------
-+
-+** What vs. how
-+===============
-+There are two different interfaces: "what" (simple) & "how" (detailed).
-+
-+ - Simple: you specify semantics: datatype (list or vector),
-+ mutability, and stability.
-+
-+ - Detailed: you specify the actual algorithm (quick, heap,
-+ insert, merge). Different algorithms have different properties,
-+ both semantic & pragmatic, so these exports are necessary.
-+
-+ It is necessarily the case that the specifications of these procedures
-+ make statements about execution "pragmatics." For example, the sole
-+ distinction between heap sort and quick sort -- both of which are
-+ provided by this library -- is one of execution time, which is not a
-+ "semantic" distinction. Similar resource-use statements are made about
-+ "iterative" procedures, meaning that they can execute on input of
-+ arbitrary size without needing to allocate an unbounded number of stack
-+ frames.
-+
-+** Consistency across function signatures
-+=========================================
-+The two interfaces share common function signatures wherever
-+possible, to facilitate switching a given call from one procedure
-+to another.
-+
-+** Data parameter first, less-than parameter after
-+==================================================
-+These procedures uniformly observe the following parameter order:
-+the data to be sorted come before the the comparison function.
-+That is, we write
-+ (sort lis <)
-+not
-+ (sort < lis).
-+This is consistent with every single implementation out there, with
-+the sole exception of Chez Scheme.
-+
-+In my opinion, it would be more consistent with other Scheme libraries
-+to put the ordering function first -- the "operation currying" convention.
-+(E.g., consider FOR-EACH or MAP or FIND.) I decided to leave things as they
-+are in favor of near-total backwards compatibility with existing practice.
-+
-+[Perhaps this should be discussed.]
-+
-+** Ordering, comparison functions & stability
-+=============================================
-+These routines take a < comparison function, not a <= comparison
-+function, and they sort into increasing order. The difference between
-+a < spec and a <= spec comes up in two places:
-+ - the definition of an ordered or sorted data set, and
-+ - the definition of a stable sorting algorithm.
-+
-++ We say that a data set (a list or vector) is *sorted* or *ordered*
-+ if it contains no adjacent pair of values ... X Y ... such that Y < X.
-+
-+ In other words, scanning across the data never takes a "downwards" step.
-+
-+ If you use a <= procedure where these algorithms expect a <
-+ procedure, you may not get the answers you expect. For example,
-+ the LIST-SORTED? function will return false if you pass it a <= comparison
-+ function and an ordered list containing adjacent equal elements.
-+
-++ A "stable" sort is one that preserves the pre-existing order of equal
-+ elements. Suppose, for example, that we sort a list of numbers by
-+ comparing their absolute values, i.e., using comparison function
-+ (lambda (x y) (< (abs x) (abs y)))
-+ If we sort a list that contains both 3 and -3:
-+ ... 3 ... -3 ...
-+ then a stable sort is an algorithm that will not swap the order
-+ of these two elements, that is, the answer will look like
-+ ... 3 -3 ...
-+ not
-+ ... -3 3 ...
-+
-+ Choosing < for the comparison function instead of <= affects how stability
-+ is coded. Given an adjacent pair X Y, (< y x) means "Y should be moved in
-+ front of X" -- otherwise, leave things as they are. So using a <= function
-+ where a < function is expected will *invert* stability.
-+
-+ This is due to the definition of equality, given a < comparator:
-+ (and (not (< x y))
-+ (not (< y x)))
-+ The definition is rather different, given a <= comparator:
-+ (and (<= x y)
-+ (<= x y))
-+
-++ A "stable" merge is one that reliably favors one of its data sets
-+ when equal items appear in both data sets. *All merge operations in
-+ this library are stable*, breaking ties between data sets in favor
-+ of the first data set -- elements of the first list come before equal
-+ elements in the second list.
-+
-+ So, if we are merging two lists of numbers ordered by absolute value
-+ using the stable merge operation LIST-MERGE
-+ (list-merge '(0 -2 4 8 -10) '(-1 3 -4 7)
-+ (lambda (x y) (< (abs x) (abs y))))
-+ reliably places the 4 of the first list before the equal-comparing -4
-+ of the second list:
-+ (0 -1 -2 4 -4 7 8 -10)
-+
-+In short, if your comparison function F answers true to (F x x), then
-+using a stable sorting or merging algorithm will not give you a stable sort
-+or merge, and LIST-SORTED? may surprise you. Note that you can synthesize a <
-+function from a <= function with
-+ (lambda (x y) (not (<= y x)))
-+if need be.
-+
-+Precise definitions give sharp edges to tools, but require care
-+in use. "Measure twice, cut once."
-+
-+I have adopted the choice of < from Common Lisp. I assume they
-+had a good reason for adopting < instead of <=. I'd love to know
-+what this reason is; send me email if you can explain it, please.
-+
-+** All vector operations accept optional subrange parameters
-+============================================================
-+The vector operations specified below all take optional START/END arguments
-+indicating a selected subrange of a vector's elements. If a START parameter or
-+START/END parameter pair is given to such a procedure, they must be exact,
-+non-negative integers, such that
-+ 0 <= START <= END <= (VECTOR-LENGTH V)
-+where V is the related vector parameter. If not specified, they default to 0
-+and the length of the vector, respectively. They are interpreted to select the
-+range [START,END), that is, all elements from index START (inclusive) up to,
-+but not including, index END.
-+
-+** Required vs. allowed side-effects
-+====================================
-+LIST-SORT! and LIST-STABLE-SORT! are allowed, but not required,
-+to alter their arguments' cons cells to construct the result list. This is
-+consistent with the what-not-how character of the group of procedures
-+to which they belong (the "sort-lib" package).
-+
-+The LIST-DELETE-NEIGHBOR-DUPS!, LIST-MERGE! and LIST-MERGE-SORT! procedures,
-+on the other hand, provide specific algorithms, and, as such, explicitly
-+commit to the use of side-effects on their input lists in order to guarantee
-+their key algorithmic properties (e.g., linear-time operation, constant-space
-+stack use).
-+
-+-------------------------------------------------------------------------------
-+* Procedure specification
-+-------------------------
-+The procedures are split into several packages. In a Scheme system that has a
-+module or package system, these procedures should be contained in modules
-+named as follows:
-+ Package name Functionality
-+ ------------ -------------
-+ sort-lib General sorting for lists & vectors
-+ sorted?-lib Sorted predicates for lists & vectors
-+ list-merge-sort-lib List merge sort
-+ vector-merge-sort-lib Vector merge sort
-+ vector-heap-sort-lib Vector heap sort
-+ vector-quick-sort-lib Vector quick sort
-+ vector-insert-sort-lib Vector insertion sort
-+ delndup-lib List and vector delete neighbor duplicates
-+
-+A Scheme system without a module system should provide all of the bindings
-+defined in all of these modules as components of the "SRFI-32" package.
-+
-+Note that there is no list insert sort package, as you might as well always
-+use list merge sort. The reference implementation's destructive list merge
-+sort will do fewer SET-CDR!s than a destructive insert sort.
-+
-+** Procedure naming and functionality
-+=====================================
-+Almost all of the procedures described below are variants of two basic
-+operations: sorting and merging. These procedures are consistently named
-+by composing a set of basic lexemes to indicate what they do.
-+
-+Lexeme Meaning
-+------ -------
-+"sort" The procedure sorts its input data set by some < comparison function.
-+
-+"merge" The procedure merges two ordered data sets into a single ordered
-+ result.
-+
-+"stable" This lexeme indicates that the sort is a stable one.
-+
-+"vector" The procedure operates upon vectors.
-+
-+"list" The procedure operates upon lists.
-+
-+"!" Procedures that end in "!" are allowed, and sometimes required,
-+ to reuse their input storage to construct their answer.
-+
-+** Types of parameters and return values
-+========================================
-+In the procedures specified below,
-+ - A LIS parameter is a list;
-+
-+ - A V parameter is a vector;
-+
-+ - A < or = parameter is a procedure accepting two arguments taken from the
-+ specified procedure's data set(s), and returning a boolean;
-+
-+ - START and END parameters are exact, non-negative integers that
-+ serve as vector indices selecting a subrange of some associated vector.
-+ When specified, they must satisfy the relation
-+ 0 <= start <= end <= (vector-length v)
-+ where V is the associated vector.
-+
-+Passing values to procedures with these parameters that do not satisfy these
-+types is an error.
-+
-+If a procedure is said to return "unspecified," this means that nothing at all
-+is said about what the procedure returns, not even the number of return
-+values. Such a procedure is not even required to be consistent from call to
-+call in the nature or number of its return values. It is simply required to
-+return a value (or values) that may be passed to a command continuation, e.g.
-+as the value of an expression appearing as a non-terminal subform of a BEGIN
-+expression. Note that in R5RS, this restricts such a procedure to returning a
-+single value; non-R5RS systems may not even provide this restriction.
-+
-+** sort-lib - general sorting package
-+=====================================
-+This library provides basic sorting and merging functionality suitable for
-+general programming. The procedures are named by their semantic properties,
-+i.e., what they do to the data (sort, stable sort, merge, and so forth).
-+
-+ Procedure Suggested algorithm
-+ -------------------------------------------------------------------------
-+ list-sorted? lis < -> boolean
-+ list-merge lis1 lis2 < -> list
-+ list-merge! lis1 lis2 < -> list
-+ list-sort lis < -> list (vector heap or quick)
-+ list-sort! lis < -> list (list merge sort)
-+ list-stable-sort lis < -> list (vector merge sort)
-+ list-stable-sort! lis < -> list (list merge sort)
-+ list-delete-neighbor-dups lis = -> list
-+ list-delete-neighbor-dups! lis = -> list
-+
-+ vector-sorted? v < [start end] -> boolean
-+ vector-merge v1 v2 < [start1 end1 start2 end2] -> vector
-+ vector-merge! v v1 v2 < [start start1 end1 start2 end2] -> unspecific
-+ vector-sort v < [start end] -> vector (heap or quick sort)
-+ vector-sort! v < [start end] -> unspecific (heap or quick sort)
-+ vector-stable-sort v < [start end] -> vector (vector merge sort)
-+ vector-stable-sort! v < [start end] -> unspecific (vector merge sort)
-+ vector-delete-neighbor-dups v = [start end] -> vector
-+ vector-delete-neighbor-dups! v = [start end] -> end'
-+
-+ LIST-SORTED? and VECTOR-SORTED? return true if their input list or vector
-+ is in sorted order, as determined by their < comparison parameter.
-+
-+ All four merge operations are stable: an element of the initial list LIS1
-+ or vector V1 will come before an equal-comparing element in the second
-+ list LIS2 or vector V2 in the result.
-+
-+ The procedures
-+ LIST-MERGE
-+ LIST-SORT
-+ LIST-STABLE-SORT
-+ LIST-DELETE-NEIGHBOR-DUPS
-+ do not alter their inputs and are allowed to return a value that shares
-+ a common tail with a list argument.
-+
-+ The procedures
-+ LIST-SORT!
-+ LIST-STABLE-SORT!
-+ are "linear update" operators -- they are allowed, but not required, to
-+ alter the cons cells of their arguments to produce their results.
-+
-+ On the other hand, the procedures
-+ LIST-DELETE-NEIGHBOR-DUPS!
-+ LIST-MERGE!
-+ make only a single, iterative, linear-time pass over their argument lists,
-+ using SET-CDR!s to rearrange the cells of the lists into the final result
-+ -- they work "in place." Hence, any cons cell appearing in the result must
-+ have originally appeared in an input. The intent of this
-+ iterative-algorithm commitment is to allow the programmer to be sure that
-+ if, for example, LIST-MERGE! is asked to merge two ten-million-element
-+ lists, the operation will complete without performing some extremely
-+ (possibly twenty-million) deep recursion.
-+
-+ The vector procedures
-+ VECTOR-SORT
-+ VECTOR-STABLE-SORT
-+ VECTOR-DELETE-NEIGHBOR-DUPS
-+ do not alter their inputs, but allocate a fresh vector for their result,
-+ of length END-START.
-+
-+ The vector procedures
-+ VECTOR-SORT!
-+ VECTOR-STABLE-SORT!
-+ sort their data in-place. (But note that VECTOR-STABLE-SORT! may
-+ allocate temporary storage proportional to the size of the input --
-+ I am not aware of O(n lg n) stable vector sorting algorithms that
-+ run in constant space.)
-+
-+ VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
-+
-+ VECTOR-MERGE! writes its result into vector V, beginning at index START0,
-+ for indices less than END0 = START0 + (END1-START1) + (END2-START2). The
-+ target subvector
-+ V[start0,end0)
-+ may not overlap either source subvector
-+ V1[start1,end1)
-+ V2[start2,end2).
-+
-+ The DELETE-NEIGHBOR-DUP-... procedures:
-+ These procedures delete adjacent duplicate elements from a list or a
-+ vector, using a given element-equality procedure. The first/leftmost
-+ element of a run of equal elements is the one that survives. The list or
-+ vector is not otherwise disordered.
-+
-+ These procedures are linear time -- much faster than the O(n^2) general
-+ duplicate-element deletors that do not assume any "bunching" of elements
-+ (such as the ones provided by SRFI-1). If you want to delete duplicate
-+ elements from a large list or vector, sort the elements to bring equal
-+ items together, then use one of these procedures, for a total time of
-+ O(n lg n).
-+
-+ The comparison function = passed to these procedures is always applied
-+ (= x y)
-+ where X comes before Y in the containing list or vector.
-+
-+ - LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
-+ may share storage with the input list.
-+
-+ - VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
-+ rather allocates a fresh vector to hold the result.
-+
-+ - LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
-+ mutate its input list in order to construct its answer.
-+
-+ - VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
-+ answer, packing its answer into the index range [start,end'), where
-+ END' is the non-negative exact integer returned as its value. It
-+ returns END' as its result. The vector is not altered outside the range
-+ [start,end').
-+
-+ [Maybe this procedure should take a "target" vector to write?]
-+
-+ - Examples:
-+ (list-delete-neighbor-dups '(1 1 2 7 7 7 0 -2 -2) =)
-+ => (1 2 7 0 -2)
-+
-+ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) =)
-+ => #(1 2 7 0 -2)
-+
-+ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) = 3 7)
-+ => #(7 0 -2)
-+
-+ ;; Result left in v[3,9):
-+ (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
-+ (cons (vector-delete-neighbor-dups! v = 3)
-+ v))
-+ => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
-+
-+
-+** Algorithm-specific sorting packages
-+======================================
-+These packages provide more specific sorting functionality, that is,
-+specific committment to particular algorithms that have particular
-+pragmatic consequences (such as memory locality, asymptotic running time)
-+beyond their semantic behaviour (sorting, stable sorting, merging, etc.).
-+Programmers that need a particular algorithm can use one of these packages.
-+
-+sorted?-lib - sorted predicates
-+ list-sorted? lis < -> boolean
-+ vector-sorted? v < [start end] -> boolean
-+
-+ Return #f iff there is an adjacent pair ... X Y ... in the input
-+ list or vector such that Y < X. The optional START/END range
-+ arguments restrict VECTOR-SORTED? to the indicated subvector.
-+
-+list-merge-sort-lib - list merge sort
-+ list-merge-sort lis < -> list
-+ list-merge-sort! lis < -> list
-+ list-merge lis1 lis2 < -> list
-+ list-merge! lis1 lis2 < -> list
-+
-+ The sort procedures sort their data using a list merge sort, which is
-+ stable. (The reference implementation is, additionally, a "natural" sort.
-+ See below for the properties of this algorithm.)
-+
-+ The ! procedures are destructive -- they use SET-CDR!s to rearrange the
-+ cells of the lists into the proper order. As such, they do not allocate
-+ any extra cons cells -- they are "in place" sorts. Additionally,
-+ LIST-MERGE! is iterative, not recursive -- it can operate on arguments of
-+ arbitrary size without requiring an unbounded amount of stack space.
-+
-+ The merge operations are stable: an element of LIS1 will come before an
-+ equal-comparing element in LIS2 in the result list.
-+
-+vector-merge-sort-lib - vector merge sort
-+ vector-merge-sort v < [start end temp] -> vector
-+ vector-merge-sort! v < [start end temp] -> unspecific
-+ vector-merge v1 v2 < [start1 end1 start2 end2] -> vector
-+ vector-merge! v v1 v2 < [start0 start1 end1 start2 end2] -> unspecific
-+
-+ The sort procedures sort their data using vector merge sort, which is
-+ stable. (The reference implementation is, additionally, a "natural" sort.
-+ See below for the properties of this algorithm.)
-+
-+ The optional START/END arguments provide for sorting of subranges, and
-+ default to 0 and the length of the corresponding vector.
-+
-+ Merge-sorting a vector requires the allocation of a temporary "scratch"
-+ work vector for the duration of the sort. This scratch vector can be
-+ passed in by the client as the optional TEMP argument; if so, the supplied
-+ vector must be of size >= END, and will not be altered outside the range
-+ [start,end). If not supplied, the sort routines allocate one themselves.
-+
-+ The merge operations are stable: an element of V1 will come before an
-+ equal-comparing element in V2 in the result vector.
-+
-+ VECTOR-MERGE-SORT! leaves its result in V[start,end).
-+
-+ VECTOR-MERGE-SORT returns a vector of length END-START.
-+
-+ VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
-+
-+ VECTOR-MERGE! writes its result into vector V, beginning at index START0,
-+ for indices less than END0 = START0 + (END1-START1) + (END2-START2). The
-+ target subvector
-+ V[start0,end0)
-+ may not overlap either source subvector
-+ V1[start1,end1)
-+ V2[start2,end2).
-+
-+vector-heap-sort-lib - vector heap sort
-+ heap-sort v < [start end] -> vector
-+ heap-sort! v < [start end] -> unspecific
-+
-+ These procedures sort their data using heap sort,
-+ which is not a stable sorting algorithm.
-+
-+ HEAP-SORT returns a vector of length END-START.
-+ HEAP-SORT! is in-place, leaving its result in V[start,end).
-+
-+vector-quick-sort-lib - vector quick sort
-+ quick-sort v < [start end] -> vector
-+ quick-sort! v < [start end] -> unspecific
-+
-+ These procedures sort their data using quick sort,
-+ which is not a stable sorting algorithm.
-+
-+ QUICK-SORT returns a vector of length END-START.
-+ QUICK-SORT! is in-place, leaving its result in V[start,end).
-+
-+vector-insert-sort-lib - vector insertion sort
-+ insert-sort v < [start end] -> vector
-+ insert-sort! v < [start end] -> unspecific
-+
-+ These procedures stably sort their data using insertion sort.
-+
-+ INSERT-SORT returns a vector of length END-START.
-+ INSERT-SORT! is in-place, leaving its result in V[start,end).
-+
-+delndup-lib - list and vector delete neighbor duplicates
-+ list-delete-neighbor-dups lis = -> list
-+ list-delete-neighbor-dups! lis = -> list
-+
-+ vector-delete-neighbor-dups v = [start end] -> vector
-+ vector-delete-neighbor-dups! v = [start end] -> end'
-+
-+ These procedures delete adjacent duplicate elements from a list or
-+ a vector, using a given element-equality procedure =. The first/leftmost
-+ element of a run of equal elements is the one that survives. The list
-+ or vector is not otherwise disordered.
-+
-+ These procedures are linear time -- much faster than the O(n^2) general
-+ duplicate-element deletors that do not assume any "bunching" of elements
-+ (such as the ones provided by SRFI-1). If you want to delete duplicate
-+ elements from a large list or vector, sort the elements to bring equal
-+ items together, then use one of these procedures, for a total time of
-+ O(n lg n).
-+
-+ The comparison function = passed to these procedures is always applied
-+ (= x y)
-+ where X comes before Y in the containing list or vector.
-+
-+ LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
-+ may share storage with the input list.
-+
-+ VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
-+ rather allocates a fresh vector to hold the result.
-+
-+ LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
-+ mutate its input list in order to construct its answer.
-+
-+ VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
-+ answer, packing its answer into the index range [start,end'), where
-+ END' is the non-negative exact integer returned as its value. It
-+ returns END' as its result. The vector is not altered outside the range
-+ [start,end').
-+
-+ Examples:
-+ (list-delete-neighbor-dups '(1 1 2 7 7 7 0 -2 -2) =)
-+ => (1 2 7 0 -2)
-+
-+ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) =)
-+ => #(1 2 7 0 -2)
-+
-+ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) = 3 7)
-+ => #(7 0 -2)
-+
-+ ;; Result left in v[3,9):
-+ (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
-+ (cons (vector-delete-neighbor-dups! v = 3)
-+ v))
-+ => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
-+
-+
-+-------------------------------------------------------------------------------
-+* Algorithmic properties
-+------------------------
-+Different sort and merge algorithms have different properties.
-+Choose the algorithm that matches your needs:
-+
-+Vector insert sort
-+ Stable, but only suitable for small vectors -- O(n^2).
-+
-+Vector quick sort
-+ Not stable. Is fast on average -- O(n lg n) -- but has bad worst-case
-+ behaviour. Has good memory locality for big vectors (unlike heap sort).
-+ A clever pivot-picking trick (median of three samples) helps avoid
-+ worst-case behaviour, but pathological cases can still blow up.
-+
-+Vector heap sort
-+ Not stable. Guaranteed fast -- O(n lg n) *worst* case. Poor locality
-+ on large vectors. A very reliable workhorse.
-+
-+Vector merge sort
-+ Stable. Not in-place -- requires a temporary buffer of equal size.
-+ Fast -- O(n lg n) -- and has good memory locality for large vectors.
-+
-+ The implementation of vector merge sort provided by this SRFI's reference
-+ implementation is, additionally, a "natural" sort, meaning that it
-+ exploits existing order in the input data, providing O(n) best case.
-+
-+Destructive list merge sort
-+ Stable, fast and in-place (i.e., allocates no new cons cells). "Fast"
-+ means O(n lg n) worse-case, and substantially better if the data
-+ is already mostly ordered, all the way down to linear time for
-+ a completely-ordered input list (i.e., it is a "natural" sort).
-+
-+ Note that sorting lists involves chasing pointers through memory, which
-+ can be a loser on modern machine architectures because of poor cache &
-+ page locality. Pointer *writing*, which is what the SET-CDR!s of a
-+ destructive list-sort algorithm do, is even worse, especially if your
-+ Scheme has a generational GC -- the writes will thrash the write-barrier.
-+ Sorting vectors has inherently better locality.
-+
-+ This SRFIs destructive list merge and merge sort implementations are
-+ opportunistic -- they avoid redundant SET-CDR!s, and try to take long
-+ already-ordered runs of list structure as-is when doing the merges.
-+
-+Pure list merge sort
-+ Stable and fast -- O(n lg n) worst-case, and possibly better, depending
-+ upon the input list (see above).
-+
-+
-+Algorithm Stable? Worst case Average case In-place
-+------------------------------------------------------
-+V insert Yes O(n^2) O(n^2) Yes
-+V quick No O(n^2) O(n lg n) Yes
-+V heap No O(n lg n) O(n lg n) Yes
-+V merge Yes O(n lg n) O(n lg n) No
-+L merge Yes O(n lg n) O(n lg n) Either
-+
-+
-+-------------------------------------------------------------------------------
-+* Topics to be resolved during discussion phase
-+-----------------------------------------------
-+I particularly solicit comments about the following topics.
-+
-+- Include VECTOR-BINARY-SEARCH ?
-+ Should we include
-+ (VECTOR-BINARY-SEARCH v key< elt->key key [start end])
-+ in the SRFI? It sort of goes with sorting; it's exactly ten lines of code.
-+
-+- Comparison function before or after the list/vector argument?
-+ Should it be
-+ (list-sort < lis)
-+ or
-+ (list-sort lis <)
-+ There is overwhelming consistency among the implementations: data first,
-+ < after. Only Chez does it differently.
-+
-+ I have done it in the backwards-compatible way. But I prefer the < first,
-+ data after way.
-+
-+
-+-------------------------------------------------------------------------------
-+* Porting and optimisation
-+--------------------------
-+This package should be trivial to port. There are only four non-R4RS bits
-+in the code:
-+- Use of multiple-value return, with the R5RS VALUES procedure, and the
-+ simple (RECEIVE (var ...) mv-exp body ...) multiple-value binding macro.
-+
-+- A VECTOR-COPY procedure. This is a tiny little procedure:
-+ (vector-copy v [start end])
-+
-+- Use of the LET-OPTIONALS macro from scsh to parse and default optional
-+ arguments to three routines. Again, easy to port the macro or rewrite
-+ the code to parse, default, and error check the args by hand.
-+
-+- Calls to an ERROR function for complaining about bad arguments.
-+
-+This code is tightly bummed, as far as I can go in portable Scheme.
-+
-+You could speed up the vector code a lot by error-checking the procedure
-+parameters and then shifting over to fixnum-specific arithmetic and
-+dangerous vector-indexing and vector-setting primitives. The comments
-+in the code indicate where the initial error checks would have to be
-+added. There are several (QUOTIENT N 2)'s that could be changed to a
-+fixnum right-shift, as well, in both the list and vector code. The code
-+is designed to enable this -- each file usually exports one or two "safe"
-+procedures that end up calling an internal "dangerous" primitive. The
-+little exported cover procedures are where you move the error checks.
-+
-+This should provide *big* speedups. In fact, all the code bumming I've done
-+pretty much disappears in the noise unless you have a good compiler and also
-+can dump the vector-index checks and generic arithmetic -- so I've really just
-+set things up for you to exploit.
-+
-+The optional-arg parsing, defaulting, and error checking is done with a
-+portable R4RS macro. But if your Scheme has a faster mechanism (e.g., Chez),
-+you should definitely port over to it. Note that argument defaulting and
-+error-checking are interleaved -- you don't have to error-check defaulted
-+START/END args to see if they are fixnums that are legal vector indices for
-+the corresponding vector, etc.
-+
-+
-+-------------------------------------------------------------------------------
-+* References & Links
-+--------------------
-+
-+This document, in HTML:
-+ http://srfi.schemers.org/srfi-32/srfi-32.html
-+ [This link may not be valid while the SRFI is in draft form.]
-+
-+This document, in simple text format:
-+ http://srfi.schemers.org/srfi-32/srfi-32.txt
-+
-+Archive of SRFI-32 discussion-list email:
-+ http://srfi.schemers.org/srfi-32/mail-archive/maillist.html
-+
-+SRFI web site:
-+ http://srfi.schemers.org/
-+
-+[CommonLisp]
-+ Common Lisp: the Language
-+ Guy L. Steele Jr. (editor).
-+ Digital Press, Maynard, Mass., second edition 1990.
-+ Available at http://www.elwood.com/alu/table/references.htm#cltl2
-+
-+ The Common Lisp "HyperSpec," produced by Kent Pitman, is essentially
-+ the ANSI spec for Common Lisp:
-+ http://www.xanalys.com/software_tools/reference/HyperSpec/
-+
-+[R5RS]
-+ Revised^5 Report on the Algorithmic Language Scheme,
-+ R. Kelsey, W. Clinger, J. Rees (editors).
-+ Higher-Order and Symbolic Computation, Vol. 11, No. 1, September, 1998.
-+ and ACM SIGPLAN Notices, Vol. 33, No. 9, October, 1998.
-+
-+ Available at http://www.schemers.org/Documents/Standards/
-+
-+
-+-------------------------------------------------------------------------------
-+* Acknowledgements
-+------------------
-+
-+I thank the authors of the open source I consulted when designing this
-+library, particularly Richard O'Keefe, Donovan Kolby and the MIT Scheme Team.
-+
-+
-+-------------------------------------------------------------------------------
-+* Copyright
-+-----------
-+
-+** SRFI text
-+============
-+This document is copyright (C) Olin Shivers (1998, 1999).
-+All Rights Reserved.
-+
-+Permission is hereby granted, free of charge, to any person obtaining
-+a copy of this software and associated documentation files (the
-+"Software"), to deal in the Software without restriction, including
-+without limitation the rights to use, copy, modify, merge, publish,
-+distribute, sublicense, and/or sell copies of the Software, and to
-+permit persons to whom the Software is furnished to do so, subject to
-+the following conditions:
-+
-+The above copyright notice and this permission notice shall be
-+included in all copies or substantial portions of the Software.
-+
-+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
-+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
-+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
-+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
-+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
-+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
-+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
-+
-+** Reference implementation
-+===========================
-+Short summary: no restrictions.
-+
-+While I wrote all of this code myself, I read a lot of code before I began
-+writing. However, all such code is, itself, either open source or public
-+domain, rendering irrelevant any issue of "copyright taint."
-+
-+The natural merge sorts (pure list, destructive list, and vector) are not only
-+my own code, but are implementations of an algorithm of my own devising. They
-+run in O(n lg n) worst case, O(n) best case, and require only a logarithmic
-+number of stack frames. And they are stable. And the destructive-list variant
-+allocates zero cons cells; it simply rearranges the cells of the input list.
-+
-+Hence the reference implementation is
-+ Copyright (c) 1998 by Olin Shivers.
-+and made available under the same copyright as the SRFI text (see above).
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib/srfi/%3a5/let.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5/let.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1 +0,0 @@
--#lang s-exp srfi/provider srfi/%3a5
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib/srfi/%3a5.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,3 +0,0 @@
--#lang s-exp srfi/provider srfi/5
--
--;; FIXME: "rest arguments" need to generate mutable lists
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib/srfi/5/let.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/5/let.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,86 +0,0 @@
--;;;
--;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
--;;; Time-stamp: <03/04/08 09:56:06 solsona>
--;;;
--;;; Usually, I would add a copyright notice, and the announce that
--;;; this code is under the LGPL licence. Nevertheless, I only did the
--;;; port to PLT Scheme v200, and here is the copyright notice,
--;;; comments, and licence from the original source:
--;;;
--;;; Copyright (C) Andy Gaynor (1999-2003)
--;;;
--;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
--;;; Renamed to s:let by Eli Barzilay
--
--#lang scheme/base
--(provide s:let)
--
--(define-syntax s:let
-- (syntax-rules ()
-- ;; standard
-- ((s:let () body ...)
-- (let () body ...))
-- ((s:let ((var val) ...) body ...)
-- (let ((var val) ...) body ...))
--
-- ;; rest style
-- ((s:let ((var val) . bindings) body ...)
-- (let-loop #f bindings (var) (val) (body ...)))
--
-- ;; signature style
-- ((s:let (name bindings ...) body ...)
-- (let-loop name (bindings ...) () () (body ...)))
--
-- ;; standard named style
-- ((s:let name (bindings ...) body ...)
-- (let-loop name (bindings ...) () () (body ...)))
--
-- ))
--
--;; A loop to walk down the list of bindings.
--
--(define-syntax let-loop
-- (syntax-rules ()
--
-- ;; No more bindings - make a LETREC.
-- ((let-loop name () (vars ...) (vals ...) body)
-- ((letrec ((name (lambda (vars ...) . body)))
-- name)
-- vals ...))
--
-- ;; Rest binding, no name
-- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-- (let ((var val) ... (rest-var (list rest-val ...))) . body))
--
-- ;; Process a (var val) pair.
-- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
-- (let-loop name (more ...) (vars ... var) (vals ... val) body))
--
-- ;; End with a rest variable - make a LETREC.
-- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
-- ((letrec ((name (lambda (vars ... . rest-var) . body)))
-- name)
-- vals ... rest-vals ...))))
--
--;; Four loops - normal and `signature-style', each with and without a rest
--;; binding.
--;;
--;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
--;; (if (= i n)
--;; f0
--;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
--;;
--;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
--;; (if (= i n)
--;; f0
--;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
--;;
--;;(let fibonacci ((n 10) (i 0) . (f 0 1))
--;; (if (= i n)
--;; (car f)
--;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
--;;
--;;(let (fibonacci (n 10) (i 0) . (f 0 1))
--;; (if (= i n)
--;; (car f)
--;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib/srfi/5.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/5.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,2 +0,0 @@
--;; module loader for SRFI-5
--#lang s-exp srfi/provider srfi/5/let #:unprefix s:
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/info.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1,14 @@
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define deps '("scheme-lib"
-+ "base"
-+ "r6rs-lib"
-+ "srfi-lib"
-+ "compatibility-lib"))
-+
-+
-+(define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1 @@
-+#lang s-exp srfi/provider srfi/%3a5
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1,3 @@
-+#lang s-exp srfi/provider srfi/5
-+
-+;; FIXME: "rest arguments" need to generate mutable lists
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1,86 @@
-+;;;
-+;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
-+;;; Time-stamp: <03/04/08 09:56:06 solsona>
-+;;;
-+;;; Usually, I would add a copyright notice, and the announce that
-+;;; this code is under the LGPL licence. Nevertheless, I only did the
-+;;; port to PLT Scheme v200, and here is the copyright notice,
-+;;; comments, and licence from the original source:
-+;;;
-+;;; Copyright (C) Andy Gaynor (1999-2003)
-+;;;
-+;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
-+;;; Renamed to s:let by Eli Barzilay
-+
-+#lang scheme/base
-+(provide s:let)
-+
-+(define-syntax s:let
-+ (syntax-rules ()
-+ ;; standard
-+ ((s:let () body ...)
-+ (let () body ...))
-+ ((s:let ((var val) ...) body ...)
-+ (let ((var val) ...) body ...))
-+
-+ ;; rest style
-+ ((s:let ((var val) . bindings) body ...)
-+ (let-loop #f bindings (var) (val) (body ...)))
-+
-+ ;; signature style
-+ ((s:let (name bindings ...) body ...)
-+ (let-loop name (bindings ...) () () (body ...)))
-+
-+ ;; standard named style
-+ ((s:let name (bindings ...) body ...)
-+ (let-loop name (bindings ...) () () (body ...)))
-+
-+ ))
-+
-+;; A loop to walk down the list of bindings.
-+
-+(define-syntax let-loop
-+ (syntax-rules ()
-+
-+ ;; No more bindings - make a LETREC.
-+ ((let-loop name () (vars ...) (vals ...) body)
-+ ((letrec ((name (lambda (vars ...) . body)))
-+ name)
-+ vals ...))
-+
-+ ;; Rest binding, no name
-+ ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-+ (let ((var val) ... (rest-var (list rest-val ...))) . body))
-+
-+ ;; Process a (var val) pair.
-+ ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
-+ (let-loop name (more ...) (vars ... var) (vals ... val) body))
-+
-+ ;; End with a rest variable - make a LETREC.
-+ ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
-+ ((letrec ((name (lambda (vars ... . rest-var) . body)))
-+ name)
-+ vals ... rest-vals ...))))
-+
-+;; Four loops - normal and `signature-style', each with and without a rest
-+;; binding.
-+;;
-+;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
-+;; (if (= i n)
-+;; f0
-+;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+;;
-+;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
-+;; (if (= i n)
-+;; f0
-+;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+;;
-+;;(let fibonacci ((n 10) (i 0) . (f 0 1))
-+;; (if (= i n)
-+;; (car f)
-+;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
-+;;
-+;;(let (fibonacci (n 10) (i 0) . (f 0 1))
-+;; (if (= i n)
-+;; (car f)
-+;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1,2 @@
-+;; module loader for SRFI-5
-+#lang s-exp srfi/provider srfi/5/let #:unprefix s:
-diff -urN racket-6.12-orig/share/pkgs/srfi-lite-lib/info.rkt racket-6.12/share/pkgs/srfi-lite-lib/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-lite-lib/info.rkt 2018-01-26 16:10:04.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lite-lib/info.rkt 2018-01-31 12:22:58.160501940 -0500
-@@ -1 +1,9 @@
--(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("base"))) (define pkg-desc "implementation of the most widely used \"srfi\" libraries") (define pkg-authors (quote (mflatt)))))
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define deps '("base"))
-+
-+(define pkg-desc "implementation of the most widely used \"srfi\" libraries")
-+
-+(define pkg-authors '(mflatt))
-Binary files racket-6.12-orig/share/pkgs/srfi-lite-lib/.info.rkt.swp and racket-6.12/share/pkgs/srfi-lite-lib/.info.rkt.swp differ
-diff -urN racket-6.12-orig/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt racket-6.12/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt
---- racket-6.12-orig/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt 2018-01-31 12:22:58.162501933 -0500
-@@ -2,6 +2,7 @@
-
- (require racket/contract/base
- racket/file
-+ racket/list
- (only-in racket/runtime-path define-runtime-path)
- racket/string racket/format
- syntax/modread
-@@ -113,10 +114,6 @@
- (lambda ()
- (with-input-from-file path read))))
- #t))))
-- (define (rdc ls)
-- (if (null? (cdr ls))
-- '()
-- (cons (car ls) (rdc (cdr ls)))))
-
- ;;Retrieve a localized template given its package name and a template name
- (define (localized-template package-name template-name)
-@@ -127,5 +124,5 @@
- (let ((bundle (hash-ref *localization-bundles* specifier #f)))
- (cond ((and bundle (assq template-name bundle)) => cdr)
- ((null? (cdr specifier)) #f)
-- (else (loop (rdc specifier))))))))
-+ (else (loop (drop-right specifier 1))))))))
- )
diff --git a/racket-7.0-fix-doc-open-url.patch b/racket-7.0-fix-doc-open-url.patch
new file mode 100644
index 0000000..7c497f5
--- /dev/null
+++ b/racket-7.0-fix-doc-open-url.patch
@@ -0,0 +1,69 @@
+diff -urN racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt racket-6.12/share/pkgs/scribble-lib/help/search.rkt
+--- racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:14:56.017273013 -0400
++++ racket-6.12/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:15:37.839113929 -0400
+@@ -17,33 +17,38 @@
+ #:notify [notify void])
+ (define open-url (get-doc-open-url))
+ (cond
+- [open-url
+- (define dest-url (let ([u (string->url open-url)])
+- (combine-url/relative
+- u
+- (string-join
+- (for/list ([s (explode-path sub)])
+- (if (path? s)
+- (path-element->string s)
+- (format "~a" s)))
+- "/"))))
+- (notify (url->string dest-url))
+- (send-url (url->string
+- (struct-copy url dest-url
+- [fragment (or fragment
+- (url-fragment dest-url))]
+- [query (append
+- (url-query dest-url)
+- (if query
+- (url-query
+- (string->url
+- (format "q?~a" query)))
+- null))])))]
+- [else
+- (let* ([path (build-path (find-user-doc-dir) sub)]
+- [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
+- (notify path)
+- (send-url/file path #:fragment fragment #:query query))]))
++ [open-url
++ (define dest-url (let ([u (string->url open-url)])
++ (combine-url/relative
++ u
++ (string-join
++ (for/list ([s (explode-path sub)])
++ (if (path? s)
++ (path-element->string s)
++ (format "~a" s)))
++ "/"))))
++ (notify (url->string dest-url))
++ (send-url (url->string
++ (struct-copy url dest-url
++ [fragment (or fragment
++ (url-fragment dest-url))]
++ [query (append
++ (url-query dest-url)
++ (if query
++ (url-query
++ (string->url
++ (format "q?~a" query)))
++ null))])))]
++ [else
++ (let* ([path (build-path (find-user-doc-dir) sub)]
++ [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
++ (notify path)
++ (if (file-exists? path)
++ (send-url/file path #:fragment fragment #:query query)
++ (let ([part (lambda (pfx x) (if x (string-append pfx x) ""))])
++ (send-url (string-append
++ "https://docs.racket-lang.org/"
++ sub (part "#" fragment) (part "?" query))))))]))
+
+ ;; This is an example of changing this code to use the online manuals.
+ ;; Normally, it's better to set `doc-open-url` in "etc/config.rktd",
diff --git a/racket-7.0-remove-nonfree.patch b/racket-7.0-remove-nonfree.patch
new file mode 100644
index 0000000..778001a
--- /dev/null
+++ b/racket-7.0-remove-nonfree.patch
@@ -0,0 +1,1074 @@
+diff -urN racket-7.0.orig/share/links.rktd racket-7.0/share/links.rktd
+--- racket-7.0.orig/share/links.rktd 2018-08-03 15:23:48.604561727 -0400
++++ racket-7.0/share/links.rktd 2018-08-03 15:23:45.481540939 -0400
+@@ -175,9 +175,7 @@
+ ("sasl" "pkgs/sasl-doc")
+ (root "pkgs/slideshow-exe")
+ (root "pkgs/slideshow-plugin")
+- (root "pkgs/srfi-lib-nonfree")
+ (root "pkgs/srfi-doc")
+- (root "pkgs/srfi-doc-nonfree")
+ (root "pkgs/syntax-color-doc")
+ (root "pkgs/web-server-lib")
+ (root "pkgs/unix-socket-lib")
+diff -urN racket-7.0.orig/share/pkgs/srfi/info.rkt racket-7.0/share/pkgs/srfi/info.rkt
+--- racket-7.0.orig/share/pkgs/srfi/info.rkt 2018-08-03 15:23:48.891563638 -0400
++++ racket-7.0/share/pkgs/srfi/info.rkt 2018-08-03 15:23:45.424540559 -0400
+@@ -1 +1 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "7.0"))) (define collection (quote multi)) (define deps (quote ("srfi-lib" "srfi-lib-nonfree" "srfi-doc" "srfi-doc-nonfree"))) (define implies (quote ("srfi-lib" "srfi-lib-nonfree" "srfi-doc" "srfi-doc-nonfree"))) (define pkg-desc "Legacy SRFI (Scheme) libraries") (define pkg-authors (quote (mflatt noel chongkai jay)))))
++(module info setup/infotab (#%module-begin (define package-content-state (quote (built "7.0"))) (define collection (quote multi)) (define deps (quote ("srfi-lib" "srfi-doc"))) (define implies (quote ("srfi-lib" "srfi-doc"))) (define pkg-desc "Legacy SRFI (Scheme) libraries") (define pkg-authors (quote (mflatt noel chongkai jay)))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/info.rkt racket-7.0/share/pkgs/srfi-doc-nonfree/info.rkt
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/info.rkt 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "7.0"))) (define collection (quote multi)) (define build-deps (quote ("mzscheme-doc" "scheme-lib" "base" "scribble-lib" "srfi-doc" "srfi-lib-nonfree" "racket-doc" "r5rs-doc" "r6rs-doc" "compatibility-lib"))) (define update-implies (quote ("srfi-lib-nonfree"))) (define pkg-desc "documentation part of \"srfi nonfree\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang info
+-
+-(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
+@@ -1,66 +0,0 @@
+-#lang scribble/doc
+-@(require srfi/scribblings/util
+- scribble/manual
+- scribble/eval
+- scriblib/render-cond
+- scribble/core
+- scribble/html-properties
+- (for-syntax scheme/base)
+- (for-label scheme/base
+- racket/stream))
+-
+-@; ----------------------------------------------------------------------
+-
+-@title{SRFI Nonfree Libraries and Documentation}
+-
+-The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
+-Implementation} (a.k.a. @deftech{SRFI}) process allows individual
+-members of the Scheme community to propose libraries and extensions to
+-be supported by multiple Scheme implementations.
+-
+-Racket is distributed with implementations of many SRFIs, most of
+-which can be implemented as libraries. To import the bindings of SRFI
+-@math{n}, use
+-
+-@racketblock[
+-(require @#,elem{@racketidfont{srfi/}@math{n}})
+-]
+-
+-This document lists the SRFIs that are supported by Racket and
+-provides a link to the original SRFI specification (which is also
+-distributed as part of Racket's documentation).
+-
+-The following SRFI specification documents are licensed restrictively.
+-
+-@table-of-contents[]
+-
+-
+-@; ----------------------------------------
+-
+-@srfi[5]{A compatible let form with signatures and rest arguments}
+-
+-@redirect[5 '(
+- (let #t "unnamed")
+-)]
+-
+-Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
+-
+-@; ----------------------------------------
+-
+-@srfi[29]{Localization}
+-
+-@redirect[29 '(
+- (current-language #f "current-language")
+- (current-country #f "current-country")
+- (current-locale-details #f "current-locale-details")
+- (declare-bundle! #f "declare-bundle!")
+- (store-bundle #f "store-bundle")
+- (load-bundle! #f "load-bundle!")
+- (localized-template #f "localized-template")
+-)]
+-
+-Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
+-
+-@; ----------------------------------------
+-
+-@index-section[]
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,507 +0,0 @@
+-<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
+-<html>
+- <head>
+- <meta name="generator" content="HTML Tidy, see www.w3.org">
+- <title>SRFI 29: Localization</title>
+- <meta name="author" content="Scott G. Miller">
+- <meta name="description" content="Localization">
+- </head>
+- <body>
+- <H1>Title</H1>
+-
+- SRFI 29: Localization
+-
+- <H1>Author</H1>
+-
+- Scott G. Miller
+-
+- <H1>Abstract</H1>
+-
+- This document specifies an interface to retrieving and
+- displaying locale sensitive messages. A Scheme program can
+- register one or more translations of templated messages, and
+- then write Scheme code that can transparently retrieve the
+- appropriate message for the locale under which the Scheme
+- system is running. <br>
+-
+-
+- <H1>Rationale</H1>
+-
+- <p>As any programmer that has ever had to deal with making his
+- or her code readable in more than one locale, the process of
+- sufficiently abstracting program messages from their
+- presentation to the user is non-trivial without help from the
+- programming language. Most modern programming language
+- libraries do provide some mechanism for performing this
+- separation.</p>
+-
+- <p>A portable API that allows a piece of code to run without
+- modification in different countries and under different
+- languages is a must for any non-trivial software project.
+- The interface should separate the logic of a program from
+- the myriad of translations that may be necessary.</p>
+-
+- <p>The interface described in this document provides such
+- functionality. The underlying implementation is also allowed to
+- use whatever datastructures it likes to provide access to the
+- translations in the most efficient manner possible. In
+- addition, the implementation is provided with standardized
+- functions that programs will use for accessing an external,
+- unspecified repository of translations.</p>
+-
+- <p>This interface <i>does not</i> cover all aspects of
+- localization, including support for non-latin characters,
+- number and date formatting, etc. Such functionality is the
+- scope of a future SRFI that may extend this one.</p>
+-
+- <H1>Dependencies</H1>
+-
+- An SRFI-29 conformant implementation must also implement
+- SRFI-28, Basic Format Strings. Message templates are strings
+- that must be processed by the <tt>format</tt> function
+- specified in that SRFI.
+-
+- <H1>Specification</H1>
+-
+- <h3>Message Bundles</h3>
+-
+- <p>A Message Bundle is a set of message templates and their
+- identifying keys. Each bundle contains one or more such
+- key/value pairs. The bundle itself is associated with a
+- <i>bundle specifier</i> which uniquely identifies the
+- bundle.</p>
+-
+- <h3>Bundle Specifiers</h3>
+-
+- <p>A Bundle Specifier is a Scheme list that describes, in order
+- of importance, the package and locale of a message bundle.
+- In most cases, a locale specifier will have between one
+- and three elements. The first element is a symbol denoting the
+- package for which this bundle applies. The second and third
+- elements denote a <i>locale</i>. The second element (first
+- element of the locale) if present, is the two letter, ISO 639-1
+- language code for the bundle. The third element, if present, is
+- a two letter ISO 3166-1 country code. In some cases, a
+- fourth element may be present, specifying the encoding used for
+- the bundle. All bundle specifier elements are Scheme
+- symbols.</p>
+-
+- <p>If only one translation is provided, it should be designated
+- only by a package name, for example <tt>(mathlib)</tt>. This
+- translation is called the <i>default</i> translation.</p>
+-
+- <h3>Bundle Searching</h3>
+-
+- <p>When a message template is retrieved from a bundle, the
+- Scheme implementation will provide the locale under which the
+- system is currently running. When the template is retrieved,
+- the package name will be specified. The Scheme system should
+- construct a Bundle Specifier from the provided package name and
+- the active locale. For example, when retrieving a message
+- template for French Canadian, in the <tt>mathlib</tt> package,
+- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
+- program may also retrieve the elements of the current locale
+- using the no-argument procedures:</p>
+-
+- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-language</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- When given no arguments, returns the current ISO 639-1
+- language code as a symbol. If provided with an
+- argument, the current language is set to that named by the
+- symbol for the currently executing Scheme thread (or for the
+- entire Scheme system if such a distinction is not possible).
+-
+- </blockquote>
+-
+- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-country</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- returns the current ISO 3166-1 country code as a symbol.
+- If provided with an argument, the current country is
+- set to that named by the symbol for the currently executing
+- Scheme thread (or for the entire Scheme system if such a
+- distinction is not possible).
+- </blockquote>
+-
+- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
+- symbol</i></tt>s<br>
+- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- Returns a list of additional locale details as a list of
+- symbols. This list may contain information about
+- encodings or other more specific information. If
+- provided with an argument, the current locale details are set
+- to those given in the currently executing Scheme thread (or
+- for the entire Scheme system if such a distinction is not
+- possible).
+- </blockquote>
+-
+- <p>The Scheme System should first check for a bundle with the
+- exact name provided. If no such bundle is found, the last
+- element from the list is removed and a search is tried for a
+- bundle with that name. If no bundle is then found, the list is
+- shortened by removing the last element again. If no message is
+- found and the bundle specifier is now the empty list, an error
+- should be raised.</p>
+-
+- <p>The reason for this search order is to provide the most
+- locale sensitive template possible, but to fall back on more
+- general templates if a translation has not yet been provided
+- for the given locale.</p>
+-
+- <h3>Message Templates</h3>
+-
+- <p>A message template is a localized message that may or may
+- not contain one of a number of formatting codes. A message
+- template is a Scheme string. The string is of a form that can
+- be processed by the <tt>format</tt> procedure found in many
+- Scheme systems and formally specified in SRFI-28 (Basic Format
+- Strings).</p>
+-
+- <p>This SRFI also extends SRFI-28 to provide an additional
+- <tt>format</tt> escape code:</p>
+-
+- <blockquote>
+- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
+- follows this code immediately to reference the [N]'th
+- optional value absolutely, rather than the next unconsumed
+- value. The referenced value is <i>not</i> consumed.
+- </blockquote>
+- This extension allows optional values to be positionally
+- referenced, so that message templates can be constructed that
+- can produce the proper word ordering for a language.
+-
+- <h3>Preparing Bundles</h3>
+- Before a bundle may be used by the Scheme system to retrieve
+- localized template messages, they must be made available to the
+- Scheme system. This SRFI specifies a way to portably
+- define the bundles, as well as store them in and retrieve them
+- from an unspecified system which may be provided by resources
+- outside the Scheme system.<br>
+-
+-
+- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
+- association-list</i> -> undefined<br>
+- </tt></p>
+-
+- <blockquote>
+- Declares a new bundle named by the given bundle-specifier.
+- The contents of the bundle are defined by the provided
+- association list. The list contains associations
+- between Scheme symbols and the message templates (Scheme
+- strings) they name. If a bundle already exists with the
+- given name, it is overwritten with the newly declared
+- bundle.<br>
+- </blockquote>
+- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to store a bundle named by the given bundle
+- specifier, and previously made available using
+- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
+- unspecified mechanism that may be persistent across Scheme
+- system restarts. If successful, a non-false value is
+- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
+- </blockquote>
+- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to retrieve a bundle from an unspecified mechanism
+- which stores bundles outside the Scheme system. If the
+- bundle was retrieved successfully, the function returns a
+- non-false value, and the bundle is immediately available to
+- the Scheme system. If the bundle could not be found or loaded
+- successfully, the function returns <tt>#f</tt>, and the
+- Scheme system's bundle registry remains unaffected.<br>
+- </blockquote>
+- A compliant Scheme system may choose not to provide any
+- external mechanism to store localized bundles. If it does
+- not, it must still provide implementations for
+- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
+- case, both functions must return <tt>#f</tt> regardless of the
+- arguments given. Users of this SRFI should recognize that the
+- inability to load or store a localized bundle in an external
+- repository is <i>not</i> a fatal error.<br>
+-
+-
+- <h3>Retrieving Localized Message Templates</h3>
+-
+- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
+- message-template-name</tt></i> <tt>-> <i>string or #f<br>
+- </i></tt></p>
+-
+- <blockquote>
+- Retrieves a localized message template for the given package
+- name and the given message template name (both symbols).
+- If no such message could be found, false (#f) is
+- returned.<br>
+- <br>
+- </blockquote>
+- After retrieving a template, the calling program can use
+- <tt>format</tt> to produce a string that can be displayed to
+- the user.<br>
+-
+-
+- <h2>Examples</h2>
+- The below example makes use of SRFI-29 to display simple,
+- localized messages. It also defines its bundles in such a
+- way that the Scheme system may store and retrieve the bundles
+- from a more efficient system catalog, if available.<br>
+-
+-<pre>
+-(let ((translations
+- '(((en) . ((time . "Its ~a, ~a.")
+- (goodbye . "Goodbye, ~a.")))
+- ((fr) . ((time . "~1@*~a, c'est ~a.")
+- (goodbye . "Au revoir, ~a."))))))
+- (for-each (lambda (translation)
+- (let ((bundle-name (cons 'hello-program (car translation))))
+- (if (not (load-bundle! bundle-name))
+- (begin
+- (declare-bundle! bundle-name (cdr translation))
+- (store-bundle! bundle-name)))))
+- translations))
+-
+-(define localized-message
+- (lambda (message-name . args)
+- (apply format (cons (localized-template 'hello-program
+- message-name)
+- args))))
+-
+-(let ((myname "Fred"))
+- (display (localized-message 'time "12:00" myname))
+- (display #\newline)
+-
+- (display (localized-message 'goodbye myname))
+- (display #\newline))
+-
+-;; Displays (English):
+-;; Its 12:00, Fred.
+-;; Goodbye, Fred.
+-;;
+-;; French:
+-;; Fred, c'est 12:00.
+-;; Au revoir, Fred.
+-</pre>
+-
+- <H1>Implementation</H1>
+-
+- <p>The implementation requires that the Scheme system provide a
+- definition for <tt>current-language</tt> and
+- <tt>current-country</tt> capable of distinguishing the correct
+- locale present during a Scheme session. The definitions of
+- those functions in the reference implementation are not capable
+- of that distinction. Their implementation is provided only so
+- that the following code can run in any R4RS scheme system.
+- <br>
+- </p>
+-
+- <p>In addition, the below implementation of a compliant
+- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
+- SRFI-23 (Error reporting)</p>
+-<pre>
+-;; The association list in which bundles will be stored
+-(define *localization-bundles* '())
+-
+-;; The current-language and current-country functions provided
+-;; here must be rewritten for each Scheme system to default to the
+-;; actual locale of the session
+-(define current-language
+- (let ((current-language-value 'en))
+- (lambda args
+- (if (null? args)
+- current-language-value
+- (set! current-language-value (car args))))))
+-
+-(define current-country
+- (let ((current-country-value 'us))
+- (lambda args
+- (if (null? args)
+- current-country-value
+- (set! current-country-value (car args))))))
+-
+-;; The load-bundle! and store-bundle! both return #f in this
+-;; reference implementation. A compliant implementation need
+-;; not rewrite these procedures.
+-(define load-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-(define store-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-;; Declare a bundle of templates with a given bundle specifier
+-(define declare-bundle!
+- (letrec ((remove-old-bundle
+- (lambda (specifier bundle)
+- (cond ((null? bundle) '())
+- ((equal? (caar bundle) specifier)
+- (cdr bundle))
+- (else (cons (car bundle)
+- (remove-old-bundle specifier
+- (cdr bundle))))))))
+- (lambda (bundle-specifier bundle-assoc-list)
+- (set! *localization-bundles*
+- (cons (cons bundle-specifier bundle-assoc-list)
+- (remove-old-bundle bundle-specifier
+- *localization-bundles*))))))
+-
+-;;Retrieve a localized template given its package name and a template name
+-(define localized-template
+- (letrec ((rdc
+- (lambda (ls)
+- (if (null? (cdr ls))
+- '()
+- (cons (car ls) (rdc (cdr ls))))))
+- (find-bundle
+- (lambda (specifier template-name)
+- (cond ((assoc specifier *localization-bundles*) =>
+- (lambda (bundle) bundle))
+- ((null? specifier) #f)
+- (else (find-bundle (rdc specifier)
+- template-name))))))
+- (lambda (package-name template-name)
+- (let loop ((specifier (cons package-name
+- (list (current-language)
+- (current-country)))))
+- (and (not (null? specifier))
+- (let ((bundle (find-bundle specifier template-name)))
+- (and bundle
+- (cond ((assq template-name bundle) => cdr)
+- ((null? (cdr specifier)) #f)
+- (else (loop (rdc specifier)))))))))))
+-
+-;;An SRFI-28 and SRFI-29 compliant version of format. It requires
+-;;SRFI-23 for error reporting.
+-(define format
+- (lambda (format-string . objects)
+- (let ((buffer (open-output-string)))
+- (let loop ((format-list (string->list format-string))
+- (objects objects)
+- (object-override #f))
+- (cond ((null? format-list) (get-output-string buffer))
+- ((char=? (car format-list) #\~)
+- (cond ((null? (cdr format-list))
+- (error 'format "Incomplete escape sequence"))
+- ((char-numeric? (cadr format-list))
+- (let posloop ((fl (cddr format-list))
+- (pos (string->number
+- (string (cadr format-list)))))
+- (cond ((null? fl)
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (null? (cdr fl)))
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (eq? (cadr fl) '#\*))
+- (loop (cddr fl) objects (list-ref objects pos)))
+- (else
+- (posloop (cdr fl)
+- (+ (* 10 pos)
+- (string->number
+- (string (car fl)))))))))
+- (else
+- (case (cadr format-list)
+- ((#\a)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (display (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\s)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (write (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\%)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\newline buffer)
+- (loop (cddr format-list) objects #f))
+- ((#\~)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\~ buffer)
+- (loop (cddr format-list) objects #f))
+- (else
+- (error 'format "Unrecognized escape sequence"))))))
+- (else (display (car format-list) buffer)
+- (loop (cdr format-list) objects #f)))))))
+-
+-</pre>
+-
+- <H1>Copyright</H1>
+-
+- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
+-
+- <p>This document and translations of it may be copied and
+- furnished to others, and derivative works that comment on or
+- otherwise explain it or assist in its implementation may be
+- prepared, copied, published and distributed, in whole or in
+- part, without restriction of any kind, provided that the above
+- copyright notice and this paragraph are included on all such
+- copies and derivative works. However, this document itself may
+- not be modified in any way, such as by removing the copyright
+- notice or references to the Scheme Request For Implementation
+- process or editors, except as needed for the purpose of
+- developing SRFIs in which case the procedures for copyrights
+- defined in the SRFI process must be followed, or as required to
+- translate it into languages other than English.</p>
+-
+- <p>The limited permissions granted above are perpetual and will
+- not be revoked by the authors or their successors or
+- assigns.</p>
+-
+- <p>This document and the information contained herein is
+- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
+- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
+- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
+- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
+- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
+- <hr>
+-
+- <address>
+- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
+- Rush</a>
+- </address>
+-
+- <address>
+- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
+- Miller</a>
+- </address>
+- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
+- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
+- Daylight Time 2002 <!-- hhmts end --> <br>
+- </body>
+-</html>
+-
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,345 +0,0 @@
+-<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
+-<html>
+-<head>
+- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
+-</head>
+-<body>
+-
+-<H1>Title</H1>
+-
+-SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
+-
+-<H1>Author</H1>
+-
+-Andy Gaynor
+-
+-<H1>Status</H1>
+-
+-This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
+-You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
+-<P><UL>
+-<LI>Received: 1999/2/2
+-<LI>Draft: 1999/2/10-1999/04/12
+-<LI>Final: 1999/4/26
+-<LI>Revised reference implementation: 2003/01/27
+-</UL>
+-
+-<H1>Abstract</H1>
+-
+-The <i>named-let</i> incarnation of the <code>let</code> form has two slight
+-inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
+-form makes no accommodation for rest arguments, an issue of functionality
+-and consistency. As defined, the <code>let</code> form does not accommodate
+-signature-style syntax, an issue of aesthetics and consistency. Both
+-issues are addressed here in a manner which is compatible with the traditional
+-<code>let</code> form but for minor extensions.
+-
+-<H1>Rationale</H1>
+-
+-<H2>Signature-style Syntax</H2>
+-
+-Consider the following two equivalent definitions:
+-
+-<p><pre>
+-(define fibonacci
+- (lambda (n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
+-
+-(define (fibonacci n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-Although there is a named-let analog for the former form, there is none
+-for the latter. To wit, suppose one wished to compute the 10th element
+-of the Fibonacci sequence using a named let:
+-
+-<p>
+-<pre>
+-(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-Values: 55
+-</pre>
+-
+-As it stands, one cannot equivalently write
+-
+-<p>
+-<pre>
+-(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-which is consistent with <code>define</code>'s signature-style form.
+-<p>Those that favor the signature style may prefer this extension.
+-In any case, it may be more appropriate to include all bound names within
+-the binding section. As presented, this straightforward extension
+-introduces no ambiguity or incompatibility with the existing definition
+-of let.
+-
+-<H2>Rest Arguments</H2>
+-
+-As it stands, one cannot write a named let with rest arguments, as in
+-
+-<p>
+-<pre>
+-(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))
+-</pre>
+-
+-otherwise equivalent to
+-
+-<p>
+-<pre>
+-(letrec ((blast (lambda (port . x)
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))))
+- (blast (current-output-port) (+ 1 2) 4 5))
+-</pre>
+-
+-While this example is rather contrived, the functionality is not.
+-There are several times when the author has used this construct in practice.
+-Regardless, there is little reason to deny the <code>let</code> form access to
+-all the features of lambda functionality.
+-
+-<H2>Symbols in Binding Sections</H2>
+-
+-Both the features above rely upon the placement of symbols in <code>let</code>
+-binding lists (this statement is intentially simplistic). The only
+-other apparent use of such symbol placement is to tersely bind variables
+-to unspecified values. For example, one might desire to use
+-<code>(let (foo bar baz) ...)</code>
+-to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
+-unspecified values.
+-
+-<p>This usage is considered less important in light of the rationales
+-presented above, and an alternate syntax is immediately apparent, as
+-in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
+-be preferable, consistently parenthesizing normal binding clauses.
+-
+-<H1>Specification</H1>
+-
+-<H2>Syntax</H2>
+-
+-<p>
+-A formal specification of the syntax follows. Below, body, expression,
+-and identifier are free. Each instantiation of binding-name must be
+-unique.
+-</p>
+-
+-<p>
+-<pre>
+- let = "(" "let" let-bindings body ")"
+- expressions = nothing | expression expressions
+- let-bindings = let-name bindings
+- |"(" let-name "." bindings ")"
+- let-name = identifier
+- bindings = "(" ")"
+- | rest-binding
+- | "(" normal-bindings ["." rest-binding] ")"
+-normal-bindings = nothing
+- | normal-binding normal-bindings
+-normal-binding = "(" binding-name expression ")"
+- binding-name = identifier
+- rest-binding = "(" binding-name expressions ")"
+-</pre>
+-
+-<p>
+-For clarity and convenience, an informal specification follows.
+-</p>
+-
+-<ol>
+-<li><a name="unnamed">Unnamed</a>
+-
+-<p><pre>
+-(let ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig">
+-Named, non-signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-sig">Named, signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-
+-<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-</li>
+-</ol>
+-
+-<H2>Semantics</H2>
+-
+-Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
+-and <code>letrec</code> forms, respectively.
+-
+-<ul>
+-<li>For informal syntax 1:
+-
+-<p><pre>
+-(($lambda (<parameter>...) <body>...) <argument>...)
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 2 and 3:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...) <body>...)))
+- (<name> <argument>...))
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 4 and 5:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...
+-
+-. <rest-parameter>) <body>...)))
+- (<name> <argument>... <rest-argument>...))
+-</pre>
+-</li>
+-</ul>
+-
+-<H1>Implementation</H1>
+-
+-Here is an implementation using <code>SYNTAX-RULES</code>.
+-
+-<p>
+-<pre>
+-;; Use your own standard let.
+-;; Or call a lambda.
+-;; (define-syntax standard-let
+-;;
+-;; (syntax-rules ()
+-;;
+-;; ((let ((var val) ...) body ...)
+-;; ((lambda (var ...) body ...) val ...))))
+-
+-(define-syntax let
+-
+- (syntax-rules ()
+-
+- ;; No bindings: use standard-let.
+- ((let () body ...)
+- (standard-let () body ...))
+- ;; Or call a lambda.
+- ;; ((lambda () body ...))
+-
+- ;; All standard bindings: use standard-let.
+- ((let ((var val) ...) body ...)
+- (standard-let ((var val) ...) body ...))
+- ;; Or call a lambda.
+- ;; ((lambda (var ...) body ...) val ...)
+-
+- ;; One standard binding: loop.
+- ;; The all-standard-bindings clause didn't match,
+- ;; so there must be a rest binding.
+- ((let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; Signature-style name: loop.
+- ((let (name binding ...) body ...)
+- (let-loop name (binding ...) () () (body ...)))
+-
+- ;; defun-style name: loop.
+- ((let name bindings body ...)
+- (let-loop name bindings () () (body ...)))))
+-
+-(define-syntax let-loop
+-
+- (syntax-rules ()
+-
+- ;; Standard binding: destructure and loop.
+- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
+- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
+-
+- ;; Rest binding, no name: use standard-let, listing the rest values.
+- ;; Because of let's first clause, there is no "no bindings, no name" clause.
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
+- ;; Or call a lambda with a rest parameter on all values.
+- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
+- ;; Or use one of several other reasonable alternatives.
+-
+- ;; No bindings, name: call a letrec'ed lambda.
+- ((let-loop name () (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ...) . body)))
+- name)
+- val ...))
+-
+- ;; Rest binding, name: call a letrec'ed lambda.
+- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ... . rest-var) . body)))
+- name)
+- val ... rest-val ...))))
+-</pre>
+-
+-
+-<H1>Copyright</H1>
+-
+-Copyright (C) Andy Gaynor (1999). All Rights Reserved.
+-<p>This document and translations of it may be copied and furnished to
+-others, and derivative works that comment on or otherwise explain it or
+-assist in its implementation may be prepared, copied, published and distributed,
+-in whole or in part, without restriction of any kind, provided that the
+-above copyright notice and this paragraph are included on all such copies
+-and derivative works. However, this document itself may not be modified
+-in any way, such as by removing the copyright notice or references to the
+-Scheme Request For Implementation process or editors, except as needed
+-for the purpose of developing SRFIs in which case the procedures for copyrights
+-defined in the SRFI process must be followed, or as required to translate
+-it into languages other than English.
+-<p>The limited permissions granted above are perpetual and will not be
+-revoked by the authors or their successors or assigns.
+-<p>This document and the information contained herein is provided on an
+-"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
+-EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
+-USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
+-WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
+-
+- <hr>
+- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
+-
+-</body>
+-</html>
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/info.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/info.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/info.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "7.0"))) (define collection (quote multi)) (define deps (quote ("scheme-lib" "base" "r6rs-lib" "srfi-lib" "compatibility-lib"))) (define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-#lang s-exp srfi/provider srfi/%3a5
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang s-exp srfi/provider srfi/5
+-
+-;; FIXME: "rest arguments" need to generate mutable lists
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,86 +0,0 @@
+-;;;
+-;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
+-;;; Time-stamp: <03/04/08 09:56:06 solsona>
+-;;;
+-;;; Usually, I would add a copyright notice, and the announce that
+-;;; this code is under the LGPL licence. Nevertheless, I only did the
+-;;; port to PLT Scheme v200, and here is the copyright notice,
+-;;; comments, and licence from the original source:
+-;;;
+-;;; Copyright (C) Andy Gaynor (1999-2003)
+-;;;
+-;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
+-;;; Renamed to s:let by Eli Barzilay
+-
+-#lang scheme/base
+-(provide s:let)
+-
+-(define-syntax s:let
+- (syntax-rules ()
+- ;; standard
+- ((s:let () body ...)
+- (let () body ...))
+- ((s:let ((var val) ...) body ...)
+- (let ((var val) ...) body ...))
+-
+- ;; rest style
+- ((s:let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; signature style
+- ((s:let (name bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ;; standard named style
+- ((s:let name (bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ))
+-
+-;; A loop to walk down the list of bindings.
+-
+-(define-syntax let-loop
+- (syntax-rules ()
+-
+- ;; No more bindings - make a LETREC.
+- ((let-loop name () (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ...) . body)))
+- name)
+- vals ...))
+-
+- ;; Rest binding, no name
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (let ((var val) ... (rest-var (list rest-val ...))) . body))
+-
+- ;; Process a (var val) pair.
+- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
+- (let-loop name (more ...) (vars ... var) (vals ... val) body))
+-
+- ;; End with a rest variable - make a LETREC.
+- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ... . rest-var) . body)))
+- name)
+- vals ... rest-vals ...))))
+-
+-;; Four loops - normal and `signature-style', each with and without a rest
+-;; binding.
+-;;
+-;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let fibonacci ((n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,2 +0,0 @@
+-;; module loader for SRFI-5
+-#lang s-exp srfi/provider srfi/5/let #:unprefix s:
diff --git a/racket.spec b/racket.spec
index d3dc907..5fb06c9 100644
--- a/racket.spec
+++ b/racket.spec
@@ -1,38 +1,24 @@
Name: racket
-Version: 6.12
-Release: 8%{?dist}
+Version: 7.0
+Release: 1%{?dist}
Summary: General purpose programming language
License: GPLv3 and LGPLv3 and MIT
URL: https://racket-lang.org
Source0: https://mirror.racket-lang.org/installers/%{version}/%{name}-%{version}-src…
-# Update SRFI libraries to include upstream PR 5.
-# See: https://github.com/racket/srfi/pull/5 (merged)
-Patch0: racket-6.12-update-srfi.patch
-
# Remove SRFI library and docs with restrictive licensing.
# See: https://github.com/racket/srfi/issues/4 (open)
# Note: Upstream maintainers have confirmed this
# is safe, since the removed components are
# extra elements which nothing else in the
# package depends on.
-Patch1: racket-6.12-remove-nonfree.patch
-
-# Backport upstream fix of rpaths in racket/racket
-# which break check-buildroot
-# https://github.com/racket/racket/pull/1947 (merged)
-Patch2: racket-6.12-fix-rpaths.patch
-
-# Backport upstream fix of rpaths in racket/web-server
-# which break check-buildroot
-# https://github.com/racket/web-server/issues/36 (closed)
-Patch3: racket-6.12-fix-webserver-rpaths.patch
+Patch0: racket-7.0-remove-nonfree.patch
# Update racket-doc/search to use https://docs.racket-lang.org
# when no docs exist locally
# https://github.com/racket/scribble/pull/164 (open)
-Patch4: racket-6.12-fix-doc-open-url.patch
+Patch1: racket-7.0-fix-doc-open-url.patch
# Issue Building for s390x and armv7hl in koji
ExcludeArch: s390x armv7hl
@@ -131,6 +117,7 @@ for i in $DOCS_TO_FIX; do
${RPM_BUILD_ROOT}/%{_datadir}/doc/racket/$i
done
+# Remove the executable bit on legacy template file
chmod -x ${RPM_BUILD_ROOT}%{_libdir}/racket/starter-sh
%ldconfig_scriptlets
@@ -244,6 +231,10 @@ A local installation of the Racket documentation system.
%{_datadir}/doc/racket
%changelog
+* Fri Aug 3 2018 David Benoit <dbenoit(a)redhat.com> 7.0-1
+- Remove 6.12 patches and add update remove nonfree
+ srfi patch to 7.0
+
* Mon Jul 30 2018 David Benoit <dbenoit(a)redhat.com> 6.12-8
- Annotate dependencies with links to source code
- Move dependencies to racket-pkgs, since they are only used
diff --git a/sources b/sources
index 0fa455a..e43e6a3 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (racket-6.12-src.tgz) = e373cc1dd94f98d84dd2a105dad1cc2f2068118777b455af3f30ea3d190e6d601392bf17337652eeb84fa86d9c60aaa17563599b1772dc312211a8ac6e2c7060
+SHA512 (racket-7.0-src.tgz) = 98be3a7f0fbe38ae5febd2a9a3c81baede7dd7b3f3a9ed3ce68321389a68f74c6a8e24170013f7ca4fd725a79a3d5cfbda2e26f0f75163076e3b98f43094507f
commit 1b0a0799af2a5acee95845192bd3a772fd4a0f55
Author: dbenoit <dbenoit(a)redhat.com>
Date: Fri Aug 31 14:54:24 2018 -0400
fix sources file
diff --git a/.gitignore b/.gitignore
index e69de29..d013960 100644
--- a/.gitignore
+++ b/.gitignore
@@ -0,0 +1 @@
+/racket-6.12-src.tgz
diff --git a/sources b/sources
index 452da4f..0fa455a 100644
--- a/sources
+++ b/sources
@@ -1,6 +1 @@
-SHA512 (racket-6.12-fix-doc-open-url.patch) = 66a6a7f0f005a0da7acaf1ffbd53831eb796171994aa033bae5a2a973a79bb1c65212eaf4bf7ec02dfae5a186efcfa3e2a06b08bc31ff162cd4e08ab5b97bb94
-SHA512 (racket-6.12-fix-rpaths.patch) = d844d4d174b1cd92c61d16912aaf48afddbc9010acb2e5950d2d277d3110fc9256b68cbc38f06f69f6e2009f34085b32690a535049934d46d820e5b1491434e1
-SHA512 (racket-6.12-fix-webserver-rpaths.patch) = 807e356726dff3087bbcf505835082bcd2f0cdfb1d45e899d98bdf9b23715586b5c2a4a5703fff9ff2af9e46feb9936eaf685b6fb331388cd77dcc57389fbd76
-SHA512 (racket-6.12-remove-nonfree.patch) = 76e24e03728742073bdf4dc927da734195ddb6400cbcdaedecc09d94e7eac4168de4c9d30423bfd293b5fcd3580b0147b38c9061bded1205ec91e24f4868be2c
-SHA512 (racket-6.12-update-srfi.patch) = cfa97cc312ec3a40ad3dd4c4d662c949274325d6acc991c0d9b2327fc03b6c532cec0aa6e7f9a5563d0c23526742177367264f601f8e92055a1d6c3347e763a4
-SHA512 (racket.spec) = 5c53ad1906aa228848f7cabd72f377dd6da8f711669af515650d35fb40b5d3cb627c0b3daaf347cd2066ed250ce7fbcced7192e24f1aea7509eba1ceee487f49
+SHA512 (racket-6.12-src.tgz) = e373cc1dd94f98d84dd2a105dad1cc2f2068118777b455af3f30ea3d190e6d601392bf17337652eeb84fa86d9c60aaa17563599b1772dc312211a8ac6e2c7060
commit 335d72219631d93c3f29c29fd82070c0af593ef4
Author: dbenoit <dbenoit(a)redhat.com>
Date: Fri Aug 31 14:44:26 2018 -0400
fix gitignore
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..e69de29
diff --git a/sources b/sources
new file mode 100644
index 0000000..452da4f
--- /dev/null
+++ b/sources
@@ -0,0 +1,6 @@
+SHA512 (racket-6.12-fix-doc-open-url.patch) = 66a6a7f0f005a0da7acaf1ffbd53831eb796171994aa033bae5a2a973a79bb1c65212eaf4bf7ec02dfae5a186efcfa3e2a06b08bc31ff162cd4e08ab5b97bb94
+SHA512 (racket-6.12-fix-rpaths.patch) = d844d4d174b1cd92c61d16912aaf48afddbc9010acb2e5950d2d277d3110fc9256b68cbc38f06f69f6e2009f34085b32690a535049934d46d820e5b1491434e1
+SHA512 (racket-6.12-fix-webserver-rpaths.patch) = 807e356726dff3087bbcf505835082bcd2f0cdfb1d45e899d98bdf9b23715586b5c2a4a5703fff9ff2af9e46feb9936eaf685b6fb331388cd77dcc57389fbd76
+SHA512 (racket-6.12-remove-nonfree.patch) = 76e24e03728742073bdf4dc927da734195ddb6400cbcdaedecc09d94e7eac4168de4c9d30423bfd293b5fcd3580b0147b38c9061bded1205ec91e24f4868be2c
+SHA512 (racket-6.12-update-srfi.patch) = cfa97cc312ec3a40ad3dd4c4d662c949274325d6acc991c0d9b2327fc03b6c532cec0aa6e7f9a5563d0c23526742177367264f601f8e92055a1d6c3347e763a4
+SHA512 (racket.spec) = 5c53ad1906aa228848f7cabd72f377dd6da8f711669af515650d35fb40b5d3cb627c0b3daaf347cd2066ed250ce7fbcced7192e24f1aea7509eba1ceee487f49
commit f241dafce908418595fbb717060cbbefb6867e61
Author: dbenoit <dbenoit(a)redhat.com>
Date: Fri Aug 31 14:25:40 2018 -0400
add spec and patches
diff --git a/racket-6.12-fix-doc-open-url.patch b/racket-6.12-fix-doc-open-url.patch
new file mode 100644
index 0000000..7c497f5
--- /dev/null
+++ b/racket-6.12-fix-doc-open-url.patch
@@ -0,0 +1,69 @@
+diff -urN racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt racket-6.12/share/pkgs/scribble-lib/help/search.rkt
+--- racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:14:56.017273013 -0400
++++ racket-6.12/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:15:37.839113929 -0400
+@@ -17,33 +17,38 @@
+ #:notify [notify void])
+ (define open-url (get-doc-open-url))
+ (cond
+- [open-url
+- (define dest-url (let ([u (string->url open-url)])
+- (combine-url/relative
+- u
+- (string-join
+- (for/list ([s (explode-path sub)])
+- (if (path? s)
+- (path-element->string s)
+- (format "~a" s)))
+- "/"))))
+- (notify (url->string dest-url))
+- (send-url (url->string
+- (struct-copy url dest-url
+- [fragment (or fragment
+- (url-fragment dest-url))]
+- [query (append
+- (url-query dest-url)
+- (if query
+- (url-query
+- (string->url
+- (format "q?~a" query)))
+- null))])))]
+- [else
+- (let* ([path (build-path (find-user-doc-dir) sub)]
+- [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
+- (notify path)
+- (send-url/file path #:fragment fragment #:query query))]))
++ [open-url
++ (define dest-url (let ([u (string->url open-url)])
++ (combine-url/relative
++ u
++ (string-join
++ (for/list ([s (explode-path sub)])
++ (if (path? s)
++ (path-element->string s)
++ (format "~a" s)))
++ "/"))))
++ (notify (url->string dest-url))
++ (send-url (url->string
++ (struct-copy url dest-url
++ [fragment (or fragment
++ (url-fragment dest-url))]
++ [query (append
++ (url-query dest-url)
++ (if query
++ (url-query
++ (string->url
++ (format "q?~a" query)))
++ null))])))]
++ [else
++ (let* ([path (build-path (find-user-doc-dir) sub)]
++ [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
++ (notify path)
++ (if (file-exists? path)
++ (send-url/file path #:fragment fragment #:query query)
++ (let ([part (lambda (pfx x) (if x (string-append pfx x) ""))])
++ (send-url (string-append
++ "https://docs.racket-lang.org/"
++ sub (part "#" fragment) (part "?" query))))))]))
+
+ ;; This is an example of changing this code to use the online manuals.
+ ;; Normally, it's better to set `doc-open-url` in "etc/config.rktd",
diff --git a/racket-6.12-fix-rpaths.patch b/racket-6.12-fix-rpaths.patch
new file mode 100644
index 0000000..f76730e
--- /dev/null
+++ b/racket-6.12-fix-rpaths.patch
@@ -0,0 +1,13 @@
+diff -urN racket-6.12-f/collects/racket/runtime-path.rkt racket-6.12/collects/racket/runtime-path.rkt
+--- racket-6.12-f/collects/racket/runtime-path.rkt 2018-02-06 16:04:29.969878308 -0500
++++ racket-6.12/collects/racket/runtime-path.rkt 2018-02-06 16:07:10.549176282 -0500
+@@ -186,7 +186,8 @@
+ (path-of
+ #,(datum->syntax
+ #'orig-stx
+- `(,#'this-expression-source-file))))
++ `(,#'this-expression-source-file)
++ #'orig-stx)))
+ #'void)])
+ (apply to-values (resolve-paths (#%variable-reference)
+ get-dir
diff --git a/racket-6.12-fix-webserver-rpaths.patch b/racket-6.12-fix-webserver-rpaths.patch
new file mode 100644
index 0000000..78614cf
--- /dev/null
+++ b/racket-6.12-fix-webserver-rpaths.patch
@@ -0,0 +1,35 @@
+diff -urN racket-6.12-rp/share/pkgs/web-server-lib/web-server/formlets/lib.rkt racket-6.12-wbsvr/share/pkgs/web-server-lib/web-server/formlets/lib.rkt
+--- racket-6.12-rp/share/pkgs/web-server-lib/web-server/formlets/lib.rkt 2018-02-22 18:42:06.131455756 -0500
++++ racket-6.12-wbsvr/share/pkgs/web-server-lib/web-server/formlets/lib.rkt 2018-03-20 15:07:09.666222028 -0400
+@@ -7,6 +7,7 @@
+ racket/function
+ racket/serialize
+ syntax/location
++ setup/collects
+ (for-syntax racket/base
+ syntax/parse))
+
+@@ -70,20 +71,20 @@
+ #:range-contracts (map (curry coerce-contract 'formlet/c)
+ contracts)))]))
+ (define quote-this-module-path
+- (quote-module-path))
++ (path->collects-relative (quote-module-path)))
+ (define-syntax formlet/c
+ (syntax-parser
+ [(_ range ...)
+ #:declare range (expr/c #'contract?
+ #:name "range contract argument")
+ #'(formlet/c** (-> listof-binding
+- (values (coerce-contract 'formlet/c range.c) ...)))]
++ (values (coerce-contract 'formlet/c range #;.c) ...)))]
+ [name:id
+ #`(contract
+ (-> contract? (... ...) contract?)
+ dynamic-formlet/c
+ quote-this-module-path
+- (quote-module-path)
++ (path->collects-relative (quote-module-path))
+ "formlet/c"
+ #'name)]))
+
diff --git a/racket-6.12-remove-nonfree.patch b/racket-6.12-remove-nonfree.patch
new file mode 100644
index 0000000..4e72a3e
--- /dev/null
+++ b/racket-6.12-remove-nonfree.patch
@@ -0,0 +1,1124 @@
+diff -urN racket-6.12-nf/share/links.rktd racket-6.12/share/links.rktd
+--- racket-6.12-nf/share/links.rktd 2018-01-31 13:04:06.757393819 -0500
++++ racket-6.12/share/links.rktd 2018-01-31 13:04:25.419309681 -0500
+@@ -133,7 +133,6 @@
+ (root "pkgs/r5rs-doc")
+ (root "pkgs/srfi-lite-lib")
+ (root "pkgs/srfi-lib")
+- (root "pkgs/srfi-lib-nonfree")
+ (root "pkgs/html-doc")
+ (root "pkgs/images-gui-lib")
+ (root "pkgs/images-doc")
+@@ -177,7 +176,6 @@
+ (root "pkgs/slideshow-exe")
+ (root "pkgs/slideshow-plugin")
+ (root "pkgs/srfi-doc")
+- (root "pkgs/srfi-doc-nonfree")
+ (root "pkgs/syntax-color-doc")
+ (root "pkgs/web-server-lib")
+ (root "pkgs/unix-socket-lib")
+diff -urN racket-6.12-nf/share/pkgs/srfi/info.rkt racket-6.12/share/pkgs/srfi/info.rkt
+--- racket-6.12-nf/share/pkgs/srfi/info.rkt 2018-01-31 13:04:07.216391749 -0500
++++ racket-6.12/share/pkgs/srfi/info.rkt 2018-01-31 13:05:32.601025227 -0500
+@@ -3,13 +3,9 @@
+ (define collection 'multi)
+
+ (define deps '("srfi-lib"
+- "srfi-lib-nonfree"
+- "srfi-doc"
+- "srfi-doc-nonfree"))
++ "srfi-doc"))
+ (define implies '("srfi-lib"
+- "srfi-lib-nonfree"
+- "srfi-doc"
+- "srfi-foc-nonfree"))
++ "srfi-doc"))
+
+ (define pkg-desc "Legacy SRFI (Scheme) libraries")
+
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/info.rkt 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,19 +0,0 @@
+-#lang info
+-
+-(define collection 'multi)
+-
+-(define build-deps '("mzscheme-doc"
+- "scheme-lib"
+- "base"
+- "scribble-lib"
+- "srfi-doc"
+- "srfi-lib-nonfree"
+- "racket-doc"
+- "r5rs-doc"
+- "r6rs-doc"
+- "compatibility-lib"))
+-(define update-implies '("srfi-lib-nonfree"))
+-
+-(define pkg-desc "documentation part of \"srfi nonfree\"")
+-
+-(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang info
+-
+-(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
+@@ -1,66 +0,0 @@
+-#lang scribble/doc
+-@(require srfi/scribblings/util
+- scribble/manual
+- scribble/eval
+- scriblib/render-cond
+- scribble/core
+- scribble/html-properties
+- (for-syntax scheme/base)
+- (for-label scheme/base
+- racket/stream))
+-
+-@; ----------------------------------------------------------------------
+-
+-@title{SRFI Nonfree Libraries and Documentation}
+-
+-The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
+-Implementation} (a.k.a. @deftech{SRFI}) process allows individual
+-members of the Scheme community to propose libraries and extensions to
+-be supported by multiple Scheme implementations.
+-
+-Racket is distributed with implementations of many SRFIs, most of
+-which can be implemented as libraries. To import the bindings of SRFI
+-@math{n}, use
+-
+-@racketblock[
+-(require @#,elem{@racketidfont{srfi/}@math{n}})
+-]
+-
+-This document lists the SRFIs that are supported by Racket and
+-provides a link to the original SRFI specification (which is also
+-distributed as part of Racket's documentation).
+-
+-The following SRFI specification documents are licensed restrictively.
+-
+-@table-of-contents[]
+-
+-
+-@; ----------------------------------------
+-
+-@srfi[5]{A compatible let form with signatures and rest arguments}
+-
+-@redirect[5 '(
+- (let #t "unnamed")
+-)]
+-
+-Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
+-
+-@; ----------------------------------------
+-
+-@srfi[29]{Localization}
+-
+-@redirect[29 '(
+- (current-language #f "current-language")
+- (current-country #f "current-country")
+- (current-locale-details #f "current-locale-details")
+- (declare-bundle! #f "declare-bundle!")
+- (store-bundle #f "store-bundle")
+- (load-bundle! #f "load-bundle!")
+- (localized-template #f "localized-template")
+-)]
+-
+-Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
+-
+-@; ----------------------------------------
+-
+-@index-section[]
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,507 +0,0 @@
+-<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
+-<html>
+- <head>
+- <meta name="generator" content="HTML Tidy, see www.w3.org">
+- <title>SRFI 29: Localization</title>
+- <meta name="author" content="Scott G. Miller">
+- <meta name="description" content="Localization">
+- </head>
+- <body>
+- <H1>Title</H1>
+-
+- SRFI 29: Localization
+-
+- <H1>Author</H1>
+-
+- Scott G. Miller
+-
+- <H1>Abstract</H1>
+-
+- This document specifies an interface to retrieving and
+- displaying locale sensitive messages. A Scheme program can
+- register one or more translations of templated messages, and
+- then write Scheme code that can transparently retrieve the
+- appropriate message for the locale under which the Scheme
+- system is running. <br>
+-
+-
+- <H1>Rationale</H1>
+-
+- <p>As any programmer that has ever had to deal with making his
+- or her code readable in more than one locale, the process of
+- sufficiently abstracting program messages from their
+- presentation to the user is non-trivial without help from the
+- programming language. Most modern programming language
+- libraries do provide some mechanism for performing this
+- separation.</p>
+-
+- <p>A portable API that allows a piece of code to run without
+- modification in different countries and under different
+- languages is a must for any non-trivial software project.
+- The interface should separate the logic of a program from
+- the myriad of translations that may be necessary.</p>
+-
+- <p>The interface described in this document provides such
+- functionality. The underlying implementation is also allowed to
+- use whatever datastructures it likes to provide access to the
+- translations in the most efficient manner possible. In
+- addition, the implementation is provided with standardized
+- functions that programs will use for accessing an external,
+- unspecified repository of translations.</p>
+-
+- <p>This interface <i>does not</i> cover all aspects of
+- localization, including support for non-latin characters,
+- number and date formatting, etc. Such functionality is the
+- scope of a future SRFI that may extend this one.</p>
+-
+- <H1>Dependencies</H1>
+-
+- An SRFI-29 conformant implementation must also implement
+- SRFI-28, Basic Format Strings. Message templates are strings
+- that must be processed by the <tt>format</tt> function
+- specified in that SRFI.
+-
+- <H1>Specification</H1>
+-
+- <h3>Message Bundles</h3>
+-
+- <p>A Message Bundle is a set of message templates and their
+- identifying keys. Each bundle contains one or more such
+- key/value pairs. The bundle itself is associated with a
+- <i>bundle specifier</i> which uniquely identifies the
+- bundle.</p>
+-
+- <h3>Bundle Specifiers</h3>
+-
+- <p>A Bundle Specifier is a Scheme list that describes, in order
+- of importance, the package and locale of a message bundle.
+- In most cases, a locale specifier will have between one
+- and three elements. The first element is a symbol denoting the
+- package for which this bundle applies. The second and third
+- elements denote a <i>locale</i>. The second element (first
+- element of the locale) if present, is the two letter, ISO 639-1
+- language code for the bundle. The third element, if present, is
+- a two letter ISO 3166-1 country code. In some cases, a
+- fourth element may be present, specifying the encoding used for
+- the bundle. All bundle specifier elements are Scheme
+- symbols.</p>
+-
+- <p>If only one translation is provided, it should be designated
+- only by a package name, for example <tt>(mathlib)</tt>. This
+- translation is called the <i>default</i> translation.</p>
+-
+- <h3>Bundle Searching</h3>
+-
+- <p>When a message template is retrieved from a bundle, the
+- Scheme implementation will provide the locale under which the
+- system is currently running. When the template is retrieved,
+- the package name will be specified. The Scheme system should
+- construct a Bundle Specifier from the provided package name and
+- the active locale. For example, when retrieving a message
+- template for French Canadian, in the <tt>mathlib</tt> package,
+- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
+- program may also retrieve the elements of the current locale
+- using the no-argument procedures:</p>
+-
+- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-language</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- When given no arguments, returns the current ISO 639-1
+- language code as a symbol. If provided with an
+- argument, the current language is set to that named by the
+- symbol for the currently executing Scheme thread (or for the
+- entire Scheme system if such a distinction is not possible).
+-
+- </blockquote>
+-
+- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-country</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- returns the current ISO 3166-1 country code as a symbol.
+- If provided with an argument, the current country is
+- set to that named by the symbol for the currently executing
+- Scheme thread (or for the entire Scheme system if such a
+- distinction is not possible).
+- </blockquote>
+-
+- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
+- symbol</i></tt>s<br>
+- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- Returns a list of additional locale details as a list of
+- symbols. This list may contain information about
+- encodings or other more specific information. If
+- provided with an argument, the current locale details are set
+- to those given in the currently executing Scheme thread (or
+- for the entire Scheme system if such a distinction is not
+- possible).
+- </blockquote>
+-
+- <p>The Scheme System should first check for a bundle with the
+- exact name provided. If no such bundle is found, the last
+- element from the list is removed and a search is tried for a
+- bundle with that name. If no bundle is then found, the list is
+- shortened by removing the last element again. If no message is
+- found and the bundle specifier is now the empty list, an error
+- should be raised.</p>
+-
+- <p>The reason for this search order is to provide the most
+- locale sensitive template possible, but to fall back on more
+- general templates if a translation has not yet been provided
+- for the given locale.</p>
+-
+- <h3>Message Templates</h3>
+-
+- <p>A message template is a localized message that may or may
+- not contain one of a number of formatting codes. A message
+- template is a Scheme string. The string is of a form that can
+- be processed by the <tt>format</tt> procedure found in many
+- Scheme systems and formally specified in SRFI-28 (Basic Format
+- Strings).</p>
+-
+- <p>This SRFI also extends SRFI-28 to provide an additional
+- <tt>format</tt> escape code:</p>
+-
+- <blockquote>
+- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
+- follows this code immediately to reference the [N]'th
+- optional value absolutely, rather than the next unconsumed
+- value. The referenced value is <i>not</i> consumed.
+- </blockquote>
+- This extension allows optional values to be positionally
+- referenced, so that message templates can be constructed that
+- can produce the proper word ordering for a language.
+-
+- <h3>Preparing Bundles</h3>
+- Before a bundle may be used by the Scheme system to retrieve
+- localized template messages, they must be made available to the
+- Scheme system. This SRFI specifies a way to portably
+- define the bundles, as well as store them in and retrieve them
+- from an unspecified system which may be provided by resources
+- outside the Scheme system.<br>
+-
+-
+- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
+- association-list</i> -> undefined<br>
+- </tt></p>
+-
+- <blockquote>
+- Declares a new bundle named by the given bundle-specifier.
+- The contents of the bundle are defined by the provided
+- association list. The list contains associations
+- between Scheme symbols and the message templates (Scheme
+- strings) they name. If a bundle already exists with the
+- given name, it is overwritten with the newly declared
+- bundle.<br>
+- </blockquote>
+- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to store a bundle named by the given bundle
+- specifier, and previously made available using
+- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
+- unspecified mechanism that may be persistent across Scheme
+- system restarts. If successful, a non-false value is
+- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
+- </blockquote>
+- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to retrieve a bundle from an unspecified mechanism
+- which stores bundles outside the Scheme system. If the
+- bundle was retrieved successfully, the function returns a
+- non-false value, and the bundle is immediately available to
+- the Scheme system. If the bundle could not be found or loaded
+- successfully, the function returns <tt>#f</tt>, and the
+- Scheme system's bundle registry remains unaffected.<br>
+- </blockquote>
+- A compliant Scheme system may choose not to provide any
+- external mechanism to store localized bundles. If it does
+- not, it must still provide implementations for
+- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
+- case, both functions must return <tt>#f</tt> regardless of the
+- arguments given. Users of this SRFI should recognize that the
+- inability to load or store a localized bundle in an external
+- repository is <i>not</i> a fatal error.<br>
+-
+-
+- <h3>Retrieving Localized Message Templates</h3>
+-
+- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
+- message-template-name</tt></i> <tt>-> <i>string or #f<br>
+- </i></tt></p>
+-
+- <blockquote>
+- Retrieves a localized message template for the given package
+- name and the given message template name (both symbols).
+- If no such message could be found, false (#f) is
+- returned.<br>
+- <br>
+- </blockquote>
+- After retrieving a template, the calling program can use
+- <tt>format</tt> to produce a string that can be displayed to
+- the user.<br>
+-
+-
+- <h2>Examples</h2>
+- The below example makes use of SRFI-29 to display simple,
+- localized messages. It also defines its bundles in such a
+- way that the Scheme system may store and retrieve the bundles
+- from a more efficient system catalog, if available.<br>
+-
+-<pre>
+-(let ((translations
+- '(((en) . ((time . "Its ~a, ~a.")
+- (goodbye . "Goodbye, ~a.")))
+- ((fr) . ((time . "~1@*~a, c'est ~a.")
+- (goodbye . "Au revoir, ~a."))))))
+- (for-each (lambda (translation)
+- (let ((bundle-name (cons 'hello-program (car translation))))
+- (if (not (load-bundle! bundle-name))
+- (begin
+- (declare-bundle! bundle-name (cdr translation))
+- (store-bundle! bundle-name)))))
+- translations))
+-
+-(define localized-message
+- (lambda (message-name . args)
+- (apply format (cons (localized-template 'hello-program
+- message-name)
+- args))))
+-
+-(let ((myname "Fred"))
+- (display (localized-message 'time "12:00" myname))
+- (display #\newline)
+-
+- (display (localized-message 'goodbye myname))
+- (display #\newline))
+-
+-;; Displays (English):
+-;; Its 12:00, Fred.
+-;; Goodbye, Fred.
+-;;
+-;; French:
+-;; Fred, c'est 12:00.
+-;; Au revoir, Fred.
+-</pre>
+-
+- <H1>Implementation</H1>
+-
+- <p>The implementation requires that the Scheme system provide a
+- definition for <tt>current-language</tt> and
+- <tt>current-country</tt> capable of distinguishing the correct
+- locale present during a Scheme session. The definitions of
+- those functions in the reference implementation are not capable
+- of that distinction. Their implementation is provided only so
+- that the following code can run in any R4RS scheme system.
+- <br>
+- </p>
+-
+- <p>In addition, the below implementation of a compliant
+- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
+- SRFI-23 (Error reporting)</p>
+-<pre>
+-;; The association list in which bundles will be stored
+-(define *localization-bundles* '())
+-
+-;; The current-language and current-country functions provided
+-;; here must be rewritten for each Scheme system to default to the
+-;; actual locale of the session
+-(define current-language
+- (let ((current-language-value 'en))
+- (lambda args
+- (if (null? args)
+- current-language-value
+- (set! current-language-value (car args))))))
+-
+-(define current-country
+- (let ((current-country-value 'us))
+- (lambda args
+- (if (null? args)
+- current-country-value
+- (set! current-country-value (car args))))))
+-
+-;; The load-bundle! and store-bundle! both return #f in this
+-;; reference implementation. A compliant implementation need
+-;; not rewrite these procedures.
+-(define load-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-(define store-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-;; Declare a bundle of templates with a given bundle specifier
+-(define declare-bundle!
+- (letrec ((remove-old-bundle
+- (lambda (specifier bundle)
+- (cond ((null? bundle) '())
+- ((equal? (caar bundle) specifier)
+- (cdr bundle))
+- (else (cons (car bundle)
+- (remove-old-bundle specifier
+- (cdr bundle))))))))
+- (lambda (bundle-specifier bundle-assoc-list)
+- (set! *localization-bundles*
+- (cons (cons bundle-specifier bundle-assoc-list)
+- (remove-old-bundle bundle-specifier
+- *localization-bundles*))))))
+-
+-;;Retrieve a localized template given its package name and a template name
+-(define localized-template
+- (letrec ((rdc
+- (lambda (ls)
+- (if (null? (cdr ls))
+- '()
+- (cons (car ls) (rdc (cdr ls))))))
+- (find-bundle
+- (lambda (specifier template-name)
+- (cond ((assoc specifier *localization-bundles*) =>
+- (lambda (bundle) bundle))
+- ((null? specifier) #f)
+- (else (find-bundle (rdc specifier)
+- template-name))))))
+- (lambda (package-name template-name)
+- (let loop ((specifier (cons package-name
+- (list (current-language)
+- (current-country)))))
+- (and (not (null? specifier))
+- (let ((bundle (find-bundle specifier template-name)))
+- (and bundle
+- (cond ((assq template-name bundle) => cdr)
+- ((null? (cdr specifier)) #f)
+- (else (loop (rdc specifier)))))))))))
+-
+-;;An SRFI-28 and SRFI-29 compliant version of format. It requires
+-;;SRFI-23 for error reporting.
+-(define format
+- (lambda (format-string . objects)
+- (let ((buffer (open-output-string)))
+- (let loop ((format-list (string->list format-string))
+- (objects objects)
+- (object-override #f))
+- (cond ((null? format-list) (get-output-string buffer))
+- ((char=? (car format-list) #\~)
+- (cond ((null? (cdr format-list))
+- (error 'format "Incomplete escape sequence"))
+- ((char-numeric? (cadr format-list))
+- (let posloop ((fl (cddr format-list))
+- (pos (string->number
+- (string (cadr format-list)))))
+- (cond ((null? fl)
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (null? (cdr fl)))
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (eq? (cadr fl) '#\*))
+- (loop (cddr fl) objects (list-ref objects pos)))
+- (else
+- (posloop (cdr fl)
+- (+ (* 10 pos)
+- (string->number
+- (string (car fl)))))))))
+- (else
+- (case (cadr format-list)
+- ((#\a)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (display (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\s)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (write (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\%)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\newline buffer)
+- (loop (cddr format-list) objects #f))
+- ((#\~)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\~ buffer)
+- (loop (cddr format-list) objects #f))
+- (else
+- (error 'format "Unrecognized escape sequence"))))))
+- (else (display (car format-list) buffer)
+- (loop (cdr format-list) objects #f)))))))
+-
+-</pre>
+-
+- <H1>Copyright</H1>
+-
+- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
+-
+- <p>This document and translations of it may be copied and
+- furnished to others, and derivative works that comment on or
+- otherwise explain it or assist in its implementation may be
+- prepared, copied, published and distributed, in whole or in
+- part, without restriction of any kind, provided that the above
+- copyright notice and this paragraph are included on all such
+- copies and derivative works. However, this document itself may
+- not be modified in any way, such as by removing the copyright
+- notice or references to the Scheme Request For Implementation
+- process or editors, except as needed for the purpose of
+- developing SRFIs in which case the procedures for copyrights
+- defined in the SRFI process must be followed, or as required to
+- translate it into languages other than English.</p>
+-
+- <p>The limited permissions granted above are perpetual and will
+- not be revoked by the authors or their successors or
+- assigns.</p>
+-
+- <p>This document and the information contained herein is
+- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
+- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
+- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
+- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
+- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
+- <hr>
+-
+- <address>
+- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
+- Rush</a>
+- </address>
+-
+- <address>
+- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
+- Miller</a>
+- </address>
+- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
+- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
+- Daylight Time 2002 <!-- hhmts end --> <br>
+- </body>
+-</html>
+-
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,345 +0,0 @@
+-<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
+-<html>
+-<head>
+- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
+-</head>
+-<body>
+-
+-<H1>Title</H1>
+-
+-SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
+-
+-<H1>Author</H1>
+-
+-Andy Gaynor
+-
+-<H1>Status</H1>
+-
+-This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
+-You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
+-<P><UL>
+-<LI>Received: 1999/2/2
+-<LI>Draft: 1999/2/10-1999/04/12
+-<LI>Final: 1999/4/26
+-<LI>Revised reference implementation: 2003/01/27
+-</UL>
+-
+-<H1>Abstract</H1>
+-
+-The <i>named-let</i> incarnation of the <code>let</code> form has two slight
+-inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
+-form makes no accommodation for rest arguments, an issue of functionality
+-and consistency. As defined, the <code>let</code> form does not accommodate
+-signature-style syntax, an issue of aesthetics and consistency. Both
+-issues are addressed here in a manner which is compatible with the traditional
+-<code>let</code> form but for minor extensions.
+-
+-<H1>Rationale</H1>
+-
+-<H2>Signature-style Syntax</H2>
+-
+-Consider the following two equivalent definitions:
+-
+-<p><pre>
+-(define fibonacci
+- (lambda (n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
+-
+-(define (fibonacci n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-Although there is a named-let analog for the former form, there is none
+-for the latter. To wit, suppose one wished to compute the 10th element
+-of the Fibonacci sequence using a named let:
+-
+-<p>
+-<pre>
+-(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-Values: 55
+-</pre>
+-
+-As it stands, one cannot equivalently write
+-
+-<p>
+-<pre>
+-(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-which is consistent with <code>define</code>'s signature-style form.
+-<p>Those that favor the signature style may prefer this extension.
+-In any case, it may be more appropriate to include all bound names within
+-the binding section. As presented, this straightforward extension
+-introduces no ambiguity or incompatibility with the existing definition
+-of let.
+-
+-<H2>Rest Arguments</H2>
+-
+-As it stands, one cannot write a named let with rest arguments, as in
+-
+-<p>
+-<pre>
+-(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))
+-</pre>
+-
+-otherwise equivalent to
+-
+-<p>
+-<pre>
+-(letrec ((blast (lambda (port . x)
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))))
+- (blast (current-output-port) (+ 1 2) 4 5))
+-</pre>
+-
+-While this example is rather contrived, the functionality is not.
+-There are several times when the author has used this construct in practice.
+-Regardless, there is little reason to deny the <code>let</code> form access to
+-all the features of lambda functionality.
+-
+-<H2>Symbols in Binding Sections</H2>
+-
+-Both the features above rely upon the placement of symbols in <code>let</code>
+-binding lists (this statement is intentially simplistic). The only
+-other apparent use of such symbol placement is to tersely bind variables
+-to unspecified values. For example, one might desire to use
+-<code>(let (foo bar baz) ...)</code>
+-to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
+-unspecified values.
+-
+-<p>This usage is considered less important in light of the rationales
+-presented above, and an alternate syntax is immediately apparent, as
+-in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
+-be preferable, consistently parenthesizing normal binding clauses.
+-
+-<H1>Specification</H1>
+-
+-<H2>Syntax</H2>
+-
+-<p>
+-A formal specification of the syntax follows. Below, body, expression,
+-and identifier are free. Each instantiation of binding-name must be
+-unique.
+-</p>
+-
+-<p>
+-<pre>
+- let = "(" "let" let-bindings body ")"
+- expressions = nothing | expression expressions
+- let-bindings = let-name bindings
+- |"(" let-name "." bindings ")"
+- let-name = identifier
+- bindings = "(" ")"
+- | rest-binding
+- | "(" normal-bindings ["." rest-binding] ")"
+-normal-bindings = nothing
+- | normal-binding normal-bindings
+-normal-binding = "(" binding-name expression ")"
+- binding-name = identifier
+- rest-binding = "(" binding-name expressions ")"
+-</pre>
+-
+-<p>
+-For clarity and convenience, an informal specification follows.
+-</p>
+-
+-<ol>
+-<li><a name="unnamed">Unnamed</a>
+-
+-<p><pre>
+-(let ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig">
+-Named, non-signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-sig">Named, signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-
+-<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-</li>
+-</ol>
+-
+-<H2>Semantics</H2>
+-
+-Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
+-and <code>letrec</code> forms, respectively.
+-
+-<ul>
+-<li>For informal syntax 1:
+-
+-<p><pre>
+-(($lambda (<parameter>...) <body>...) <argument>...)
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 2 and 3:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...) <body>...)))
+- (<name> <argument>...))
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 4 and 5:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...
+-
+-. <rest-parameter>) <body>...)))
+- (<name> <argument>... <rest-argument>...))
+-</pre>
+-</li>
+-</ul>
+-
+-<H1>Implementation</H1>
+-
+-Here is an implementation using <code>SYNTAX-RULES</code>.
+-
+-<p>
+-<pre>
+-;; Use your own standard let.
+-;; Or call a lambda.
+-;; (define-syntax standard-let
+-;;
+-;; (syntax-rules ()
+-;;
+-;; ((let ((var val) ...) body ...)
+-;; ((lambda (var ...) body ...) val ...))))
+-
+-(define-syntax let
+-
+- (syntax-rules ()
+-
+- ;; No bindings: use standard-let.
+- ((let () body ...)
+- (standard-let () body ...))
+- ;; Or call a lambda.
+- ;; ((lambda () body ...))
+-
+- ;; All standard bindings: use standard-let.
+- ((let ((var val) ...) body ...)
+- (standard-let ((var val) ...) body ...))
+- ;; Or call a lambda.
+- ;; ((lambda (var ...) body ...) val ...)
+-
+- ;; One standard binding: loop.
+- ;; The all-standard-bindings clause didn't match,
+- ;; so there must be a rest binding.
+- ((let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; Signature-style name: loop.
+- ((let (name binding ...) body ...)
+- (let-loop name (binding ...) () () (body ...)))
+-
+- ;; defun-style name: loop.
+- ((let name bindings body ...)
+- (let-loop name bindings () () (body ...)))))
+-
+-(define-syntax let-loop
+-
+- (syntax-rules ()
+-
+- ;; Standard binding: destructure and loop.
+- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
+- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
+-
+- ;; Rest binding, no name: use standard-let, listing the rest values.
+- ;; Because of let's first clause, there is no "no bindings, no name" clause.
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
+- ;; Or call a lambda with a rest parameter on all values.
+- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
+- ;; Or use one of several other reasonable alternatives.
+-
+- ;; No bindings, name: call a letrec'ed lambda.
+- ((let-loop name () (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ...) . body)))
+- name)
+- val ...))
+-
+- ;; Rest binding, name: call a letrec'ed lambda.
+- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ... . rest-var) . body)))
+- name)
+- val ... rest-val ...))))
+-</pre>
+-
+-
+-<H1>Copyright</H1>
+-
+-Copyright (C) Andy Gaynor (1999). All Rights Reserved.
+-<p>This document and translations of it may be copied and furnished to
+-others, and derivative works that comment on or otherwise explain it or
+-assist in its implementation may be prepared, copied, published and distributed,
+-in whole or in part, without restriction of any kind, provided that the
+-above copyright notice and this paragraph are included on all such copies
+-and derivative works. However, this document itself may not be modified
+-in any way, such as by removing the copyright notice or references to the
+-Scheme Request For Implementation process or editors, except as needed
+-for the purpose of developing SRFIs in which case the procedures for copyrights
+-defined in the SRFI process must be followed, or as required to translate
+-it into languages other than English.
+-<p>The limited permissions granted above are perpetual and will not be
+-revoked by the authors or their successors or assigns.
+-<p>This document and the information contained herein is provided on an
+-"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
+-EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
+-USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
+-WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
+-
+- <hr>
+- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
+-
+-</body>
+-</html>
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/info.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/info.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,14 +0,0 @@
+-#lang info
+-
+-(define collection 'multi)
+-
+-(define deps '("scheme-lib"
+- "base"
+- "r6rs-lib"
+- "srfi-lib"
+- "compatibility-lib"))
+-
+-
+-(define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"")
+-
+-(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-#lang s-exp srfi/provider srfi/%3a5
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang s-exp srfi/provider srfi/5
+-
+-;; FIXME: "rest arguments" need to generate mutable lists
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,86 +0,0 @@
+-;;;
+-;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
+-;;; Time-stamp: <03/04/08 09:56:06 solsona>
+-;;;
+-;;; Usually, I would add a copyright notice, and the announce that
+-;;; this code is under the LGPL licence. Nevertheless, I only did the
+-;;; port to PLT Scheme v200, and here is the copyright notice,
+-;;; comments, and licence from the original source:
+-;;;
+-;;; Copyright (C) Andy Gaynor (1999-2003)
+-;;;
+-;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
+-;;; Renamed to s:let by Eli Barzilay
+-
+-#lang scheme/base
+-(provide s:let)
+-
+-(define-syntax s:let
+- (syntax-rules ()
+- ;; standard
+- ((s:let () body ...)
+- (let () body ...))
+- ((s:let ((var val) ...) body ...)
+- (let ((var val) ...) body ...))
+-
+- ;; rest style
+- ((s:let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; signature style
+- ((s:let (name bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ;; standard named style
+- ((s:let name (bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ))
+-
+-;; A loop to walk down the list of bindings.
+-
+-(define-syntax let-loop
+- (syntax-rules ()
+-
+- ;; No more bindings - make a LETREC.
+- ((let-loop name () (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ...) . body)))
+- name)
+- vals ...))
+-
+- ;; Rest binding, no name
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (let ((var val) ... (rest-var (list rest-val ...))) . body))
+-
+- ;; Process a (var val) pair.
+- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
+- (let-loop name (more ...) (vars ... var) (vals ... val) body))
+-
+- ;; End with a rest variable - make a LETREC.
+- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ... . rest-var) . body)))
+- name)
+- vals ... rest-vals ...))))
+-
+-;; Four loops - normal and `signature-style', each with and without a rest
+-;; binding.
+-;;
+-;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let fibonacci ((n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,2 +0,0 @@
+-;; module loader for SRFI-5
+-#lang s-exp srfi/provider srfi/5/let #:unprefix s:
diff --git a/racket-6.12-update-srfi.patch b/racket-6.12-update-srfi.patch
new file mode 100644
index 0000000..1894e32
--- /dev/null
+++ b/racket-6.12-update-srfi.patch
@@ -0,0 +1,4369 @@
+diff -urN racket-6.12-orig/share/links.rktd racket-6.12/share/links.rktd
+--- racket-6.12-orig/share/links.rktd 2018-01-26 16:12:13.000000000 -0500
++++ racket-6.12/share/links.rktd 2018-01-31 12:22:57.600504036 -0500
+@@ -133,6 +133,7 @@
+ (root "pkgs/r5rs-doc")
+ (root "pkgs/srfi-lite-lib")
+ (root "pkgs/srfi-lib")
++ (root "pkgs/srfi-lib-nonfree")
+ (root "pkgs/html-doc")
+ (root "pkgs/images-gui-lib")
+ (root "pkgs/images-doc")
+@@ -176,6 +177,7 @@
+ (root "pkgs/slideshow-exe")
+ (root "pkgs/slideshow-plugin")
+ (root "pkgs/srfi-doc")
++ (root "pkgs/srfi-doc-nonfree")
+ (root "pkgs/syntax-color-doc")
+ (root "pkgs/web-server-lib")
+ (root "pkgs/unix-socket-lib")
+diff -urN racket-6.12-orig/share/pkgs/srfi/info.rkt racket-6.12/share/pkgs/srfi/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi/info.rkt 2018-01-26 16:10:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi/info.rkt 2018-01-31 12:22:58.139502019 -0500
+@@ -1 +1,16 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("srfi-lib" "srfi-doc"))) (define implies (quote ("srfi-lib" "srfi-doc"))) (define pkg-desc "Legacy SRFI (Scheme) libraries") (define pkg-authors (quote (mflatt noel chongkai jay)))))
++#lang info
++
++(define collection 'multi)
++
++(define deps '("srfi-lib"
++ "srfi-lib-nonfree"
++ "srfi-doc"
++ "srfi-doc-nonfree"))
++(define implies '("srfi-lib"
++ "srfi-lib-nonfree"
++ "srfi-doc"
++ "srfi-foc-nonfree"))
++
++(define pkg-desc "Legacy SRFI (Scheme) libraries")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/info.rkt racket-6.12/share/pkgs/srfi-doc/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-doc/info.rkt 2018-01-26 16:10:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/info.rkt 2018-01-31 12:22:58.139502019 -0500
+@@ -1 +1,24 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define build-deps (quote ("mzscheme-doc" "scheme-lib" "base" "scribble-lib" "srfi-lib" "racket-doc" "r5rs-doc" "r6rs-doc" "compatibility-lib"))) (define update-implies (quote ("srfi-lib"))) (define pkg-desc "documentation part of \"srfi\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
++#lang info
++
++(define collection 'multi)
++
++(define build-deps '("mzscheme-doc"
++ "scheme-lib"
++ "base"
++ "scribble-lib"
++ "srfi-lib"
++ "racket-doc"
++ "r5rs-doc"
++ "r6rs-doc"
++ "compatibility-lib"))
++
++(define deps '("scheme-lib"
++ "base"
++ "scribble-lib"
++ "compatibility-lib"))
++
++(define update-implies '("srfi-lib"))
++
++(define pkg-desc "documentation part of \"srfi\"")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl
+--- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl 2018-01-31 12:22:58.144502000 -0500
+@@ -1,5 +1,6 @@
+ #lang scribble/doc
+-@(require scribble/manual
++@(require srfi/scribblings/util
++ scribble/manual
+ scribble/eval
+ scriblib/render-cond
+ scribble/core
+@@ -8,54 +9,6 @@
+ (for-label scheme/base
+ racket/stream))
+
+-@(define-syntax (srfi stx)
+- (syntax-case stx ()
+- [(_ num #:subdir subdir? . title)
+- (with-syntax ([srfi/n (string->symbol (format "srfi/~a" (syntax-e #'num)))])
+- #'(begin
+- (section #:tag (format "srfi-~a" num)
+- #:style 'unnumbered
+- (format "SRFI ~a: " num)
+- . title)
+- (defmodule srfi/n)
+- "Original specification: "
+- (let* ([label (format "SRFI ~a" num)]
+- [sub (if subdir? (format "srfi-~a/" num) "")]
+- [url ( (b) (format "~a/srfi-std/~asrfi-~a.html" b sub num))])
+- (cond-element
+- [(or latex text) @link[(url "http://docs.racket-lang.org") label]]
+- [else @link[(url ".") label]]))))]
+- [(_ num . title) #'(srfi num #:subdir #f . title)]))
+-
+-@;{ The `lst' argument is a list of
+- (list sym syntactic-form? html-anchor) }
+-@(define (redirect n lst #:subdir [subdir? #f])
+- (let ([file (if subdir?
+- (format "srfi-~a/srfi-~a.html" n n)
+- (format "srfi-~a.html" n))]
+- [mod-path (string->symbol (format "srfi/~a" n))])
+- (make-binding-redirect-elements mod-path
+- (map (lambda (b)
+- (list (car b) (cadr b)
+- (build-path "srfi-std" file)
+- (caddr b)))
+- lst))))
+-
+-@(define in-core
+- (case-lambda
+- [() (in-core ".")]
+- [(k) @elem{This SRFI's bindings are also available in
+- @racketmodname[racket/base]@|k|}]))
+-
+-@(begin
+- (define-syntax-rule (def-mz mz-if)
+- (begin
+- (require (for-label mzscheme))
+- (define mz-if (racket if))))
+- (def-mz mz-if))
+-
+-@(define srfi-std (style #f (list (install-resource "srfi-std"))))
+-
+ @; ----------------------------------------------------------------------
+
+ @title{SRFIs: Libraries}
+@@ -245,14 +198,6 @@
+
+ @; ----------------------------------------
+
+-@srfi[5]{A compatible let form with signatures and rest arguments}
+-
+-@redirect[5 '(
+- (let #t "unnamed")
+-)]
+-
+-@; ----------------------------------------
+-
+ @srfi[6]{Basic String Ports}
+
+ @redirect[6 '(
+@@ -671,20 +616,6 @@
+
+ @; ----------------------------------------
+
+-@srfi[29]{Localization}
+-
+-@redirect[29 '(
+- (current-language #f "current-language")
+- (current-country #f "current-country")
+- (current-locale-details #f "current-locale-details")
+- (declare-bundle! #f "declare-bundle!")
+- (store-bundle #f "store-bundle")
+- (load-bundle! #f "load-bundle!")
+- (localized-template #f "localized-template")
+-)]
+-
+-@; ----------------------------------------
+-
+ @srfi[30]{Nested Multi-line Comments}
+
+ This SRFI's syntax is part of Racket's default reader.
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html
+--- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,507 +0,0 @@
+-<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
+-<html>
+- <head>
+- <meta name="generator" content="HTML Tidy, see www.w3.org">
+- <title>SRFI 29: Localization</title>
+- <meta name="author" content="Scott G. Miller">
+- <meta name="description" content="Localization">
+- </head>
+- <body>
+- <H1>Title</H1>
+-
+- SRFI 29: Localization
+-
+- <H1>Author</H1>
+-
+- Scott G. Miller
+-
+- <H1>Abstract</H1>
+-
+- This document specifies an interface to retrieving and
+- displaying locale sensitive messages. A Scheme program can
+- register one or more translations of templated messages, and
+- then write Scheme code that can transparently retrieve the
+- appropriate message for the locale under which the Scheme
+- system is running. <br>
+-
+-
+- <H1>Rationale</H1>
+-
+- <p>As any programmer that has ever had to deal with making his
+- or her code readable in more than one locale, the process of
+- sufficiently abstracting program messages from their
+- presentation to the user is non-trivial without help from the
+- programming language. Most modern programming language
+- libraries do provide some mechanism for performing this
+- separation.</p>
+-
+- <p>A portable API that allows a piece of code to run without
+- modification in different countries and under different
+- languages is a must for any non-trivial software project.
+- The interface should separate the logic of a program from
+- the myriad of translations that may be necessary.</p>
+-
+- <p>The interface described in this document provides such
+- functionality. The underlying implementation is also allowed to
+- use whatever datastructures it likes to provide access to the
+- translations in the most efficient manner possible. In
+- addition, the implementation is provided with standardized
+- functions that programs will use for accessing an external,
+- unspecified repository of translations.</p>
+-
+- <p>This interface <i>does not</i> cover all aspects of
+- localization, including support for non-latin characters,
+- number and date formatting, etc. Such functionality is the
+- scope of a future SRFI that may extend this one.</p>
+-
+- <H1>Dependencies</H1>
+-
+- An SRFI-29 conformant implementation must also implement
+- SRFI-28, Basic Format Strings. Message templates are strings
+- that must be processed by the <tt>format</tt> function
+- specified in that SRFI.
+-
+- <H1>Specification</H1>
+-
+- <h3>Message Bundles</h3>
+-
+- <p>A Message Bundle is a set of message templates and their
+- identifying keys. Each bundle contains one or more such
+- key/value pairs. The bundle itself is associated with a
+- <i>bundle specifier</i> which uniquely identifies the
+- bundle.</p>
+-
+- <h3>Bundle Specifiers</h3>
+-
+- <p>A Bundle Specifier is a Scheme list that describes, in order
+- of importance, the package and locale of a message bundle.
+- In most cases, a locale specifier will have between one
+- and three elements. The first element is a symbol denoting the
+- package for which this bundle applies. The second and third
+- elements denote a <i>locale</i>. The second element (first
+- element of the locale) if present, is the two letter, ISO 639-1
+- language code for the bundle. The third element, if present, is
+- a two letter ISO 3166-1 country code. In some cases, a
+- fourth element may be present, specifying the encoding used for
+- the bundle. All bundle specifier elements are Scheme
+- symbols.</p>
+-
+- <p>If only one translation is provided, it should be designated
+- only by a package name, for example <tt>(mathlib)</tt>. This
+- translation is called the <i>default</i> translation.</p>
+-
+- <h3>Bundle Searching</h3>
+-
+- <p>When a message template is retrieved from a bundle, the
+- Scheme implementation will provide the locale under which the
+- system is currently running. When the template is retrieved,
+- the package name will be specified. The Scheme system should
+- construct a Bundle Specifier from the provided package name and
+- the active locale. For example, when retrieving a message
+- template for French Canadian, in the <tt>mathlib</tt> package,
+- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
+- program may also retrieve the elements of the current locale
+- using the no-argument procedures:</p>
+-
+- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-language</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- When given no arguments, returns the current ISO 639-1
+- language code as a symbol. If provided with an
+- argument, the current language is set to that named by the
+- symbol for the currently executing Scheme thread (or for the
+- entire Scheme system if such a distinction is not possible).
+-
+- </blockquote>
+-
+- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-country</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- returns the current ISO 3166-1 country code as a symbol.
+- If provided with an argument, the current country is
+- set to that named by the symbol for the currently executing
+- Scheme thread (or for the entire Scheme system if such a
+- distinction is not possible).
+- </blockquote>
+-
+- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
+- symbol</i></tt>s<br>
+- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- Returns a list of additional locale details as a list of
+- symbols. This list may contain information about
+- encodings or other more specific information. If
+- provided with an argument, the current locale details are set
+- to those given in the currently executing Scheme thread (or
+- for the entire Scheme system if such a distinction is not
+- possible).
+- </blockquote>
+-
+- <p>The Scheme System should first check for a bundle with the
+- exact name provided. If no such bundle is found, the last
+- element from the list is removed and a search is tried for a
+- bundle with that name. If no bundle is then found, the list is
+- shortened by removing the last element again. If no message is
+- found and the bundle specifier is now the empty list, an error
+- should be raised.</p>
+-
+- <p>The reason for this search order is to provide the most
+- locale sensitive template possible, but to fall back on more
+- general templates if a translation has not yet been provided
+- for the given locale.</p>
+-
+- <h3>Message Templates</h3>
+-
+- <p>A message template is a localized message that may or may
+- not contain one of a number of formatting codes. A message
+- template is a Scheme string. The string is of a form that can
+- be processed by the <tt>format</tt> procedure found in many
+- Scheme systems and formally specified in SRFI-28 (Basic Format
+- Strings).</p>
+-
+- <p>This SRFI also extends SRFI-28 to provide an additional
+- <tt>format</tt> escape code:</p>
+-
+- <blockquote>
+- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
+- follows this code immediately to reference the [N]'th
+- optional value absolutely, rather than the next unconsumed
+- value. The referenced value is <i>not</i> consumed.
+- </blockquote>
+- This extension allows optional values to be positionally
+- referenced, so that message templates can be constructed that
+- can produce the proper word ordering for a language.
+-
+- <h3>Preparing Bundles</h3>
+- Before a bundle may be used by the Scheme system to retrieve
+- localized template messages, they must be made available to the
+- Scheme system. This SRFI specifies a way to portably
+- define the bundles, as well as store them in and retrieve them
+- from an unspecified system which may be provided by resources
+- outside the Scheme system.<br>
+-
+-
+- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
+- association-list</i> -> undefined<br>
+- </tt></p>
+-
+- <blockquote>
+- Declares a new bundle named by the given bundle-specifier.
+- The contents of the bundle are defined by the provided
+- association list. The list contains associations
+- between Scheme symbols and the message templates (Scheme
+- strings) they name. If a bundle already exists with the
+- given name, it is overwritten with the newly declared
+- bundle.<br>
+- </blockquote>
+- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to store a bundle named by the given bundle
+- specifier, and previously made available using
+- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
+- unspecified mechanism that may be persistent across Scheme
+- system restarts. If successful, a non-false value is
+- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
+- </blockquote>
+- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to retrieve a bundle from an unspecified mechanism
+- which stores bundles outside the Scheme system. If the
+- bundle was retrieved successfully, the function returns a
+- non-false value, and the bundle is immediately available to
+- the Scheme system. If the bundle could not be found or loaded
+- successfully, the function returns <tt>#f</tt>, and the
+- Scheme system's bundle registry remains unaffected.<br>
+- </blockquote>
+- A compliant Scheme system may choose not to provide any
+- external mechanism to store localized bundles. If it does
+- not, it must still provide implementations for
+- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
+- case, both functions must return <tt>#f</tt> regardless of the
+- arguments given. Users of this SRFI should recognize that the
+- inability to load or store a localized bundle in an external
+- repository is <i>not</i> a fatal error.<br>
+-
+-
+- <h3>Retrieving Localized Message Templates</h3>
+-
+- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
+- message-template-name</tt></i> <tt>-> <i>string or #f<br>
+- </i></tt></p>
+-
+- <blockquote>
+- Retrieves a localized message template for the given package
+- name and the given message template name (both symbols).
+- If no such message could be found, false (#f) is
+- returned.<br>
+- <br>
+- </blockquote>
+- After retrieving a template, the calling program can use
+- <tt>format</tt> to produce a string that can be displayed to
+- the user.<br>
+-
+-
+- <h2>Examples</h2>
+- The below example makes use of SRFI-29 to display simple,
+- localized messages. It also defines its bundles in such a
+- way that the Scheme system may store and retrieve the bundles
+- from a more efficient system catalog, if available.<br>
+-
+-<pre>
+-(let ((translations
+- '(((en) . ((time . "Its ~a, ~a.")
+- (goodbye . "Goodbye, ~a.")))
+- ((fr) . ((time . "~1@*~a, c'est ~a.")
+- (goodbye . "Au revoir, ~a."))))))
+- (for-each (lambda (translation)
+- (let ((bundle-name (cons 'hello-program (car translation))))
+- (if (not (load-bundle! bundle-name))
+- (begin
+- (declare-bundle! bundle-name (cdr translation))
+- (store-bundle! bundle-name)))))
+- translations))
+-
+-(define localized-message
+- (lambda (message-name . args)
+- (apply format (cons (localized-template 'hello-program
+- message-name)
+- args))))
+-
+-(let ((myname "Fred"))
+- (display (localized-message 'time "12:00" myname))
+- (display #\newline)
+-
+- (display (localized-message 'goodbye myname))
+- (display #\newline))
+-
+-;; Displays (English):
+-;; Its 12:00, Fred.
+-;; Goodbye, Fred.
+-;;
+-;; French:
+-;; Fred, c'est 12:00.
+-;; Au revoir, Fred.
+-</pre>
+-
+- <H1>Implementation</H1>
+-
+- <p>The implementation requires that the Scheme system provide a
+- definition for <tt>current-language</tt> and
+- <tt>current-country</tt> capable of distinguishing the correct
+- locale present during a Scheme session. The definitions of
+- those functions in the reference implementation are not capable
+- of that distinction. Their implementation is provided only so
+- that the following code can run in any R4RS scheme system.
+- <br>
+- </p>
+-
+- <p>In addition, the below implementation of a compliant
+- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
+- SRFI-23 (Error reporting)</p>
+-<pre>
+-;; The association list in which bundles will be stored
+-(define *localization-bundles* '())
+-
+-;; The current-language and current-country functions provided
+-;; here must be rewritten for each Scheme system to default to the
+-;; actual locale of the session
+-(define current-language
+- (let ((current-language-value 'en))
+- (lambda args
+- (if (null? args)
+- current-language-value
+- (set! current-language-value (car args))))))
+-
+-(define current-country
+- (let ((current-country-value 'us))
+- (lambda args
+- (if (null? args)
+- current-country-value
+- (set! current-country-value (car args))))))
+-
+-;; The load-bundle! and store-bundle! both return #f in this
+-;; reference implementation. A compliant implementation need
+-;; not rewrite these procedures.
+-(define load-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-(define store-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-;; Declare a bundle of templates with a given bundle specifier
+-(define declare-bundle!
+- (letrec ((remove-old-bundle
+- (lambda (specifier bundle)
+- (cond ((null? bundle) '())
+- ((equal? (caar bundle) specifier)
+- (cdr bundle))
+- (else (cons (car bundle)
+- (remove-old-bundle specifier
+- (cdr bundle))))))))
+- (lambda (bundle-specifier bundle-assoc-list)
+- (set! *localization-bundles*
+- (cons (cons bundle-specifier bundle-assoc-list)
+- (remove-old-bundle bundle-specifier
+- *localization-bundles*))))))
+-
+-;;Retrieve a localized template given its package name and a template name
+-(define localized-template
+- (letrec ((rdc
+- (lambda (ls)
+- (if (null? (cdr ls))
+- '()
+- (cons (car ls) (rdc (cdr ls))))))
+- (find-bundle
+- (lambda (specifier template-name)
+- (cond ((assoc specifier *localization-bundles*) =>
+- (lambda (bundle) bundle))
+- ((null? specifier) #f)
+- (else (find-bundle (rdc specifier)
+- template-name))))))
+- (lambda (package-name template-name)
+- (let loop ((specifier (cons package-name
+- (list (current-language)
+- (current-country)))))
+- (and (not (null? specifier))
+- (let ((bundle (find-bundle specifier template-name)))
+- (and bundle
+- (cond ((assq template-name bundle) => cdr)
+- ((null? (cdr specifier)) #f)
+- (else (loop (rdc specifier)))))))))))
+-
+-;;An SRFI-28 and SRFI-29 compliant version of format. It requires
+-;;SRFI-23 for error reporting.
+-(define format
+- (lambda (format-string . objects)
+- (let ((buffer (open-output-string)))
+- (let loop ((format-list (string->list format-string))
+- (objects objects)
+- (object-override #f))
+- (cond ((null? format-list) (get-output-string buffer))
+- ((char=? (car format-list) #\~)
+- (cond ((null? (cdr format-list))
+- (error 'format "Incomplete escape sequence"))
+- ((char-numeric? (cadr format-list))
+- (let posloop ((fl (cddr format-list))
+- (pos (string->number
+- (string (cadr format-list)))))
+- (cond ((null? fl)
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (null? (cdr fl)))
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (eq? (cadr fl) '#\*))
+- (loop (cddr fl) objects (list-ref objects pos)))
+- (else
+- (posloop (cdr fl)
+- (+ (* 10 pos)
+- (string->number
+- (string (car fl)))))))))
+- (else
+- (case (cadr format-list)
+- ((#\a)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (display (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\s)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (write (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\%)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\newline buffer)
+- (loop (cddr format-list) objects #f))
+- ((#\~)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\~ buffer)
+- (loop (cddr format-list) objects #f))
+- (else
+- (error 'format "Unrecognized escape sequence"))))))
+- (else (display (car format-list) buffer)
+- (loop (cdr format-list) objects #f)))))))
+-
+-</pre>
+-
+- <H1>Copyright</H1>
+-
+- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
+-
+- <p>This document and translations of it may be copied and
+- furnished to others, and derivative works that comment on or
+- otherwise explain it or assist in its implementation may be
+- prepared, copied, published and distributed, in whole or in
+- part, without restriction of any kind, provided that the above
+- copyright notice and this paragraph are included on all such
+- copies and derivative works. However, this document itself may
+- not be modified in any way, such as by removing the copyright
+- notice or references to the Scheme Request For Implementation
+- process or editors, except as needed for the purpose of
+- developing SRFIs in which case the procedures for copyrights
+- defined in the SRFI process must be followed, or as required to
+- translate it into languages other than English.</p>
+-
+- <p>The limited permissions granted above are perpetual and will
+- not be revoked by the authors or their successors or
+- assigns.</p>
+-
+- <p>This document and the information contained herein is
+- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
+- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
+- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
+- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
+- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
+- <hr>
+-
+- <address>
+- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
+- Rush</a>
+- </address>
+-
+- <address>
+- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
+- Miller</a>
+- </address>
+- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
+- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
+- Daylight Time 2002 <!-- hhmts end --> <br>
+- </body>
+-</html>
+-
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html
+--- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,345 +0,0 @@
+-<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
+-<html>
+-<head>
+- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
+-</head>
+-<body>
+-
+-<H1>Title</H1>
+-
+-SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
+-
+-<H1>Author</H1>
+-
+-Andy Gaynor
+-
+-<H1>Status</H1>
+-
+-This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
+-You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
+-<P><UL>
+-<LI>Received: 1999/2/2
+-<LI>Draft: 1999/2/10-1999/04/12
+-<LI>Final: 1999/4/26
+-<LI>Revised reference implementation: 2003/01/27
+-</UL>
+-
+-<H1>Abstract</H1>
+-
+-The <i>named-let</i> incarnation of the <code>let</code> form has two slight
+-inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
+-form makes no accommodation for rest arguments, an issue of functionality
+-and consistency. As defined, the <code>let</code> form does not accommodate
+-signature-style syntax, an issue of aesthetics and consistency. Both
+-issues are addressed here in a manner which is compatible with the traditional
+-<code>let</code> form but for minor extensions.
+-
+-<H1>Rationale</H1>
+-
+-<H2>Signature-style Syntax</H2>
+-
+-Consider the following two equivalent definitions:
+-
+-<p><pre>
+-(define fibonacci
+- (lambda (n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
+-
+-(define (fibonacci n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-Although there is a named-let analog for the former form, there is none
+-for the latter. To wit, suppose one wished to compute the 10th element
+-of the Fibonacci sequence using a named let:
+-
+-<p>
+-<pre>
+-(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-Values: 55
+-</pre>
+-
+-As it stands, one cannot equivalently write
+-
+-<p>
+-<pre>
+-(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-which is consistent with <code>define</code>'s signature-style form.
+-<p>Those that favor the signature style may prefer this extension.
+-In any case, it may be more appropriate to include all bound names within
+-the binding section. As presented, this straightforward extension
+-introduces no ambiguity or incompatibility with the existing definition
+-of let.
+-
+-<H2>Rest Arguments</H2>
+-
+-As it stands, one cannot write a named let with rest arguments, as in
+-
+-<p>
+-<pre>
+-(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))
+-</pre>
+-
+-otherwise equivalent to
+-
+-<p>
+-<pre>
+-(letrec ((blast (lambda (port . x)
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))))
+- (blast (current-output-port) (+ 1 2) 4 5))
+-</pre>
+-
+-While this example is rather contrived, the functionality is not.
+-There are several times when the author has used this construct in practice.
+-Regardless, there is little reason to deny the <code>let</code> form access to
+-all the features of lambda functionality.
+-
+-<H2>Symbols in Binding Sections</H2>
+-
+-Both the features above rely upon the placement of symbols in <code>let</code>
+-binding lists (this statement is intentially simplistic). The only
+-other apparent use of such symbol placement is to tersely bind variables
+-to unspecified values. For example, one might desire to use
+-<code>(let (foo bar baz) ...)</code>
+-to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
+-unspecified values.
+-
+-<p>This usage is considered less important in light of the rationales
+-presented above, and an alternate syntax is immediately apparent, as
+-in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
+-be preferable, consistently parenthesizing normal binding clauses.
+-
+-<H1>Specification</H1>
+-
+-<H2>Syntax</H2>
+-
+-<p>
+-A formal specification of the syntax follows. Below, body, expression,
+-and identifier are free. Each instantiation of binding-name must be
+-unique.
+-</p>
+-
+-<p>
+-<pre>
+- let = "(" "let" let-bindings body ")"
+- expressions = nothing | expression expressions
+- let-bindings = let-name bindings
+- |"(" let-name "." bindings ")"
+- let-name = identifier
+- bindings = "(" ")"
+- | rest-binding
+- | "(" normal-bindings ["." rest-binding] ")"
+-normal-bindings = nothing
+- | normal-binding normal-bindings
+-normal-binding = "(" binding-name expression ")"
+- binding-name = identifier
+- rest-binding = "(" binding-name expressions ")"
+-</pre>
+-
+-<p>
+-For clarity and convenience, an informal specification follows.
+-</p>
+-
+-<ol>
+-<li><a name="unnamed">Unnamed</a>
+-
+-<p><pre>
+-(let ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig">
+-Named, non-signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-sig">Named, signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-
+-<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-</li>
+-</ol>
+-
+-<H2>Semantics</H2>
+-
+-Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
+-and <code>letrec</code> forms, respectively.
+-
+-<ul>
+-<li>For informal syntax 1:
+-
+-<p><pre>
+-(($lambda (<parameter>...) <body>...) <argument>...)
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 2 and 3:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...) <body>...)))
+- (<name> <argument>...))
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 4 and 5:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...
+-
+-. <rest-parameter>) <body>...)))
+- (<name> <argument>... <rest-argument>...))
+-</pre>
+-</li>
+-</ul>
+-
+-<H1>Implementation</H1>
+-
+-Here is an implementation using <code>SYNTAX-RULES</code>.
+-
+-<p>
+-<pre>
+-;; Use your own standard let.
+-;; Or call a lambda.
+-;; (define-syntax standard-let
+-;;
+-;; (syntax-rules ()
+-;;
+-;; ((let ((var val) ...) body ...)
+-;; ((lambda (var ...) body ...) val ...))))
+-
+-(define-syntax let
+-
+- (syntax-rules ()
+-
+- ;; No bindings: use standard-let.
+- ((let () body ...)
+- (standard-let () body ...))
+- ;; Or call a lambda.
+- ;; ((lambda () body ...))
+-
+- ;; All standard bindings: use standard-let.
+- ((let ((var val) ...) body ...)
+- (standard-let ((var val) ...) body ...))
+- ;; Or call a lambda.
+- ;; ((lambda (var ...) body ...) val ...)
+-
+- ;; One standard binding: loop.
+- ;; The all-standard-bindings clause didn't match,
+- ;; so there must be a rest binding.
+- ((let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; Signature-style name: loop.
+- ((let (name binding ...) body ...)
+- (let-loop name (binding ...) () () (body ...)))
+-
+- ;; defun-style name: loop.
+- ((let name bindings body ...)
+- (let-loop name bindings () () (body ...)))))
+-
+-(define-syntax let-loop
+-
+- (syntax-rules ()
+-
+- ;; Standard binding: destructure and loop.
+- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
+- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
+-
+- ;; Rest binding, no name: use standard-let, listing the rest values.
+- ;; Because of let's first clause, there is no "no bindings, no name" clause.
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
+- ;; Or call a lambda with a rest parameter on all values.
+- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
+- ;; Or use one of several other reasonable alternatives.
+-
+- ;; No bindings, name: call a letrec'ed lambda.
+- ((let-loop name () (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ...) . body)))
+- name)
+- val ...))
+-
+- ;; Rest binding, name: call a letrec'ed lambda.
+- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ... . rest-var) . body)))
+- name)
+- val ... rest-val ...))))
+-</pre>
+-
+-
+-<H1>Copyright</H1>
+-
+-Copyright (C) Andy Gaynor (1999). All Rights Reserved.
+-<p>This document and translations of it may be copied and furnished to
+-others, and derivative works that comment on or otherwise explain it or
+-assist in its implementation may be prepared, copied, published and distributed,
+-in whole or in part, without restriction of any kind, provided that the
+-above copyright notice and this paragraph are included on all such copies
+-and derivative works. However, this document itself may not be modified
+-in any way, such as by removing the copyright notice or references to the
+-Scheme Request For Implementation process or editors, except as needed
+-for the purpose of developing SRFIs in which case the procedures for copyrights
+-defined in the SRFI process must be followed, or as required to translate
+-it into languages other than English.
+-<p>The limited permissions granted above are perpetual and will not be
+-revoked by the authors or their successors or assigns.
+-<p>This document and the information contained herein is provided on an
+-"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
+-EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
+-USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
+-WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
+-
+- <hr>
+- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
+-
+-</body>
+-</html>
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/util.rkt racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/util.rkt
+--- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/util.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/util.rkt 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,61 @@
++#lang scribble/doc
++
++@(require scribble/manual
++ scribble/eval
++ scriblib/render-cond
++ scribble/core
++ scribble/html-properties
++ (for-syntax scheme/base)
++ (for-label scheme/base
++ racket/stream))
++
++@(provide (all-defined-out))
++
++@(define-syntax (srfi stx)
++ (syntax-case stx ()
++ [(_ num #:subdir subdir? . title)
++ (with-syntax ([srfi/n (string->symbol (format "srfi/~a" (syntax-e #'num)))])
++ #'(begin
++ (section #:tag (format "srfi-~a" num)
++ #:style 'unnumbered
++ (format "SRFI ~a: " num)
++ . title)
++ (defmodule srfi/n)
++ "Original specification: "
++ (let* ([label (format "SRFI ~a" num)]
++ [sub (if subdir? (format "srfi-~a/" num) "")]
++ [url ( (b) (format "~a/srfi-std/~asrfi-~a.html" b sub num))])
++ (cond-element
++ [(or latex text) @link[(url "http://docs.racket-lang.org") label]]
++ [else @link[(url ".") label]]))))]
++ [(_ num . title) #'(srfi num #:subdir #f . title)]))
++
++@;{ The `lst' argument is a list of
++ (list sym syntactic-form? html-anchor) }
++@(define (redirect n lst #:subdir [subdir? #f])
++ (let ([file (if subdir?
++ (format "srfi-~a/srfi-~a.html" n n)
++ (format "srfi-~a.html" n))]
++ [mod-path (string->symbol (format "srfi/~a" n))])
++ (make-binding-redirect-elements mod-path
++ (map (lambda (b)
++ (list (car b) (cadr b)
++ (build-path "srfi-std" file)
++ (caddr b)))
++ lst))))
++
++@(define in-core
++ (case-lambda
++ [() (in-core ".")]
++ [(k) @elem{This SRFI's bindings are also available in
++ @racketmodname[racket/base]@|k|}]))
++
++@(begin
++ (define-syntax-rule (def-mz mz-if)
++ (begin
++ (require (for-label mzscheme))
++ (define mz-if (racket if))))
++ (def-mz mz-if))
++
++@(define srfi-std (style #f (list (install-resource "srfi-std"))))
++
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,19 @@
++#lang info
++
++(define collection 'multi)
++
++(define build-deps '("mzscheme-doc"
++ "scheme-lib"
++ "base"
++ "scribble-lib"
++ "srfi-doc"
++ "srfi-lib-nonfree"
++ "racket-doc"
++ "r5rs-doc"
++ "r6rs-doc"
++ "compatibility-lib"))
++(define update-implies '("srfi-lib-nonfree"))
++
++(define pkg-desc "documentation part of \"srfi nonfree\"")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,3 @@
++#lang info
++
++(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,66 @@
++#lang scribble/doc
++@(require srfi/scribblings/util
++ scribble/manual
++ scribble/eval
++ scriblib/render-cond
++ scribble/core
++ scribble/html-properties
++ (for-syntax scheme/base)
++ (for-label scheme/base
++ racket/stream))
++
++@; ----------------------------------------------------------------------
++
++@title{SRFI Nonfree Libraries and Documentation}
++
++The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
++Implementation} (a.k.a. @deftech{SRFI}) process allows individual
++members of the Scheme community to propose libraries and extensions to
++be supported by multiple Scheme implementations.
++
++Racket is distributed with implementations of many SRFIs, most of
++which can be implemented as libraries. To import the bindings of SRFI
++@math{n}, use
++
++@racketblock[
++(require @#,elem{@racketidfont{srfi/}@math{n}})
++]
++
++This document lists the SRFIs that are supported by Racket and
++provides a link to the original SRFI specification (which is also
++distributed as part of Racket's documentation).
++
++The following SRFI specification documents are licensed restrictively.
++
++@table-of-contents[]
++
++
++@; ----------------------------------------
++
++@srfi[5]{A compatible let form with signatures and rest arguments}
++
++@redirect[5 '(
++ (let #t "unnamed")
++)]
++
++Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
++
++@; ----------------------------------------
++
++@srfi[29]{Localization}
++
++@redirect[29 '(
++ (current-language #f "current-language")
++ (current-country #f "current-country")
++ (current-locale-details #f "current-locale-details")
++ (declare-bundle! #f "declare-bundle!")
++ (store-bundle #f "store-bundle")
++ (load-bundle! #f "load-bundle!")
++ (localized-template #f "localized-template")
++)]
++
++Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
++
++@; ----------------------------------------
++
++@index-section[]
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,507 @@
++<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
++<html>
++ <head>
++ <meta name="generator" content="HTML Tidy, see www.w3.org">
++ <title>SRFI 29: Localization</title>
++ <meta name="author" content="Scott G. Miller">
++ <meta name="description" content="Localization">
++ </head>
++ <body>
++ <H1>Title</H1>
++
++ SRFI 29: Localization
++
++ <H1>Author</H1>
++
++ Scott G. Miller
++
++ <H1>Abstract</H1>
++
++ This document specifies an interface to retrieving and
++ displaying locale sensitive messages. A Scheme program can
++ register one or more translations of templated messages, and
++ then write Scheme code that can transparently retrieve the
++ appropriate message for the locale under which the Scheme
++ system is running. <br>
++
++
++ <H1>Rationale</H1>
++
++ <p>As any programmer that has ever had to deal with making his
++ or her code readable in more than one locale, the process of
++ sufficiently abstracting program messages from their
++ presentation to the user is non-trivial without help from the
++ programming language. Most modern programming language
++ libraries do provide some mechanism for performing this
++ separation.</p>
++
++ <p>A portable API that allows a piece of code to run without
++ modification in different countries and under different
++ languages is a must for any non-trivial software project.
++ The interface should separate the logic of a program from
++ the myriad of translations that may be necessary.</p>
++
++ <p>The interface described in this document provides such
++ functionality. The underlying implementation is also allowed to
++ use whatever datastructures it likes to provide access to the
++ translations in the most efficient manner possible. In
++ addition, the implementation is provided with standardized
++ functions that programs will use for accessing an external,
++ unspecified repository of translations.</p>
++
++ <p>This interface <i>does not</i> cover all aspects of
++ localization, including support for non-latin characters,
++ number and date formatting, etc. Such functionality is the
++ scope of a future SRFI that may extend this one.</p>
++
++ <H1>Dependencies</H1>
++
++ An SRFI-29 conformant implementation must also implement
++ SRFI-28, Basic Format Strings. Message templates are strings
++ that must be processed by the <tt>format</tt> function
++ specified in that SRFI.
++
++ <H1>Specification</H1>
++
++ <h3>Message Bundles</h3>
++
++ <p>A Message Bundle is a set of message templates and their
++ identifying keys. Each bundle contains one or more such
++ key/value pairs. The bundle itself is associated with a
++ <i>bundle specifier</i> which uniquely identifies the
++ bundle.</p>
++
++ <h3>Bundle Specifiers</h3>
++
++ <p>A Bundle Specifier is a Scheme list that describes, in order
++ of importance, the package and locale of a message bundle.
++ In most cases, a locale specifier will have between one
++ and three elements. The first element is a symbol denoting the
++ package for which this bundle applies. The second and third
++ elements denote a <i>locale</i>. The second element (first
++ element of the locale) if present, is the two letter, ISO 639-1
++ language code for the bundle. The third element, if present, is
++ a two letter ISO 3166-1 country code. In some cases, a
++ fourth element may be present, specifying the encoding used for
++ the bundle. All bundle specifier elements are Scheme
++ symbols.</p>
++
++ <p>If only one translation is provided, it should be designated
++ only by a package name, for example <tt>(mathlib)</tt>. This
++ translation is called the <i>default</i> translation.</p>
++
++ <h3>Bundle Searching</h3>
++
++ <p>When a message template is retrieved from a bundle, the
++ Scheme implementation will provide the locale under which the
++ system is currently running. When the template is retrieved,
++ the package name will be specified. The Scheme system should
++ construct a Bundle Specifier from the provided package name and
++ the active locale. For example, when retrieving a message
++ template for French Canadian, in the <tt>mathlib</tt> package,
++ the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
++ program may also retrieve the elements of the current locale
++ using the no-argument procedures:</p>
++
++ <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
++ <i>symbol</i></tt><br>
++ <tt><b>current-language</b> <i>symbol</i> ->
++ undefined</tt><br>
++ </p>
++
++ <blockquote>
++ When given no arguments, returns the current ISO 639-1
++ language code as a symbol. If provided with an
++ argument, the current language is set to that named by the
++ symbol for the currently executing Scheme thread (or for the
++ entire Scheme system if such a distinction is not possible).
++
++ </blockquote>
++
++ <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
++ <i>symbol</i></tt><br>
++ <tt><b>current-country</b> <i>symbol</i> ->
++ undefined</tt><br>
++ </p>
++
++ <blockquote>
++ returns the current ISO 3166-1 country code as a symbol.
++ If provided with an argument, the current country is
++ set to that named by the symbol for the currently executing
++ Scheme thread (or for the entire Scheme system if such a
++ distinction is not possible).
++ </blockquote>
++
++ <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
++ symbol</i></tt>s<br>
++ <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
++ undefined</tt><br>
++ </p>
++
++ <blockquote>
++ Returns a list of additional locale details as a list of
++ symbols. This list may contain information about
++ encodings or other more specific information. If
++ provided with an argument, the current locale details are set
++ to those given in the currently executing Scheme thread (or
++ for the entire Scheme system if such a distinction is not
++ possible).
++ </blockquote>
++
++ <p>The Scheme System should first check for a bundle with the
++ exact name provided. If no such bundle is found, the last
++ element from the list is removed and a search is tried for a
++ bundle with that name. If no bundle is then found, the list is
++ shortened by removing the last element again. If no message is
++ found and the bundle specifier is now the empty list, an error
++ should be raised.</p>
++
++ <p>The reason for this search order is to provide the most
++ locale sensitive template possible, but to fall back on more
++ general templates if a translation has not yet been provided
++ for the given locale.</p>
++
++ <h3>Message Templates</h3>
++
++ <p>A message template is a localized message that may or may
++ not contain one of a number of formatting codes. A message
++ template is a Scheme string. The string is of a form that can
++ be processed by the <tt>format</tt> procedure found in many
++ Scheme systems and formally specified in SRFI-28 (Basic Format
++ Strings).</p>
++
++ <p>This SRFI also extends SRFI-28 to provide an additional
++ <tt>format</tt> escape code:</p>
++
++ <blockquote>
++ <tt>~[n]@*</tt> - Causes a value-requiring escape code that
++ follows this code immediately to reference the [N]'th
++ optional value absolutely, rather than the next unconsumed
++ value. The referenced value is <i>not</i> consumed.
++ </blockquote>
++ This extension allows optional values to be positionally
++ referenced, so that message templates can be constructed that
++ can produce the proper word ordering for a language.
++
++ <h3>Preparing Bundles</h3>
++ Before a bundle may be used by the Scheme system to retrieve
++ localized template messages, they must be made available to the
++ Scheme system. This SRFI specifies a way to portably
++ define the bundles, as well as store them in and retrieve them
++ from an unspecified system which may be provided by resources
++ outside the Scheme system.<br>
++
++
++ <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
++ association-list</i> -> undefined<br>
++ </tt></p>
++
++ <blockquote>
++ Declares a new bundle named by the given bundle-specifier.
++ The contents of the bundle are defined by the provided
++ association list. The list contains associations
++ between Scheme symbols and the message templates (Scheme
++ strings) they name. If a bundle already exists with the
++ given name, it is overwritten with the newly declared
++ bundle.<br>
++ </blockquote>
++ <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
++ boolean</tt><br>
++
++
++ <blockquote>
++ Attempts to store a bundle named by the given bundle
++ specifier, and previously made available using
++ <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
++ unspecified mechanism that may be persistent across Scheme
++ system restarts. If successful, a non-false value is
++ returned. If unsuccessful, <tt>#f</tt> is returned.<br>
++ </blockquote>
++ <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
++ boolean</tt><br>
++
++
++ <blockquote>
++ Attempts to retrieve a bundle from an unspecified mechanism
++ which stores bundles outside the Scheme system. If the
++ bundle was retrieved successfully, the function returns a
++ non-false value, and the bundle is immediately available to
++ the Scheme system. If the bundle could not be found or loaded
++ successfully, the function returns <tt>#f</tt>, and the
++ Scheme system's bundle registry remains unaffected.<br>
++ </blockquote>
++ A compliant Scheme system may choose not to provide any
++ external mechanism to store localized bundles. If it does
++ not, it must still provide implementations for
++ <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
++ case, both functions must return <tt>#f</tt> regardless of the
++ arguments given. Users of this SRFI should recognize that the
++ inability to load or store a localized bundle in an external
++ repository is <i>not</i> a fatal error.<br>
++
++
++ <h3>Retrieving Localized Message Templates</h3>
++
++ <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
++ message-template-name</tt></i> <tt>-> <i>string or #f<br>
++ </i></tt></p>
++
++ <blockquote>
++ Retrieves a localized message template for the given package
++ name and the given message template name (both symbols).
++ If no such message could be found, false (#f) is
++ returned.<br>
++ <br>
++ </blockquote>
++ After retrieving a template, the calling program can use
++ <tt>format</tt> to produce a string that can be displayed to
++ the user.<br>
++
++
++ <h2>Examples</h2>
++ The below example makes use of SRFI-29 to display simple,
++ localized messages. It also defines its bundles in such a
++ way that the Scheme system may store and retrieve the bundles
++ from a more efficient system catalog, if available.<br>
++
++<pre>
++(let ((translations
++ '(((en) . ((time . "Its ~a, ~a.")
++ (goodbye . "Goodbye, ~a.")))
++ ((fr) . ((time . "~1@*~a, c'est ~a.")
++ (goodbye . "Au revoir, ~a."))))))
++ (for-each (lambda (translation)
++ (let ((bundle-name (cons 'hello-program (car translation))))
++ (if (not (load-bundle! bundle-name))
++ (begin
++ (declare-bundle! bundle-name (cdr translation))
++ (store-bundle! bundle-name)))))
++ translations))
++
++(define localized-message
++ (lambda (message-name . args)
++ (apply format (cons (localized-template 'hello-program
++ message-name)
++ args))))
++
++(let ((myname "Fred"))
++ (display (localized-message 'time "12:00" myname))
++ (display #\newline)
++
++ (display (localized-message 'goodbye myname))
++ (display #\newline))
++
++;; Displays (English):
++;; Its 12:00, Fred.
++;; Goodbye, Fred.
++;;
++;; French:
++;; Fred, c'est 12:00.
++;; Au revoir, Fred.
++</pre>
++
++ <H1>Implementation</H1>
++
++ <p>The implementation requires that the Scheme system provide a
++ definition for <tt>current-language</tt> and
++ <tt>current-country</tt> capable of distinguishing the correct
++ locale present during a Scheme session. The definitions of
++ those functions in the reference implementation are not capable
++ of that distinction. Their implementation is provided only so
++ that the following code can run in any R4RS scheme system.
++ <br>
++ </p>
++
++ <p>In addition, the below implementation of a compliant
++ <tt>format</tt> requires SRFI-6 (Basic String Ports) and
++ SRFI-23 (Error reporting)</p>
++<pre>
++;; The association list in which bundles will be stored
++(define *localization-bundles* '())
++
++;; The current-language and current-country functions provided
++;; here must be rewritten for each Scheme system to default to the
++;; actual locale of the session
++(define current-language
++ (let ((current-language-value 'en))
++ (lambda args
++ (if (null? args)
++ current-language-value
++ (set! current-language-value (car args))))))
++
++(define current-country
++ (let ((current-country-value 'us))
++ (lambda args
++ (if (null? args)
++ current-country-value
++ (set! current-country-value (car args))))))
++
++;; The load-bundle! and store-bundle! both return #f in this
++;; reference implementation. A compliant implementation need
++;; not rewrite these procedures.
++(define load-bundle!
++ (lambda (bundle-specifier)
++ #f))
++
++(define store-bundle!
++ (lambda (bundle-specifier)
++ #f))
++
++;; Declare a bundle of templates with a given bundle specifier
++(define declare-bundle!
++ (letrec ((remove-old-bundle
++ (lambda (specifier bundle)
++ (cond ((null? bundle) '())
++ ((equal? (caar bundle) specifier)
++ (cdr bundle))
++ (else (cons (car bundle)
++ (remove-old-bundle specifier
++ (cdr bundle))))))))
++ (lambda (bundle-specifier bundle-assoc-list)
++ (set! *localization-bundles*
++ (cons (cons bundle-specifier bundle-assoc-list)
++ (remove-old-bundle bundle-specifier
++ *localization-bundles*))))))
++
++;;Retrieve a localized template given its package name and a template name
++(define localized-template
++ (letrec ((rdc
++ (lambda (ls)
++ (if (null? (cdr ls))
++ '()
++ (cons (car ls) (rdc (cdr ls))))))
++ (find-bundle
++ (lambda (specifier template-name)
++ (cond ((assoc specifier *localization-bundles*) =>
++ (lambda (bundle) bundle))
++ ((null? specifier) #f)
++ (else (find-bundle (rdc specifier)
++ template-name))))))
++ (lambda (package-name template-name)
++ (let loop ((specifier (cons package-name
++ (list (current-language)
++ (current-country)))))
++ (and (not (null? specifier))
++ (let ((bundle (find-bundle specifier template-name)))
++ (and bundle
++ (cond ((assq template-name bundle) => cdr)
++ ((null? (cdr specifier)) #f)
++ (else (loop (rdc specifier)))))))))))
++
++;;An SRFI-28 and SRFI-29 compliant version of format. It requires
++;;SRFI-23 for error reporting.
++(define format
++ (lambda (format-string . objects)
++ (let ((buffer (open-output-string)))
++ (let loop ((format-list (string->list format-string))
++ (objects objects)
++ (object-override #f))
++ (cond ((null? format-list) (get-output-string buffer))
++ ((char=? (car format-list) #\~)
++ (cond ((null? (cdr format-list))
++ (error 'format "Incomplete escape sequence"))
++ ((char-numeric? (cadr format-list))
++ (let posloop ((fl (cddr format-list))
++ (pos (string->number
++ (string (cadr format-list)))))
++ (cond ((null? fl)
++ (error 'format "Incomplete escape sequence"))
++ ((and (eq? (car fl) '#\@)
++ (null? (cdr fl)))
++ (error 'format "Incomplete escape sequence"))
++ ((and (eq? (car fl) '#\@)
++ (eq? (cadr fl) '#\*))
++ (loop (cddr fl) objects (list-ref objects pos)))
++ (else
++ (posloop (cdr fl)
++ (+ (* 10 pos)
++ (string->number
++ (string (car fl)))))))))
++ (else
++ (case (cadr format-list)
++ ((#\a)
++ (cond (object-override
++ (begin
++ (display object-override buffer)
++ (loop (cddr format-list) objects #f)))
++ ((null? objects)
++ (error 'format "No value for escape sequence"))
++ (else
++ (begin
++ (display (car objects) buffer)
++ (loop (cddr format-list)
++ (cdr objects) #f)))))
++ ((#\s)
++ (cond (object-override
++ (begin
++ (display object-override buffer)
++ (loop (cddr format-list) objects #f)))
++ ((null? objects)
++ (error 'format "No value for escape sequence"))
++ (else
++ (begin
++ (write (car objects) buffer)
++ (loop (cddr format-list)
++ (cdr objects) #f)))))
++ ((#\%)
++ (if object-override
++ (error 'format "Escape sequence following positional override does not require a value"))
++ (display #\newline buffer)
++ (loop (cddr format-list) objects #f))
++ ((#\~)
++ (if object-override
++ (error 'format "Escape sequence following positional override does not require a value"))
++ (display #\~ buffer)
++ (loop (cddr format-list) objects #f))
++ (else
++ (error 'format "Unrecognized escape sequence"))))))
++ (else (display (car format-list) buffer)
++ (loop (cdr format-list) objects #f)))))))
++
++</pre>
++
++ <H1>Copyright</H1>
++
++ Copyright (C) Scott G. Miller (2002). All Rights Reserved.
++
++ <p>This document and translations of it may be copied and
++ furnished to others, and derivative works that comment on or
++ otherwise explain it or assist in its implementation may be
++ prepared, copied, published and distributed, in whole or in
++ part, without restriction of any kind, provided that the above
++ copyright notice and this paragraph are included on all such
++ copies and derivative works. However, this document itself may
++ not be modified in any way, such as by removing the copyright
++ notice or references to the Scheme Request For Implementation
++ process or editors, except as needed for the purpose of
++ developing SRFIs in which case the procedures for copyrights
++ defined in the SRFI process must be followed, or as required to
++ translate it into languages other than English.</p>
++
++ <p>The limited permissions granted above are perpetual and will
++ not be revoked by the authors or their successors or
++ assigns.</p>
++
++ <p>This document and the information contained herein is
++ provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
++ EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
++ BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
++ HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
++ OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
++ <hr>
++
++ <address>
++ Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
++ Rush</a>
++ </address>
++
++ <address>
++ Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
++ Miller</a>
++ </address>
++ <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
++ <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
++ Daylight Time 2002 <!-- hhmts end --> <br>
++ </body>
++</html>
++
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,345 @@
++<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
++<html>
++<head>
++ <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
++</head>
++<body>
++
++<H1>Title</H1>
++
++SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
++
++<H1>Author</H1>
++
++Andy Gaynor
++
++<H1>Status</H1>
++
++This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
++You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
++<P><UL>
++<LI>Received: 1999/2/2
++<LI>Draft: 1999/2/10-1999/04/12
++<LI>Final: 1999/4/26
++<LI>Revised reference implementation: 2003/01/27
++</UL>
++
++<H1>Abstract</H1>
++
++The <i>named-let</i> incarnation of the <code>let</code> form has two slight
++inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
++form makes no accommodation for rest arguments, an issue of functionality
++and consistency. As defined, the <code>let</code> form does not accommodate
++signature-style syntax, an issue of aesthetics and consistency. Both
++issues are addressed here in a manner which is compatible with the traditional
++<code>let</code> form but for minor extensions.
++
++<H1>Rationale</H1>
++
++<H2>Signature-style Syntax</H2>
++
++Consider the following two equivalent definitions:
++
++<p><pre>
++(define fibonacci
++ (lambda (n i f0 f1)
++ (if (= i n)
++ f0
++ (fibonacci n (+ i 1) f1 (+ f0 f1)))))
++
++(define (fibonacci n i f0 f1)
++ (if (= i n)
++ f0
++ (fibonacci n (+ i 1) f1 (+ f0 f1))))
++</pre>
++
++Although there is a named-let analog for the former form, there is none
++for the latter. To wit, suppose one wished to compute the 10th element
++of the Fibonacci sequence using a named let:
++
++<p>
++<pre>
++(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
++ (if (= i n)
++ f0
++ (fibonacci n (+ i 1) f1 (+ f0 f1))))
++Values: 55
++</pre>
++
++As it stands, one cannot equivalently write
++
++<p>
++<pre>
++(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
++ (if (= i n)
++ f0
++ (fibonacci n (+ i 1) f1 (+ f0 f1))))
++</pre>
++
++which is consistent with <code>define</code>'s signature-style form.
++<p>Those that favor the signature style may prefer this extension.
++In any case, it may be more appropriate to include all bound names within
++the binding section. As presented, this straightforward extension
++introduces no ambiguity or incompatibility with the existing definition
++of let.
++
++<H2>Rest Arguments</H2>
++
++As it stands, one cannot write a named let with rest arguments, as in
++
++<p>
++<pre>
++(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
++ (if (null? x)
++ 'just-a-silly-contrived-example
++ (begin
++ (write (car x) port)
++ (apply blast port (cdr x)))))
++</pre>
++
++otherwise equivalent to
++
++<p>
++<pre>
++(letrec ((blast (lambda (port . x)
++ (if (null? x)
++ 'just-a-silly-contrived-example
++ (begin
++ (write (car x) port)
++ (apply blast port (cdr x)))))))
++ (blast (current-output-port) (+ 1 2) 4 5))
++</pre>
++
++While this example is rather contrived, the functionality is not.
++There are several times when the author has used this construct in practice.
++Regardless, there is little reason to deny the <code>let</code> form access to
++all the features of lambda functionality.
++
++<H2>Symbols in Binding Sections</H2>
++
++Both the features above rely upon the placement of symbols in <code>let</code>
++binding lists (this statement is intentially simplistic). The only
++other apparent use of such symbol placement is to tersely bind variables
++to unspecified values. For example, one might desire to use
++<code>(let (foo bar baz) ...)</code>
++to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
++unspecified values.
++
++<p>This usage is considered less important in light of the rationales
++presented above, and an alternate syntax is immediately apparent, as
++in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
++be preferable, consistently parenthesizing normal binding clauses.
++
++<H1>Specification</H1>
++
++<H2>Syntax</H2>
++
++<p>
++A formal specification of the syntax follows. Below, body, expression,
++and identifier are free. Each instantiation of binding-name must be
++unique.
++</p>
++
++<p>
++<pre>
++ let = "(" "let" let-bindings body ")"
++ expressions = nothing | expression expressions
++ let-bindings = let-name bindings
++ |"(" let-name "." bindings ")"
++ let-name = identifier
++ bindings = "(" ")"
++ | rest-binding
++ | "(" normal-bindings ["." rest-binding] ")"
++normal-bindings = nothing
++ | normal-binding normal-bindings
++normal-binding = "(" binding-name expression ")"
++ binding-name = identifier
++ rest-binding = "(" binding-name expressions ")"
++</pre>
++
++<p>
++For clarity and convenience, an informal specification follows.
++</p>
++
++<ol>
++<li><a name="unnamed">Unnamed</a>
++
++<p><pre>
++(let ((<parameter> <argument>)...)
++ <body>...)
++</pre>
++</li>
++
++<li><a name="named-non-sig">
++Named, non-signature-style, no rest argument</a>
++
++<p><pre>
++(let <name> ((<parameter> <argument>)...)
++ <body>...)
++</pre>
++</li>
++
++<li><a name="named-sig">Named, signature-style, no rest argument</a>
++
++<p><pre>
++(let (<name> (<parameter> <argument>)...)
++ <body>...)
++</pre>
++</li>
++
++<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
++
++<p><pre>
++(let <name> ((<parameter> <argument>)...
++
++. (<rest-parameter> <rest-argument>...))
++ <body>...)
++</pre>
++
++<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
++
++<p><pre>
++(let (<name> (<parameter> <argument>)...
++
++. (<rest-parameter> <rest-argument>...))
++ <body>...)
++</pre>
++</li>
++</ol>
++
++<H2>Semantics</H2>
++
++Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
++and <code>letrec</code> forms, respectively.
++
++<ul>
++<li>For informal syntax 1:
++
++<p><pre>
++(($lambda (<parameter>...) <body>...) <argument>...)
++</pre>
++</li>
++
++<li>For informal syntaxes 2 and 3:
++
++<p>
++<pre>
++($letrec ((<name> ($lambda (<parameter>...) <body>...)))
++ (<name> <argument>...))
++</pre>
++</li>
++
++<li>For informal syntaxes 4 and 5:
++
++<p>
++<pre>
++($letrec ((<name> ($lambda (<parameter>...
++
++. <rest-parameter>) <body>...)))
++ (<name> <argument>... <rest-argument>...))
++</pre>
++</li>
++</ul>
++
++<H1>Implementation</H1>
++
++Here is an implementation using <code>SYNTAX-RULES</code>.
++
++<p>
++<pre>
++;; Use your own standard let.
++;; Or call a lambda.
++;; (define-syntax standard-let
++;;
++;; (syntax-rules ()
++;;
++;; ((let ((var val) ...) body ...)
++;; ((lambda (var ...) body ...) val ...))))
++
++(define-syntax let
++
++ (syntax-rules ()
++
++ ;; No bindings: use standard-let.
++ ((let () body ...)
++ (standard-let () body ...))
++ ;; Or call a lambda.
++ ;; ((lambda () body ...))
++
++ ;; All standard bindings: use standard-let.
++ ((let ((var val) ...) body ...)
++ (standard-let ((var val) ...) body ...))
++ ;; Or call a lambda.
++ ;; ((lambda (var ...) body ...) val ...)
++
++ ;; One standard binding: loop.
++ ;; The all-standard-bindings clause didn't match,
++ ;; so there must be a rest binding.
++ ((let ((var val) . bindings) body ...)
++ (let-loop #f bindings (var) (val) (body ...)))
++
++ ;; Signature-style name: loop.
++ ((let (name binding ...) body ...)
++ (let-loop name (binding ...) () () (body ...)))
++
++ ;; defun-style name: loop.
++ ((let name bindings body ...)
++ (let-loop name bindings () () (body ...)))))
++
++(define-syntax let-loop
++
++ (syntax-rules ()
++
++ ;; Standard binding: destructure and loop.
++ ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
++ (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
++
++ ;; Rest binding, no name: use standard-let, listing the rest values.
++ ;; Because of let's first clause, there is no "no bindings, no name" clause.
++ ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
++ (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
++ ;; Or call a lambda with a rest parameter on all values.
++ ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
++ ;; Or use one of several other reasonable alternatives.
++
++ ;; No bindings, name: call a letrec'ed lambda.
++ ((let-loop name () (var ...) (val ...) body)
++ ((letrec ((name (lambda (var ...) . body)))
++ name)
++ val ...))
++
++ ;; Rest binding, name: call a letrec'ed lambda.
++ ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
++ ((letrec ((name (lambda (var ... . rest-var) . body)))
++ name)
++ val ... rest-val ...))))
++</pre>
++
++
++<H1>Copyright</H1>
++
++Copyright (C) Andy Gaynor (1999). All Rights Reserved.
++<p>This document and translations of it may be copied and furnished to
++others, and derivative works that comment on or otherwise explain it or
++assist in its implementation may be prepared, copied, published and distributed,
++in whole or in part, without restriction of any kind, provided that the
++above copyright notice and this paragraph are included on all such copies
++and derivative works. However, this document itself may not be modified
++in any way, such as by removing the copyright notice or references to the
++Scheme Request For Implementation process or editors, except as needed
++for the purpose of developing SRFIs in which case the procedures for copyrights
++defined in the SRFI process must be followed, or as required to translate
++it into languages other than English.
++<p>The limited permissions granted above are perpetual and will not be
++revoked by the authors or their successors or assigns.
++<p>This document and the information contained herein is provided on an
++"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
++EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
++USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
++WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
++
++ <hr>
++ <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
++
++</body>
++</html>
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/info.rkt racket-6.12/share/pkgs/srfi-lib/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/info.rkt 2018-01-26 16:10:02.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/info.rkt 2018-01-31 12:22:58.144502000 -0500
+@@ -1 +1,15 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("scheme-lib" "base" "srfi-lite-lib" "r6rs-lib" "compatibility-lib"))) (define implies (quote ("srfi-lite-lib"))) (define pkg-desc "implementation (no documentation) part of \"srfi\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
++#lang info
++
++(define collection 'multi)
++
++(define deps '("scheme-lib"
++ "base"
++ "srfi-lite-lib"
++ "r6rs-lib"
++ "compatibility-lib"))
++
++(define implies '("srfi-lite-lib"))
++
++(define pkg-desc "implementation (no documentation) part of \"srfi\"")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/sort.txt racket-6.12/share/pkgs/srfi-lib/srfi/32/sort.txt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/sort.txt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/32/sort.txt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,1069 +0,0 @@
+-The SRFI-32 sort libraries -*- outline -*-
+-Olin Shivers
+-First draft: 1998/10/19
+-Last update: 2002/7/21
+-
+-[Todo: del-list-neighbor-dups!
+- vector-copy -> subvector
+- use srfi-23 for reporting errors
+- use srfi-16 for n-aries?
+-
+-Emacs should display this document in outline mode. Say c-h m for
+-instructions on how to move through it by sections (e.g., c-c c-n, c-c c-p).
+-
+-* Table of contents
+--------------------
+-Abstract
+-Procedure index
+-Introduction
+-What's wrong with the current state of affairs?
+-Design rules
+- What vs. how
+- Consistency across function signatures
+- Data parameter first, less-than parameter after
+- Ordering, comparison functions & stability
+- All vector operations accept optional subrange parameters
+- Required vs. allowed side-effects
+-Procedure specification
+- Procedure naming and functionality
+- Types of parameters and return values
+- sort-lib - general sorting package
+- Algorithm-specific sorting packages
+-Algorithmic properties
+-Topics to be resolved during discussion phase
+-Porting and optimisation
+-References & Links
+-Acknowledgements
+-Copyright
+-
+-
+-* Abstract
+-----------
+-Current Scheme sorting packages are, every one of them, surprisingly bad. I've
+-designed the API for a full-featured sort toolkit, which I propose as a SRFI.
+-
+-The spec comes with 1200 lines of high-quality reference code: tightly
+-written, highly commented, portable code, available for free. Implementors
+-want this code. It's better than what you have.
+-
+--------------------------------------------------------------------------------
+-* Procedure index
+------------------
+-list-sorted? vector-sorted?
+-
+-list-merge vector-merge
+-list-sort vector-sort
+-list-stable-sort vector-stable-sort
+-list-delete-neighbor-dups vector-delete-neighbor-dups
+-
+-list-merge! vector-merge!
+-list-sort! vector-sort!
+-list-stable-sort! vector-stable-sort!
+-list-delete-neighbor-dups! vector-delete-neighbor-dups!
+-
+-quick-sort heap-sort insert-sort list-merge-sort vector-merge-sort
+-quick-sort! heap-sort! insert-sort! list-merge-sort! vector-merge-sort!
+-quick-sort3!
+-
+-vector-binary-search
+-vector-binary-search3
+-
+--------------------------------------------------------------------------------
+-* Introduction
+---------------
+-As I'll detail below, I wasn't very happy with the state of the Scheme
+-world for sorting and merging lists and vectors. So I have designed and
+-written a fairly comprehensive sorting & merging toolkit. It is
+-
+- - very portable,
+-
+- - much better code than what is currently in Elk, Gambit, Bigloo,
+- Scheme->C, MzScheme, RScheme, Scheme48, MIT Scheme, or slib, and
+-
+- - priced to move: free code.
+-
+-The package includes
+- - Vector insert sort (stable)
+- - Vector heap sort
+- - Vector quick sort (with median-of-3 pivot picking)
+- - Vector merge sort (stable)
+- - Pure and destructive list merge sort (stable)
+- - Stable vector and list merge
+- - Miscellaneous sort-related procedures: Vector and list merging,
+- sorted? predicates, vector binary search, vector and list
+- delete-equal-neighbor procedures.
+- - A general, non-algorithmic set of procedure names for general sorting
+- and merging.
+-
+-Scheme programmers may want to adopt this package. I'd like Scheme
+-implementors to adopt this code and its API -- in fact, the code is a bribe to
+-make it easy for implementors to converge on the suggested API. I mean, you'd
+-really have to be a boor to take this free code I wrote and mutate its
+-interface over to your incompatible, unportable API, wouldn't you? But you
+-could, of course -- it's freely available. More in the spirit of the offering,
+-you could make this API available, and then also write a little module
+-providing your old interface that is defined in terms of this API. "Scheme
+-implementors," in this context, includes slib, which is not a standalone
+-implementation of Scheme, but rather an influential collection of API's and
+-code.
+-
+-The code is tightly bummed. It is clearly written, and commented in my usual
+-voluminous style. This includes notes on porting and implementation-specific
+-optimisations.
+-
+-
+--------------------------------------------------------------------------------
+-* What's wrong with the current state of affairs?
+--------------------------------------------------
+-
+-It's just amazing to me that in 2002, sorting and merging hasn't been
+-completely put to bed. These are well-understood algorithms, each of them well
+-under a page of code. The straightforward algorithms are basic, core stuff --
+-sophomore-level. But if you tour the major Scheme implementations out there on
+-the Net, you find badly written code that provides extremely spotty coverage
+-of the algorithm space. One implementation even has a buggy implementation
+-that has been in use for about 20 years. Another has an O(n^2) algorithm...
+-implemented in C for speed.
+-
+-Open source-code is a wonderful thing. In a couple of hours, I was able to
+-download and check the sources of 9 Scheme systems. Here are my notes from the
+-systems I checked. You can skip to the next section if you aren't morbidly
+-curious.
+-
+-slib
+- sorted? vector-or-list <
+- merge list1 list2 <
+- merge! list1 list2 <
+- sort vector-or-list <
+- sort! vector-or-list <
+-
+- Richard O'Keefe's stable list merge sort is right idea, but implemented
+- using gratuitous variable side effects. It also does redundant SET-CDR!s.
+- The vector sort converts to list, merge sorts, then reconverts
+- to vector. This is a bad idea -- non-local pointer chasing bad; vector
+- shuffling good. If you must allocate temp storage, might as well allocate
+- a temp vector and use vector merge sort.
+-
+-MIT Scheme
+- sort! vector <
+- merge-sort! vector <
+- quick-sort! vector <
+-
+- sort vector-or-list <
+- merge-sort vector-or-list <
+- quick-sort vector-or-list <
+-
+- Naive vector quicksort: loser, for worst-case performance reasons.
+- List sort by "list->vector; quicksort; vector->list," hence also loser.
+- A clever stable vector merge sort, albeit not very bummed.
+-
+-Scheme 48 & T
+- sort-list list <
+- sort-list! list <
+- list-merge! list1 list2 <
+-
+- Bob Nix's implementation of online merge-sort, written in the early 80's.
+- Conses unnecessary bookkeeping structure, which isn't necessary with a
+- proper recursive formulation. Also, does redundant SET-CDR!s. No vector
+- sort. Also, has a bug -- is claimed to be a stable sort, but isn't! To see
+- this, get the S48 code, and try
+- (define (my< x y) (< (abs x) (abs y)))
+- (list-merge! (list 0 2) (list -2) my<) ; -> (0 2 -2)
+- (list-merge! (list 2) (list 0 -2) my<) ; -> (0 -2 2)
+- This could be fixed very easily, but it isn't worth it given the
+- other problems with the algorithm.
+-
+-RScheme
+- vector-sort! vector <
+- sort collection <
+-
+- Good basic implementation of vector heapsort, which has O(n lg n)
+- worst-case time. Code ugly, needs tuning. List sort by "list->vector;
+- sort; vector->list." Nothing for stable sorting.
+-
+-MzScheme
+- quicksort lis <
+- mergesort alox <
+-
+- Sorts lists with (list->vector; quicksort; vector->list) -- but the core
+- quicksort is not available for vector sorting. Nothing for stable sorting.
+- Quicksort picks pivot naively, inducing O(n^2) worse-case behaviour on a
+- fairly common case: an already-sorted list.
+-
+-Bigloo, STK
+- sort vector-or-list <
+- Uses an O(n^2) algorithm... implemented in C for speed. Hmm.
+- (See runtime/Ieee/vector.scm and runtime/Clib/cvector.c)
+-
+-Gambit
+- sort-list list <
+- Nothing for vectors. Simple, slow, unstable merge sort for lists.
+-
+-Elk
+- Another naive quicksort. Lists handled by converting to vector.
+- sort vector-or-list <
+- sort! vector-or-list <
+-
+-Chez Scheme
+- merge < list1 list2
+- merge! < list1 list2
+- sort < list
+- sort! < list
+-
+- These are stable. I have not seen the source code.
+-
+-Common Lisp
+- sort sequence < [key]
+- stable-sort sequence < [key]
+- merge result-type sequence1 sequence2 < [key]
+-
+- The sort procedures are allowed, but not required, to be destructive.
+-
+-SML/NJ
+- sort: ('a*'a -> bool) -> 'a list -> 'a list
+- "Smooth applicative merge sort," which is stable.
+- There is also a highly bummed quicksort for vectors.
+-
+-The right solution: Implement a full toolbox of carefully written standard sort
+-routines.
+-
+-Having the source of all these above-cited Schemes available for study made
+-life a lot easier writing this code. I appreciate the authors making their
+-source available under such open terms.
+-
+-
+--------------------------------------------------------------------------------
+-* Design rules
+---------------
+-
+-** What vs. how
+-===============
+-There are two different interfaces: "what" (simple) & "how" (detailed).
+-
+- - Simple: you specify semantics: datatype (list or vector),
+- mutability, and stability.
+-
+- - Detailed: you specify the actual algorithm (quick, heap,
+- insert, merge). Different algorithms have different properties,
+- both semantic & pragmatic, so these exports are necessary.
+-
+- It is necessarily the case that the specifications of these procedures
+- make statements about execution "pragmatics." For example, the sole
+- distinction between heap sort and quick sort -- both of which are
+- provided by this library -- is one of execution time, which is not a
+- "semantic" distinction. Similar resource-use statements are made about
+- "iterative" procedures, meaning that they can execute on input of
+- arbitrary size in a constant number of stack frames.
+-
+-** Consistency across function signatures
+-=========================================
+-The two interfaces share common function signatures wherever
+-possible, to facilitate switching a given call from one procedure
+-to another.
+-
+-** Less-than parameter first, data parameter after
+-==================================================
+-These procedures uniformly observe the following parameter order:
+-the data to be sorted comes after the comparison function.
+-That is, we write
+- (sort < lis)
+-not
+- (sort lis <).
+-
+-With the sole exception of Chez Scheme, this is the exact opposite of
+-every sort function out there in current use in the Scheme world. (See
+-the summary of related APIs above.) However, it is consistent with common
+-practice across Scheme libraries in general to put the ordering function
+-first -- the "operation currying" convention. (E.g., consider FOR-EACH or
+-MAP or FIND.)
+-
+-The original draft of this SRFI used the data-first/comparison-last convention
+-for backwards compatibility -- a decision I made with internal misgivings.
+-Happily, however, the overwhelming response from the discussion phase
+-supported "cleaning up" this issue and re-converging the parameter order with
+-the general Scheme "op currying" convention. So the original decision was
+-inverted in favor of the comparison-first/data-last convention.
+-
+-** Ordering, comparison functions & stability
+-=============================================
+-These routines take a < comparison function, not a <= comparison
+-function, and they sort into increasing order. The difference between
+-a < spec and a <= spec comes up in three places:
+- - the definition of an ordered or sorted data set,
+- - the definition of a stable sorting algorithm, and
+- - correctness of quicksort.
+-
+-+ We say that a data set (a list or vector) is *sorted* or *ordered*
+- if it contains no adjacent pair of values ... X Y ... such that Y < X.
+-
+- In other words, scanning across the data never takes a "downwards" step.
+-
+- If you use a <= procedure where these algorithms expect a <
+- procedure, you may not get the answers you expect. For example,
+- the LIST-SORTED? function will return false if you pass it a <= comparison
+- function and an ordered list containing adjacent equal elements.
+-
+-+ A "stable" sort is one that preserves the pre-existing order of equal
+- elements. Suppose, for example, that we sort a list of numbers by
+- comparing their absolute values, i.e., using comparison function
+- (lambda (x y) (< (abs x) (abs y)))
+- If we sort a list that contains both 3 and -3:
+- ... 3 ... -3 ...
+- then a stable sort is an algorithm that will not swap the order
+- of these two elements, that is, the answer is guaranteed to to look like
+- ... 3 -3 ...
+- not
+- ... -3 3 ...
+-
+- Choosing < for the comparison function instead of <= affects how stability
+- is coded. Given an adjacent pair X Y, (< y x) means "Y should be moved in
+- front of X" -- otherwise, leave things as they are. So using a <= function
+- where a < function is expected will *invert* stability.
+-
+- This is due to the definition of equality, given a < comparator:
+- (and (not (< x y))
+- (not (< y x)))
+- The definition is rather different, given a <= comparator:
+- (and (<= x y)
+- (<= y x))
+-
+-+ A "stable" merge is one that reliably favors one of its data sets
+- when equal items appear in both data sets. *All merge operations in
+- this library are stable*, breaking ties between data sets in favor
+- of the first data set -- elements of the first list come before equal
+- elements in the second list.
+-
+- So, if we are merging two lists of numbers ordered by absolute value,
+- the stable merge operation LIST-MERGE
+- (list-merge (lambda (x y) (< (abs x) (abs y)))
+- '(0 -2 4 8 -10) '(-1 3 -4 7))
+- reliably places the 4 of the first list before the equal-comparing -4
+- of the second list:
+- (0 -1 -2 4 -4 7 8 -10)
+-
+-+ Some sort algorithms will *not work correctly* if given a <= when they
+- expect a < comparison (or vice-versa). For example, violating quicksort's
+- spec may cause it to produce wrong answers, diverge, raise an error, or do
+- some fourth thing. To see why, consider the left-scan part of the standard
+- quicksort partition step:
+- (let ((i (let scan ((i i)) (if (elt< (vector-ref v i) pivot)
+- (scan (+ i 1))
+- i))))
+- ...)
+- Consider applying this loop to a vector of all zeroes (hence, PIVOT, as
+- well, is zero), but erroneously using <= for the ELT< function. The loop
+- will scan right off the end of the vector, producing a vector-index error.
+- The guarantee that the scan loop will terminate before running off the end
+- of the vector depends critically upon ELT< performing as a true, irreflexive
+- < relation. Running off the end of the vector is only one of a variety of
+- possibly ways to lose -- other, variant implementations of quicksort can,
+- instead, loop forever on some data sets if ELT< is a <= predicate.
+-
+-In short, if your comparison function F answers true to (F x x), then
+- - using a stable sorting or merging algorithm will not give you a
+- stable sort or merge,
+- - LIST-SORTED? may surprise you, and
+- - quicksort may fail in a variety of possible ways.
+-Note that you can synthesize a < function from a <= function with
+- (lambda (x y) (not (<= y x)))
+-if need be.
+-
+-Precise definitions give sharp edges to tools, but require care in use.
+-"Measure twice, cut once."
+-
+-I have adopted the choice of < from Common Lisp. One would assume the definers
+-of Common Lisp had a good reason for adopting < instead of <=, but canvassing
+-several of the principal actors in the definition process has turned up no
+-better reason than "an arbitrary but consistent choice." At minimum, then,
+-this SRFI extends the coverage of that consistent choice.
+-
+-** All vector operations accept optional subrange parameters
+-============================================================
+-The vector operations specified below all take optional START/END arguments
+-indicating a selected subrange of a vector's elements. If a START parameter or
+-START/END parameter pair is given to such a procedure, they must be exact,
+-non-negative integers, such that
+- 0 <= START <= END <= (VECTOR-LENGTH V)
+-where V is the related vector parameter. If not specified, they default to 0
+-and the length of the vector, respectively. They are interpreted to select the
+-range [START,END), that is, all elements from index START (inclusive) up to,
+-but not including, index END.
+-
+-** Required vs. allowed side-effects
+-====================================
+-LIST-SORT! and LIST-STABLE-SORT! are allowed, but not required,
+-to alter their arguments' cons cells to construct the result list. This is
+-consistent with the what-not-how character of the group of procedures
+-to which they belong (the "sort-lib" package).
+-
+-The LIST-DELETE-NEIGHBOR-DUPS!, LIST-MERGE! and LIST-MERGE-SORT! procedures,
+-on the other hand, provide specific algorithms, and, as such, explicitly
+-commit to the use of side-effects on their input lists in order to guarantee
+-their key algorithmic properties (e.g., linear-time operation, constant-space
+-stack use).
+-
+--------------------------------------------------------------------------------
+-* Procedure specification
+--------------------------
+-The procedures are split into several packages. In a Scheme system that has a
+-module or package system, these procedures should be contained in modules
+-named as follows:
+- Package name Functionality
+- ------------ -------------
+- sort-lib General sorting for lists & vectors
+- sorted?-lib Sorted predicates for lists & vectors
+- list-merge-sort-lib List merge sort
+- vector-merge-sort-lib Vector merge sort
+- vector-heap-sort-lib Vector heap sort
+- vector-quick-sort-lib Vector quick sort
+- vector-insert-sort-lib Vector insertion sort
+- delndup-lib List and vector delete neighbor duplicates
+- binsearch-lib Vector binary search
+-
+-A Scheme system without a module system should provide all of the bindings
+-defined in all of these modules as components of the "SRFI-32" package.
+-
+-Note that there is no "list insert sort" package, as you might as well always
+-use list merge sort. The reference implementation's destructive list merge
+-sort will do fewer SET-CDR!s than a destructive insert sort.
+-
+-** Procedure naming and functionality
+-=====================================
+-Almost all of the procedures described below are variants of two basic
+-operations: sorting and merging. These procedures are consistently named
+-by composing a set of basic lexemes to indicate what they do.
+-
+-Lexeme Meaning
+------- -------
+-"sort" The procedure sorts its input data set by some < comparison function.
+-
+-"merge" The procedure merges two ordered data sets into a single ordered
+- result.
+-
+-"stable" This lexeme indicates that the sort is a stable one.
+-
+-"vector" The procedure operates upon vectors.
+-
+-"list" The procedure operates upon lists.
+-
+-"!" Procedures that end in "!" are allowed, and sometimes required,
+- to reuse their input storage to construct their answer.
+-
+-** Types of parameters and return values
+-========================================
+-In the procedures specified below,
+- - A LIS parameter is a list;
+-
+- - A V parameter is a vector;
+-
+- - A < or = parameter is a procedure accepting two arguments taken from the
+- specified procedure's data set(s), and returning a boolean;
+-
+- - START and END parameters are exact, non-negative integers that
+- serve as vector indices selecting a subrange of some associated vector.
+- When specified, they must satisfy the relation
+- 0 <= start <= end <= (vector-length v)
+- where V is the associated vector.
+-
+-Passing values to procedures with these parameters that do not satisfy these
+-types is an error.
+-
+-If a procedure is said to return "unspecified," this means that nothing at all
+-is said about what the procedure returns, not even the number of return
+-values. Such a procedure is not even required to be consistent from call to
+-call in the nature or number of its return values. It is simply required to
+-return a value (or values) that may be passed to a command continuation, e.g.
+-as the value of an expression appearing as a non-terminal subform of a BEGIN
+-expression. Note that in R5RS, this restricts such a procedure to returning a
+-single value; non-R5RS systems may not even provide this restriction.
+-
+-** sort-lib - general sorting package
+-=====================================
+-This library provides basic sorting and merging functionality suitable for
+-general programming. The procedures are named by their semantic properties,
+-i.e., what they do to the data (sort, stable sort, merge, and so forth).
+-
+- Procedure Suggested algorithm
+- -------------------------------------------------------------------------
+- list-sorted? < lis -> boolean
+- list-merge < lis1 lis2 -> list
+- list-merge! < lis1 lis2 -> list
+- list-sort < lis -> list (vector heap or quick)
+- list-sort! < lis -> list (list merge sort)
+- list-stable-sort < lis -> list (vector merge sort)
+- list-stable-sort! < lis -> list (list merge sort)
+- list-delete-neighbor-dups = lis -> list
+- list-delete-neighbor-dups! = lis -> list
+-
+- vector-sorted? < v [start end] -> boolean
+- vector-merge < v1 v2 [start1 end1 start2 end2] -> vector
+- vector-merge! < v v1 v2 [start start1 end1 start2 end2] -> unspecified
+- vector-sort < v [start end] -> vector (heap or quick sort)
+- vector-sort! < v [start end] -> unspecified (heap or quick sort)
+- vector-stable-sort < v [start end] -> vector (vector merge sort)
+- vector-stable-sort! < v [start end] -> unspecified (vector merge sort)
+- vector-delete-neighbor-dups = v [start end] -> vector
+- vector-delete-neighbor-dups! = target source [t-start s-start s-end] -> t-end
+-
+- LIST-SORTED? and VECTOR-SORTED? return true if their input list or vector
+- is in sorted order, as determined by their < comparison parameter.
+-
+- All four merge operations are stable: an element of the initial list LIS1
+- or vector V1 will come before an equal-comparing element in the second
+- list LIS2 or vector V2 in the result.
+-
+- The procedures
+- LIST-MERGE
+- LIST-SORT
+- LIST-STABLE-SORT
+- LIST-DELETE-NEIGHBOR-DUPS
+- do not alter their inputs and are allowed to return a value that shares
+- a common tail with a list argument.
+-
+- The procedures
+- LIST-SORT!
+- LIST-STABLE-SORT!
+- are "linear update" operators -- they are allowed, but not required, to
+- alter the cons cells of their arguments to produce their results.
+-
+- On the other hand, the procedures
+- LIST-DELETE-NEIGHBOR-DUPS!
+- LIST-MERGE!
+- make only a single, iterative, linear-time pass over their argument lists,
+- using SET-CDR!s to rearrange the cells of the lists into the final result
+- -- they work "in place." Hence, any cons cell appearing in the result must
+- have originally appeared in an input. The intent of this
+- iterative-algorithm commitment is to allow the programmer to be sure that
+- if, for example, LIST-MERGE! is asked to merge two ten-million-element
+- lists, the operation will complete without performing some extremely
+- (possibly twenty-million) deep recursion.
+-
+- The vector procedures
+- VECTOR-SORT
+- VECTOR-STABLE-SORT
+- VECTOR-DELETE-NEIGHBOR-DUPS
+- do not alter their inputs, but allocate a fresh vector for their result,
+- of length END - START.
+-
+- The vector procedures
+- VECTOR-SORT!
+- VECTOR-STABLE-SORT!
+- sort their data in-place. (But note that VECTOR-STABLE-SORT! may
+- allocate temporary storage proportional to the size of the input --
+- I am not aware of O(n lg n) stable vector-sorting algorithms that
+- run in constant space.)
+-
+- VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
+-
+- VECTOR-MERGE! writes its result into vector V, beginning at index START,
+- for indices less than END = START + (END1-START1) + (END2-START2). The
+- target subvector
+- V[start,end)
+- may not overlap either source subvector
+- V1[start1,end1)
+- V2[start2,end2).
+-
+- The ...-DELETE-NEIGHBOR-DUPS-... procedures:
+- These procedures delete adjacent duplicate elements from a list or a
+- vector, using a given element-equality procedure. The first/leftmost
+- element of a run of equal elements is the one that survives. The list or
+- vector is not otherwise disordered.
+-
+- These procedures are linear time -- much faster than the O(n^2) general
+- duplicate-element deletors that do not assume any "bunching" of elements
+- (such as the ones provided by SRFI-1). If you want to delete duplicate
+- elements from a large list or vector, you can sort the elements to bring
+- equal items together, then use one of these procedures, for a total time
+- of O(n lg n).
+-
+- The comparison function = passed to these procedures is always applied
+- (= x y)
+- where X comes before Y in the containing list or vector.
+-
+- - LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
+- may share storage with the input list.
+-
+- - VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
+- rather allocates a fresh vector to hold the result.
+-
+- - LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
+- mutate its input list in order to construct its answer.
+-
+- - VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
+- answer, packing its answer into the index range [start,end'), where
+- END' is the non-negative exact integer returned as its value. It
+- returns END' as its result. The vector is not altered outside the range
+- [start,end').
+-
+- - VECTOR-DELETE-NEIGHBOR-DUPS! scans vector SOURCE in range
+- [S-START,S-END), writing its result to vector TARGET beginning at index
+- T-START. It returns exact, non-negative integer T-END, which indicates
+- that the results of the operation are found in index range
+- [T-START,T-END) of TARGET; elements of TARGET outside this range
+- are unaltered.
+-
+- It is an error for memory cell TARGET[T-START] to be a memory cell in
+- the region SOURCE[1 + S-START, S-END). In a Scheme implementation
+- that does not allow distinct vectors to share storage, this means
+- that one of the following must be true:
+- 1. (not (eq? source target))
+- 2. t-start not-in [s-start + 1, s-end)
+-
+- - Examples:
+- (list-delete-neighbor-dups = '(1 1 2 7 7 7 0 -2 -2))
+- => (1 2 7 0 -2)
+-
+- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2))
+- => #(1 2 7 0 -2)
+-
+- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2) 3 7)
+- => #(7 0 -2)
+-
+- ;; Result left in v[3,9):
+- (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
+- (cons (vector-delete-neighbor-dups! = v 3)
+- v))
+- => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
+-
+-
+-** Algorithm-specific sorting packages
+-======================================
+-These packages provide more specific sorting functionality, that is,
+-specific commitment to particular algorithms that have particular
+-pragmatic consequences (such as memory locality, asymptotic running time)
+-beyond their semantic behaviour (sorting, stable sorting, merging, etc.).
+-Programmers that need a particular algorithm can use one of these packages.
+-
+-sorted?-lib - sorted predicates
+- list-sorted? < lis -> boolean
+- vector-sorted? < v [start end] -> boolean
+-
+- Return #f iff there is an adjacent pair ... X Y ... in the input
+- list or vector such that Y < X. The optional START/END range
+- arguments restrict VECTOR-SORTED? to the indicated subvector.
+-
+-list-merge-sort-lib - list merge sort
+- list-merge-sort < lis -> list
+- list-merge-sort! < lis -> list
+- list-merge lis1 < lis2 -> list
+- list-merge! lis1 < lis2 -> list
+-
+- The sort procedures sort their data using a list merge sort, which is
+- stable. (The reference implementation is, additionally, a "natural" sort.
+- See below for the properties of this algorithm.)
+-
+- The ! procedures are destructive -- they use SET-CDR!s to rearrange the
+- cells of the lists into the proper order. As such, they do not allocate
+- any extra cons cells -- they are "in place" sorts. Additionally,
+- LIST-MERGE! is iterative -- it can operate on arguments of arbitrary size
+- with a constant number of stack frames.
+-
+- The merge operations are stable: an element of LIS1 will come before an
+- equal-comparing element in LIS2 in the result list.
+-
+-vector-merge-sort-lib - vector merge sort
+- vector-merge-sort < v [start end temp] -> vector
+- vector-merge-sort! < v [start end temp] -> unspecified
+- vector-merge < v1 v2 [start1 end1 start2 end2] -> vector
+- vector-merge! < v v1 v2 [start start1 end1 start2 end2] -> unspecified
+-
+- The sort procedures sort their data using vector merge sort, which is
+- stable. (The reference implementation is, additionally, a "natural" sort.
+- See below for the properties of this algorithm.)
+-
+- The optional START/END arguments provide for sorting of subranges, and
+- default to 0 and the length of the corresponding vector.
+-
+- Merge-sorting a vector requires the allocation of a temporary "scratch"
+- work vector for the duration of the sort. This scratch vector can be
+- passed in by the client as the optional TEMP argument; if so, the supplied
+- vector must be of size >= END, and will not be altered outside the range
+- [start,end). If not supplied, the sort routines allocate one themselves.
+-
+- The merge operations are stable: an element of V1 will come before an
+- equal-comparing element in V2 in the result vector.
+-
+- VECTOR-MERGE-SORT! leaves its result in V[start,end).
+-
+- VECTOR-MERGE-SORT returns a vector of length END-START.
+-
+- VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
+-
+- VECTOR-MERGE! writes its result into vector V, beginning at index START,
+- for indices less than END = START + (END1-START1) + (END2-START2). The
+- target subvector
+- V[start,end)
+- may not overlap either source subvector
+- V1[start1,end1)
+- V2[start2,end2).
+-
+-vector-heap-sort-lib - vector heap sort
+- heap-sort < v [start end] -> vector
+- heap-sort! < v [start end] -> unspecified
+-
+- These procedures sort their data using heap sort,
+- which is not a stable sorting algorithm.
+-
+- HEAP-SORT returns a vector of length END-START.
+- HEAP-SORT! is in-place, leaving its result in V[start,end).
+-
+-vector-quick-sort-lib - vector quick sort
+- quick-sort < v [start end] -> vector
+- quick-sort! < v [start end] -> unspecified
+- quick-sort3! c v [start end] -> unspecified
+-
+- These procedures sort their data using quick sort,
+- which is not a stable sorting algorithm.
+-
+- QUICK-SORT returns a vector of length END-START.
+- QUICK-SORT! is in-place, leaving its result in V[start,end).
+-
+- QUICK-SORT3! is a variant of quick-sort that takes a three-way
+- comparison function C. C compares a pair of elements and returns
+- an exact integer whose sign indicates their relationship:
+- (c x y) < 0 => x<y
+- (c x y) = 0 => x=y
+- (c x y) > 0 => x>y
+- To help remember the relationship between the sign of the result and
+- the relation, use the function - as the model for C: (- x y) < 0
+- means that x < y; (- x y) > 0 means that x > y.
+-
+- The extra discrimination provided by the three-way comparison can
+- provide significant speedups when sorting data sets with many duplicates,
+- especially when the comparison function is relatively expensive (e.g.,
+- comparing long strings).
+-
+- WARNING: Some sort algorithms, such as insertion sort or heap sort,
+- can tolerate being passed a <= comparison function when they expect a <
+- function -- insertion and merge sort may simply invert stability; and
+- heap sort will run a bit slower, but otherwise produce a correct answer.
+-
+- Quicksort, however, is much more critically sensitive to the distinction
+- between a < and a <= comparison. If QUICK-SORT or QUICK-SORT! expect a <
+- comparison function, and are erroneously given a <= function, they may,
+- depending on implementation, produce an unsorted result, go into an
+- infinite loop, cause a run-time error, occasionally produce a correct
+- result, or do some fifth thing.
+-
+- Implementors may wish to write QUICKSORT3! so that it (a) tests the
+- comparison function (by checking that (c v[start] v[start]) produces
+- false), or (b) is tolerant of an erroneous <= function, or (c) both.
+- Clients of this function, however, should not count on this.
+-
+-vector-insert-sort-lib - vector insertion sort
+- insert-sort < v [start end] -> vector
+- insert-sort! < v [start end] -> unspecified
+-
+- These procedures stably sort their data using insertion sort.
+-
+- INSERT-SORT returns a vector of length END-START.
+- INSERT-SORT! is in-place, leaving its result in V[start,end).
+-
+-delndup-lib - list and vector delete neighbor duplicates
+- list-delete-neighbor-dups = lis -> list
+- list-delete-neighbor-dups! = lis -> list
+-
+- vector-delete-neighbor-dups = v [start end] -> vector
+- vector-delete-neighbor-dups! = v [start end] -> end'
+-
+- These procedures delete adjacent duplicate elements from a list or
+- a vector, using a given element-equality procedure =. The first/leftmost
+- element of a run of equal elements is the one that survives. The list
+- or vector is not otherwise disordered.
+-
+- These procedures are linear time -- much faster than the O(n^2) general
+- duplicate-element deletors that do not assume any "bunching" of elements
+- (such as the ones provided by SRFI-1). If you want to delete duplicate
+- elements from a large list or vector, you can sort the elements to bring
+- equal items together, then use one of these procedures, for a total time
+- of O(n lg n).
+-
+- The comparison function = passed to these procedures is always applied
+- (= x y)
+- where X comes before Y in the containing list or vector.
+-
+- LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
+- may share storage with the input list.
+-
+- VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
+- rather allocates a fresh vector to hold the result.
+-
+- LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
+- mutate its input list in order to construct its answer.
+-
+- VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
+- answer, packing its answer into the index range [start,end'), where
+- END' is the non-negative exact integer returned as its value. It
+- returns END' as its result. The vector is not altered outside the range
+- [start,end').
+-
+- Examples:
+- (list-delete-neighbor-dups = '(1 1 2 7 7 7 0 -2 -2))
+- => (1 2 7 0 -2)
+-
+- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2))
+- => #(1 2 7 0 -2)
+-
+- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2) 3 7)
+- => #(7 0 -2)
+-
+- ;; Result left in v[3,9):
+- (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
+- (cons (vector-delete-neighbor-dups! = v 3)
+- v))
+- => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
+-
+-binsearch-lib - vector binary search lib
+- vector-binary-search elt< elt->key key v [start end] -> integer-or-false
+- vector-binary-search3 c v [start end] -> integer-or-false
+-
+- VECTOR-BINARY-SEARCH searches vector V in range [START,END) (which
+- default to 0 and the length of V, respectively) for an element whose
+- associated key is equal to KEY. The procedure ELT->KEY is used to map
+- an element to its associated key. The elements of the vector are assumed
+- to be ordered by the ELT< relation on these keys. That is,
+- (vector-sorted? (lambda (x y) (elt< (elt->key x) (elt->key y)))
+- v start end) => true
+- An element E of V is a match for KEY if it's neither less nor greater
+- than the key:
+- (and (not (elt< (elt->key e) key))
+- (not (elt< key (elt->key e))))
+- If there is such an element, the procedure returns its index in the
+- vector as an exact integer. If there is no such element in the searched
+- range, the procedure returns false.
+-
+- (vector-binary-search < car 4 '#((1 . one) (3 . three)
+- (4 . four) (25 . twenty-five)))
+- => 2
+-
+- (vector-binary-search < car 7 '#((1 . one) (3 . three)
+- (4 . four) (25 . twenty-five)))
+- => #f
+-
+- VECTOR-BINARY-SEARCH3 is a variant that uses a three-way comparison
+- function C. C compares its parameter to the search key, and returns an
+- exact integer whose sign indicates its relationship to the search key.
+- (c x) < 0 => x < search-key
+- (c x) = 0 => x = search-key
+- (c x) > 0 => x > search-key
+-
+- (vector-binary-search3 (lambda (elt) (- (car elt) 4))
+- '#((1 . one) (3 . three)
+- (4 . four) (25 . twenty-five)))
+- => 2
+-
+- Rationale:
+- - Why isn't VECTOR-BINARY-SEARCH's ELT->KEY computation simply absorbed
+- into the < function? It is separated out because the < function is
+- applied twice inside the binary-search inner loop, once with the search
+- key for the first argument and the element key for the second argument,
+- and once, with the reverse argument order. This is not necessary for
+- VECTOR-BINARY-SEARCH3.
+-
+- - When a comparison operation is able to produce a three-way
+- discrimination, the inner loop of the binary search can trim the number
+- of per-iteration comparisons from an average of 1.5 to a guaranteed
+- single comparison per iteration. This can be a significant savings when
+- searching with an expensive comparison operation (e.g., one that
+- uses string compare, sends email, references a database, or queries
+- a network service such as a web server).
+-
+- - Failure is signaled by false (rather than, say, -1) so that searches
+- can be used in conditional forms such as
+- (or (vector-binary-search ...) ...)
+- or
+- (cond ((vector-binary-search ...) => index-consumer)
+- ...)
+-
+--------------------------------------------------------------------------------
+-* Algorithmic properties
+-------------------------
+-Different sort and merge algorithms have different properties.
+-Choose the algorithm that matches your needs:
+-
+-Vector insert sort
+- Stable, but only suitable for small vectors -- O(n^2).
+-
+-Vector quick sort
+- Not stable. Is fast on average -- O(n lg n) -- but has bad worst-case
+- behaviour. Has good memory locality for big vectors (unlike heap sort).
+- A clever pivot-picking trick (median of three samples) helps avoid
+- worst-case behaviour, but pathological cases can still blow up.
+-
+-Vector heap sort
+- Not stable. Guaranteed fast -- O(n lg n) *worst* case. Poor locality
+- on large vectors. A very reliable workhorse.
+-
+-Vector merge sort
+- Stable. Not in-place -- requires a temporary buffer of equal size.
+- Fast -- O(n lg n) -- and has good memory locality for large vectors.
+-
+- The implementation of vector merge sort provided by this SRFI's reference
+- implementation is, additionally, a "natural" sort, meaning that it
+- exploits existing order in the input data, providing O(n) best case.
+-
+-Destructive list merge sort
+- Stable, fast and in-place (i.e., allocates no new cons cells). "Fast"
+- means O(n lg n) worse-case, and substantially better if the data
+- is already mostly ordered, all the way down to linear time for
+- a completely-ordered input list (i.e., it is a "natural" sort).
+-
+- Note that sorting lists involves chasing pointers through memory, which
+- can be a loser on modern machine architectures because of poor cache &
+- page locality. Pointer *writing*, which is what the SET-CDR!s of a
+- destructive list-sort algorithm do, is even worse, especially if your
+- Scheme has a generational GC -- the writes will thrash the write-barrier.
+- Sorting vectors has inherently better locality.
+-
+- This SRFI's destructive list merge and merge sort implementations are
+- opportunistic -- they avoid redundant SET-CDR!s, and try to take long
+- already-ordered runs of list structure as-is when doing the merges.
+-
+-Pure list merge sort
+- Stable and fast -- O(n lg n) worst-case, and possibly O(n), depending
+- upon the input list (see discussion above).
+-
+-
+-Algorithm Stable? Worst case Average case In-place
+-------------------------------------------------------
+-Vector insert Yes O(n^2) O(n^2) Yes
+-Vector quick No O(n^2) O(n lg n) Yes
+-Vector heap No O(n lg n) O(n lg n) Yes
+-Vector merge Yes O(n lg n) O(n lg n) No
+-List merge Yes O(n lg n) O(n lg n) Either
+-
+-
+--------------------------------------------------------------------------------
+-* Porting and optimisation
+---------------------------
+-This package should be trivial to port. There are only four non-R4RS bits
+-in the code:
+-- Use of multiple-value return, with the R5RS VALUES procedure, and the
+- simple (RECEIVE (var ...) mv-exp body ...) multiple-value binding macro
+- of SRFI-8.
+-
+-- A VECTOR-COPY procedure. This is a tiny little procedure:
+- (vector-copy v [start end])
+-
+-- Use of the LET-OPTIONALS macro from scsh to parse and default optional
+- arguments to three routines. Again, easy to port the macro or rewrite
+- the code to parse, default, and error check the args by hand.
+-
+-- Calls to an ERROR function for complaining about bad arguments.
+-
+-This code is tightly bummed, as far as I can go in portable Scheme.
+-
+-You could speed up the vector code a lot by error-checking the procedure
+-parameters and then shifting over to fixnum-specific arithmetic and dangerous
+-vector-indexing and vector-setting primitives. The comments in the code
+-indicate where the initial error checks would have to be added. There are
+-several (QUOTIENT N 2)'s that could be changed to a fixnum right-shift, as
+-well, in both the list and vector code (SRFI 33 provides such an operator).
+-The code is designed to enable this -- each file usually exports one or two
+-"safe" procedures that end up calling an internal "dangerous" primitive. The
+-little exported cover procedures are where you move the error checks.
+-
+-This should provide *big* speedups. In fact, all the code bumming I've done
+-pretty much disappears in the noise unless you have a good compiler and also
+-can dump the vector-index checks and generic arithmetic -- so I've really just
+-set things up for you to exploit.
+-
+-The optional-arg parsing, defaulting, and error checking is done with a
+-portable R4RS macro. But if your Scheme has a faster mechanism (e.g., Chez),
+-you should definitely port over to it. Note that argument defaulting and
+-error-checking are interleaved -- you don't have to error-check defaulted
+-START/END args to see if they are fixnums that are legal vector indices for
+-the corresponding vector, etc.
+-
+-
+--------------------------------------------------------------------------------
+-* References & Links
+---------------------
+-
+-This document, in HTML:
+- http://srfi.schemers.org/srfi-32/srfi-32.html
+- [This link may not be valid while the SRFI is in draft form.]
+-
+-This document, in simple text format:
+- http://srfi.schemers.org/srfi-32/srfi-32.txt
+-
+-Archive of SRFI-32 discussion-list email:
+- http://srfi.schemers.org/srfi-32/mail-archive/maillist.html
+-
+-SRFI web site:
+- http://srfi.schemers.org/
+-
+-[CommonLisp]
+- Common Lisp: the Language
+- Guy L. Steele Jr. (editor).
+- Digital Press, Maynard, Mass., second edition 1990.
+- Available at http://www.elwood.com/alu/table/references.htm#cltl2
+-
+- The Common Lisp "HyperSpec," produced by Kent Pitman, is essentially
+- the ANSI spec for Common Lisp:
+- http://www.xanalys.com/software_tools/reference/HyperSpec/
+-
+-[R5RS]
+- Revised^5 Report on the Algorithmic Language Scheme,
+- R. Kelsey, W. Clinger, J. Rees (editors).
+- Higher-Order and Symbolic Computation, Vol. 11, No. 1, September, 1998.
+- and ACM SIGPLAN Notices, Vol. 33, No. 9, October, 1998.
+-
+- Available at http://www.schemers.org/Documents/Standards/
+-
+-
+--------------------------------------------------------------------------------
+-* Acknowledgements
+-------------------
+-
+-I thank the authors of the open source I consulted when designing this
+-library, particularly Richard O'Keefe, Donovan Kolby and the MIT Scheme Team.
+-
+-
+--------------------------------------------------------------------------------
+-* Copyright
+------------
+-
+-** SRFI text
+-============
+-This document is copyright (C) Olin Shivers (1998, 1999).
+-All Rights Reserved.
+-
+-This document and translations of it may be copied and furnished to others,
+-and derivative works that comment on or otherwise explain it or assist in its
+-implementation may be prepared, copied, published and distributed, in whole or
+-in part, without restriction of any kind, provided that the above copyright
+-notice and this paragraph are included on all such copies and derivative
+-works. However, this document itself may not be modified in any way, such as
+-by removing the copyright notice or references to the Scheme Request For
+-Implementation process or editors, except as needed for the purpose of
+-developing SRFIs in which case the procedures for copyrights defined in the
+-SRFI process must be followed, or as required to translate it into languages
+-other than English.
+-
+-The limited permissions granted above are perpetual and will not be revoked by
+-the authors or their successors or assigns.
+-
+-This document and the information contained herein is provided on an "AS IS"
+-basis and THE AUTHORS AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR
+-IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
+-INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
+-MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
+-
+-** Reference implementation
+-===========================
+-Short summary: no restrictions.
+-
+-While I wrote all of this code myself, I read a lot of code before I began
+-writing. However, all such code is, itself, either open source or public
+-domain, rendering irrelevant any issue of "copyright taint."
+-
+-The natural merge sorts (pure list, destructive list, and vector) are not only
+-my own code, but are implementations of an algorithm of my own devising. They
+-run in O(n lg n) worst case, O(n) best case, and require only a logarithmic
+-number of stack frames. And they are stable. And the destructive-list variant
+-allocates zero cons cells; it simply rearranges the cells of the input list.
+-
+-Hence the reference implementation is
+- Copyright (c) 1998 by Olin Shivers.
+-and made available under the same copyright as the SRFI text (see above).
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/srfi-32.txt racket-6.12/share/pkgs/srfi-lib/srfi/32/srfi-32.txt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/srfi-32.txt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/32/srfi-32.txt 2018-01-31 12:22:58.152501970 -0500
+@@ -0,0 +1,938 @@
++The SRFI-32 sort libraries -*- outline -*-
++Olin Shivers
++First draft: 1998/10/19
++Last update: 2002/7/21
++
++Emacs should display this document in outline mode. Say c-h m for
++instructions on how to move through it by sections (e.g., c-c c-n, c-c c-p).
++
++* Table of contents
++-------------------
++Abstract
++Procedure index
++Introduction
++What's wrong with the current state of affairs?
++Design rules
++ What vs. how
++ Consistency across function signatures
++ Data parameter first, less-than parameter after
++ Ordering, comparison functions & stability
++ All vector operations accept optional subrange parameters
++ Required vs. allowed side-effects
++Procedure specification
++ Procedure naming and functionality
++ Types of parameters and return values
++ sort-lib - general sorting package
++ Algorithm-specific sorting packages
++Algorithmic properties
++Topics to be resolved during discussion phase
++Porting and optimisation
++References & Links
++Acknowledgements
++Copyright
++
++
++* Abstract
++----------
++Current Scheme sorting packages are, every one of them, surprisingly bad. I've
++designed the API for a full-featured sort toolkit, which I propose as an SRFI.
++The spec comes with 1200 lines of high-quality reference code: tightly
++written, highly commented, portable code, available for free. Implementors
++want this code. It's better than what you have.
++
++-------------------------------------------------------------------------------
++* Procedure index
++-----------------
++list-sorted? vector-sorted?
++
++list-merge vector-merge
++list-sort vector-sort
++list-stable-sort vector-stable-sort
++list-delete-neighbor-dups vector-delete-neighbor-dups
++
++list-merge! vector-merge!
++list-sort! vector-sort!
++list-stable-sort! vector-stable-sort!
++list-delete-neighbor-dups! vector-delete-neighbor-dups!
++
++heap-sort quick-sort insert-sort list-merge-sort vector-merge-sort
++heap-sort! quick-sort! insert-sort! list-merge-sort! vector-merge-sort!
++
++-------------------------------------------------------------------------------
++* Introduction
++--------------
++As I'll detail bewlow, I wasn't very happy with the state of the Scheme
++world for sorting and merging lists and vectors. So I have designed and
++written a fairly comprehensive sorting & merging toolkit. It is
++
++ - very portable,
++
++ - much better code than what is currently in Elk, Gambit, Bigloo,
++ Scheme->C, MzScheme, RScheme, Scheme48, MIT Scheme, or slib, and
++
++ - priced to move: free code.
++
++The package includes
++ - Vector insert sort (stable)
++ - Vector heap sort
++ - Vector quick sort (with median-of-3 pivot picking)
++ - Vector merge sort (stable)
++ - Pure and destructive list merge sort (stable)
++ - Stable vector and list merge
++ - Miscellaneous sort-related procedures: Vector and list merging,
++ sorted? predicates, vector binary search, vector and list
++ delete-equal-neighbor procedures.
++ - A general, non-algorithmic set of procedure names for general sorting
++ and merging.
++
++Scheme programmers may want to adopt this package. I'd like Scheme
++implementors to adopt this code and its API -- in fact, the code is a bribe to
++make it easy for implementors to converge on the suggested API. I mean, you'd
++really have to be a boor to take this free code I wrote and mutate its
++interface over to your incompatible, unportable API, wouldn't you? But you
++could, of course -- it's freely available. More in the spirit of the offering,
++you could make this API available, and then also write a little module
++providing your old interface that is defined in terms of this API. "Scheme
++implementors," in this context, includes slib, which isn't really a standalone
++implementation of Scheme, but is an influential collection of API's and code.
++
++The code is tightly bummed. It is clearly written, and commented in my usual
++voluminous style. This includes notes on porting and implementation-specific
++optimisations.
++
++
++-------------------------------------------------------------------------------
++* What's wrong with the current state of affairs?
++-------------------------------------------------
++
++It's just amazing to me that in 2002, sorting and merging hasn't been
++completely put to bed. These are well-understood algorithms, each of them well
++under a page of code. The straightforward algorithms are basic, core stuff --
++sophomore-level. But if you tour the major Scheme implementations out there on
++the Net, you find badly written code that provides extremely spotty coverage
++of the algorithm space. One implementation even has a buggy implementation
++that has been in use for about 20 years!
++
++Open source-code is a wonderful thing. In a couple of hours, I was able to
++download and check the sources of 9 Scheme systems. Here are my notes from the
++systems I checked. You can skip to the next section if you aren't morbidly
++curious.
++
++slib
++ sorted? vector-or-list <
++ merge list1 list2 <
++ merge! list1 list2 <
++ sort vector-or-list <
++ sort! vector-or-list <
++
++ Richard O'Keefe's stable list merge sort is right idea, but implemented
++ using gratuitous variable side effects. It also does redundant SET-CDR!s.
++ The vector sort converts to list, merge sorts, then reconverts
++ to vector. This is a bad idea -- non-local pointer chasing bad; vector
++ shuffling good.
++
++MIT Scheme
++ sort! vector <
++ merge-sort! vector <
++ quick-sort! vector <
++
++ sort vector-or-list <
++ merge-sort vector-or-list <
++ quick-sort vector-or-list <
++
++ Naive vector quicksort: loser, for worst-case performance reasons.
++ List sort by "list->vector; quicksort; vector->list," hence also loser.
++ A clever stable vector merge sort, albeit not very bummed.
++
++Scheme 48 & T
++ sort-list list <
++ sort-list! list <
++ list-merge! list1 list2 <
++
++ Bob Nix's implementation of online merge-sort, written in the early 80's.
++ Conses unnecessary bookkeeping structure, which isn't necessary with a
++ proper recursive formulation. Also, does redundant SET-CDR!s. No vector
++ sort. Also, has a bug -- is claimed to be a stable sort, but isn't! To see
++ this, get the S48 code, and try
++ (define (my< x y) (< (quotient x 2) (quotient y 2)))
++ (list-merge! (list 0 2) (list 3) my<) ; -> (0 2 3)
++ (list-merge! (list 2) (list 0 3) my<) ; -> (0 3 2)
++ This could be fixed very easily, but it isn't worth it given the
++ other problems with the algorithm.
++
++RScheme
++ vector-sort! vector <
++ sort collection <
++
++ Good basic implementation of vector heapsort, which has O(n lg n)
++ worst-case time. Code ugly, needs tuning. List sort by "list->vector;
++ sort; vector->list", which allocates unneeded temp storage. Nothing
++ for stable sorting.
++
++MzScheme
++ Naive quicksort -- but not available for vector sorting, even
++ though it internally uses a vector. Nothing for stable sorting,
++ and naive quicksort has bad worst-case behaviour.
++
++Bigloo, Scheme->C
++ Couldn't find anything -- but maybe I didn't search for the right
++ thing, since the Bigloo names are French. (I invite correction from
++ the Bigloo implementors.)
++
++Gambit
++ sort-list list <
++ Nothing for vectors. Simple, slow, unstable merge sort for lists.
++
++Elk
++ Another naive quicksort. Lists handled by converting to vector.
++ sort vector-or-list <
++ sort! vector-or-list <
++
++Chez Scheme
++ merge < list1 list2
++ merge! < list1 list2
++ sort < list
++ sort! < list
++
++ These are stable. I have not seen the source code.
++
++Common Lisp
++ sort sequence < [key]
++ stable-sort sequence < [key]
++ merge result-type sequence1 sequence2 < [key]
++
++ The sort procedures are allowed, but not required, to be destructive.
++
++SML/NJ
++ sort: ('a*'a -> bool) -> 'a list -> 'a list
++ "Smooth applicative merge sort," which is stable.
++ There is also a highly bummed quicksort for vectors.
++
++The right solution: Implement a full toolbox of carefully written standard sort
++routines.
++
++Having the source available for all of these above-cited Schemes made
++life a lot easier writing this code. I appreciate the authors making their
++source available under such open terms.
++
++
++-------------------------------------------------------------------------------
++* Design rules
++--------------
++
++** What vs. how
++===============
++There are two different interfaces: "what" (simple) & "how" (detailed).
++
++ - Simple: you specify semantics: datatype (list or vector),
++ mutability, and stability.
++
++ - Detailed: you specify the actual algorithm (quick, heap,
++ insert, merge). Different algorithms have different properties,
++ both semantic & pragmatic, so these exports are necessary.
++
++ It is necessarily the case that the specifications of these procedures
++ make statements about execution "pragmatics." For example, the sole
++ distinction between heap sort and quick sort -- both of which are
++ provided by this library -- is one of execution time, which is not a
++ "semantic" distinction. Similar resource-use statements are made about
++ "iterative" procedures, meaning that they can execute on input of
++ arbitrary size without needing to allocate an unbounded number of stack
++ frames.
++
++** Consistency across function signatures
++=========================================
++The two interfaces share common function signatures wherever
++possible, to facilitate switching a given call from one procedure
++to another.
++
++** Data parameter first, less-than parameter after
++==================================================
++These procedures uniformly observe the following parameter order:
++the data to be sorted come before the the comparison function.
++That is, we write
++ (sort lis <)
++not
++ (sort < lis).
++This is consistent with every single implementation out there, with
++the sole exception of Chez Scheme.
++
++In my opinion, it would be more consistent with other Scheme libraries
++to put the ordering function first -- the "operation currying" convention.
++(E.g., consider FOR-EACH or MAP or FIND.) I decided to leave things as they
++are in favor of near-total backwards compatibility with existing practice.
++
++[Perhaps this should be discussed.]
++
++** Ordering, comparison functions & stability
++=============================================
++These routines take a < comparison function, not a <= comparison
++function, and they sort into increasing order. The difference between
++a < spec and a <= spec comes up in two places:
++ - the definition of an ordered or sorted data set, and
++ - the definition of a stable sorting algorithm.
++
+++ We say that a data set (a list or vector) is *sorted* or *ordered*
++ if it contains no adjacent pair of values ... X Y ... such that Y < X.
++
++ In other words, scanning across the data never takes a "downwards" step.
++
++ If you use a <= procedure where these algorithms expect a <
++ procedure, you may not get the answers you expect. For example,
++ the LIST-SORTED? function will return false if you pass it a <= comparison
++ function and an ordered list containing adjacent equal elements.
++
+++ A "stable" sort is one that preserves the pre-existing order of equal
++ elements. Suppose, for example, that we sort a list of numbers by
++ comparing their absolute values, i.e., using comparison function
++ (lambda (x y) (< (abs x) (abs y)))
++ If we sort a list that contains both 3 and -3:
++ ... 3 ... -3 ...
++ then a stable sort is an algorithm that will not swap the order
++ of these two elements, that is, the answer will look like
++ ... 3 -3 ...
++ not
++ ... -3 3 ...
++
++ Choosing < for the comparison function instead of <= affects how stability
++ is coded. Given an adjacent pair X Y, (< y x) means "Y should be moved in
++ front of X" -- otherwise, leave things as they are. So using a <= function
++ where a < function is expected will *invert* stability.
++
++ This is due to the definition of equality, given a < comparator:
++ (and (not (< x y))
++ (not (< y x)))
++ The definition is rather different, given a <= comparator:
++ (and (<= x y)
++ (<= x y))
++
+++ A "stable" merge is one that reliably favors one of its data sets
++ when equal items appear in both data sets. *All merge operations in
++ this library are stable*, breaking ties between data sets in favor
++ of the first data set -- elements of the first list come before equal
++ elements in the second list.
++
++ So, if we are merging two lists of numbers ordered by absolute value
++ using the stable merge operation LIST-MERGE
++ (list-merge '(0 -2 4 8 -10) '(-1 3 -4 7)
++ (lambda (x y) (< (abs x) (abs y))))
++ reliably places the 4 of the first list before the equal-comparing -4
++ of the second list:
++ (0 -1 -2 4 -4 7 8 -10)
++
++In short, if your comparison function F answers true to (F x x), then
++using a stable sorting or merging algorithm will not give you a stable sort
++or merge, and LIST-SORTED? may surprise you. Note that you can synthesize a <
++function from a <= function with
++ (lambda (x y) (not (<= y x)))
++if need be.
++
++Precise definitions give sharp edges to tools, but require care
++in use. "Measure twice, cut once."
++
++I have adopted the choice of < from Common Lisp. I assume they
++had a good reason for adopting < instead of <=. I'd love to know
++what this reason is; send me email if you can explain it, please.
++
++** All vector operations accept optional subrange parameters
++============================================================
++The vector operations specified below all take optional START/END arguments
++indicating a selected subrange of a vector's elements. If a START parameter or
++START/END parameter pair is given to such a procedure, they must be exact,
++non-negative integers, such that
++ 0 <= START <= END <= (VECTOR-LENGTH V)
++where V is the related vector parameter. If not specified, they default to 0
++and the length of the vector, respectively. They are interpreted to select the
++range [START,END), that is, all elements from index START (inclusive) up to,
++but not including, index END.
++
++** Required vs. allowed side-effects
++====================================
++LIST-SORT! and LIST-STABLE-SORT! are allowed, but not required,
++to alter their arguments' cons cells to construct the result list. This is
++consistent with the what-not-how character of the group of procedures
++to which they belong (the "sort-lib" package).
++
++The LIST-DELETE-NEIGHBOR-DUPS!, LIST-MERGE! and LIST-MERGE-SORT! procedures,
++on the other hand, provide specific algorithms, and, as such, explicitly
++commit to the use of side-effects on their input lists in order to guarantee
++their key algorithmic properties (e.g., linear-time operation, constant-space
++stack use).
++
++-------------------------------------------------------------------------------
++* Procedure specification
++-------------------------
++The procedures are split into several packages. In a Scheme system that has a
++module or package system, these procedures should be contained in modules
++named as follows:
++ Package name Functionality
++ ------------ -------------
++ sort-lib General sorting for lists & vectors
++ sorted?-lib Sorted predicates for lists & vectors
++ list-merge-sort-lib List merge sort
++ vector-merge-sort-lib Vector merge sort
++ vector-heap-sort-lib Vector heap sort
++ vector-quick-sort-lib Vector quick sort
++ vector-insert-sort-lib Vector insertion sort
++ delndup-lib List and vector delete neighbor duplicates
++
++A Scheme system without a module system should provide all of the bindings
++defined in all of these modules as components of the "SRFI-32" package.
++
++Note that there is no list insert sort package, as you might as well always
++use list merge sort. The reference implementation's destructive list merge
++sort will do fewer SET-CDR!s than a destructive insert sort.
++
++** Procedure naming and functionality
++=====================================
++Almost all of the procedures described below are variants of two basic
++operations: sorting and merging. These procedures are consistently named
++by composing a set of basic lexemes to indicate what they do.
++
++Lexeme Meaning
++------ -------
++"sort" The procedure sorts its input data set by some < comparison function.
++
++"merge" The procedure merges two ordered data sets into a single ordered
++ result.
++
++"stable" This lexeme indicates that the sort is a stable one.
++
++"vector" The procedure operates upon vectors.
++
++"list" The procedure operates upon lists.
++
++"!" Procedures that end in "!" are allowed, and sometimes required,
++ to reuse their input storage to construct their answer.
++
++** Types of parameters and return values
++========================================
++In the procedures specified below,
++ - A LIS parameter is a list;
++
++ - A V parameter is a vector;
++
++ - A < or = parameter is a procedure accepting two arguments taken from the
++ specified procedure's data set(s), and returning a boolean;
++
++ - START and END parameters are exact, non-negative integers that
++ serve as vector indices selecting a subrange of some associated vector.
++ When specified, they must satisfy the relation
++ 0 <= start <= end <= (vector-length v)
++ where V is the associated vector.
++
++Passing values to procedures with these parameters that do not satisfy these
++types is an error.
++
++If a procedure is said to return "unspecified," this means that nothing at all
++is said about what the procedure returns, not even the number of return
++values. Such a procedure is not even required to be consistent from call to
++call in the nature or number of its return values. It is simply required to
++return a value (or values) that may be passed to a command continuation, e.g.
++as the value of an expression appearing as a non-terminal subform of a BEGIN
++expression. Note that in R5RS, this restricts such a procedure to returning a
++single value; non-R5RS systems may not even provide this restriction.
++
++** sort-lib - general sorting package
++=====================================
++This library provides basic sorting and merging functionality suitable for
++general programming. The procedures are named by their semantic properties,
++i.e., what they do to the data (sort, stable sort, merge, and so forth).
++
++ Procedure Suggested algorithm
++ -------------------------------------------------------------------------
++ list-sorted? lis < -> boolean
++ list-merge lis1 lis2 < -> list
++ list-merge! lis1 lis2 < -> list
++ list-sort lis < -> list (vector heap or quick)
++ list-sort! lis < -> list (list merge sort)
++ list-stable-sort lis < -> list (vector merge sort)
++ list-stable-sort! lis < -> list (list merge sort)
++ list-delete-neighbor-dups lis = -> list
++ list-delete-neighbor-dups! lis = -> list
++
++ vector-sorted? v < [start end] -> boolean
++ vector-merge v1 v2 < [start1 end1 start2 end2] -> vector
++ vector-merge! v v1 v2 < [start start1 end1 start2 end2] -> unspecific
++ vector-sort v < [start end] -> vector (heap or quick sort)
++ vector-sort! v < [start end] -> unspecific (heap or quick sort)
++ vector-stable-sort v < [start end] -> vector (vector merge sort)
++ vector-stable-sort! v < [start end] -> unspecific (vector merge sort)
++ vector-delete-neighbor-dups v = [start end] -> vector
++ vector-delete-neighbor-dups! v = [start end] -> end'
++
++ LIST-SORTED? and VECTOR-SORTED? return true if their input list or vector
++ is in sorted order, as determined by their < comparison parameter.
++
++ All four merge operations are stable: an element of the initial list LIS1
++ or vector V1 will come before an equal-comparing element in the second
++ list LIS2 or vector V2 in the result.
++
++ The procedures
++ LIST-MERGE
++ LIST-SORT
++ LIST-STABLE-SORT
++ LIST-DELETE-NEIGHBOR-DUPS
++ do not alter their inputs and are allowed to return a value that shares
++ a common tail with a list argument.
++
++ The procedures
++ LIST-SORT!
++ LIST-STABLE-SORT!
++ are "linear update" operators -- they are allowed, but not required, to
++ alter the cons cells of their arguments to produce their results.
++
++ On the other hand, the procedures
++ LIST-DELETE-NEIGHBOR-DUPS!
++ LIST-MERGE!
++ make only a single, iterative, linear-time pass over their argument lists,
++ using SET-CDR!s to rearrange the cells of the lists into the final result
++ -- they work "in place." Hence, any cons cell appearing in the result must
++ have originally appeared in an input. The intent of this
++ iterative-algorithm commitment is to allow the programmer to be sure that
++ if, for example, LIST-MERGE! is asked to merge two ten-million-element
++ lists, the operation will complete without performing some extremely
++ (possibly twenty-million) deep recursion.
++
++ The vector procedures
++ VECTOR-SORT
++ VECTOR-STABLE-SORT
++ VECTOR-DELETE-NEIGHBOR-DUPS
++ do not alter their inputs, but allocate a fresh vector for their result,
++ of length END-START.
++
++ The vector procedures
++ VECTOR-SORT!
++ VECTOR-STABLE-SORT!
++ sort their data in-place. (But note that VECTOR-STABLE-SORT! may
++ allocate temporary storage proportional to the size of the input --
++ I am not aware of O(n lg n) stable vector sorting algorithms that
++ run in constant space.)
++
++ VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
++
++ VECTOR-MERGE! writes its result into vector V, beginning at index START0,
++ for indices less than END0 = START0 + (END1-START1) + (END2-START2). The
++ target subvector
++ V[start0,end0)
++ may not overlap either source subvector
++ V1[start1,end1)
++ V2[start2,end2).
++
++ The DELETE-NEIGHBOR-DUP-... procedures:
++ These procedures delete adjacent duplicate elements from a list or a
++ vector, using a given element-equality procedure. The first/leftmost
++ element of a run of equal elements is the one that survives. The list or
++ vector is not otherwise disordered.
++
++ These procedures are linear time -- much faster than the O(n^2) general
++ duplicate-element deletors that do not assume any "bunching" of elements
++ (such as the ones provided by SRFI-1). If you want to delete duplicate
++ elements from a large list or vector, sort the elements to bring equal
++ items together, then use one of these procedures, for a total time of
++ O(n lg n).
++
++ The comparison function = passed to these procedures is always applied
++ (= x y)
++ where X comes before Y in the containing list or vector.
++
++ - LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
++ may share storage with the input list.
++
++ - VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
++ rather allocates a fresh vector to hold the result.
++
++ - LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
++ mutate its input list in order to construct its answer.
++
++ - VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
++ answer, packing its answer into the index range [start,end'), where
++ END' is the non-negative exact integer returned as its value. It
++ returns END' as its result. The vector is not altered outside the range
++ [start,end').
++
++ [Maybe this procedure should take a "target" vector to write?]
++
++ - Examples:
++ (list-delete-neighbor-dups '(1 1 2 7 7 7 0 -2 -2) =)
++ => (1 2 7 0 -2)
++
++ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) =)
++ => #(1 2 7 0 -2)
++
++ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) = 3 7)
++ => #(7 0 -2)
++
++ ;; Result left in v[3,9):
++ (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
++ (cons (vector-delete-neighbor-dups! v = 3)
++ v))
++ => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
++
++
++** Algorithm-specific sorting packages
++======================================
++These packages provide more specific sorting functionality, that is,
++specific committment to particular algorithms that have particular
++pragmatic consequences (such as memory locality, asymptotic running time)
++beyond their semantic behaviour (sorting, stable sorting, merging, etc.).
++Programmers that need a particular algorithm can use one of these packages.
++
++sorted?-lib - sorted predicates
++ list-sorted? lis < -> boolean
++ vector-sorted? v < [start end] -> boolean
++
++ Return #f iff there is an adjacent pair ... X Y ... in the input
++ list or vector such that Y < X. The optional START/END range
++ arguments restrict VECTOR-SORTED? to the indicated subvector.
++
++list-merge-sort-lib - list merge sort
++ list-merge-sort lis < -> list
++ list-merge-sort! lis < -> list
++ list-merge lis1 lis2 < -> list
++ list-merge! lis1 lis2 < -> list
++
++ The sort procedures sort their data using a list merge sort, which is
++ stable. (The reference implementation is, additionally, a "natural" sort.
++ See below for the properties of this algorithm.)
++
++ The ! procedures are destructive -- they use SET-CDR!s to rearrange the
++ cells of the lists into the proper order. As such, they do not allocate
++ any extra cons cells -- they are "in place" sorts. Additionally,
++ LIST-MERGE! is iterative, not recursive -- it can operate on arguments of
++ arbitrary size without requiring an unbounded amount of stack space.
++
++ The merge operations are stable: an element of LIS1 will come before an
++ equal-comparing element in LIS2 in the result list.
++
++vector-merge-sort-lib - vector merge sort
++ vector-merge-sort v < [start end temp] -> vector
++ vector-merge-sort! v < [start end temp] -> unspecific
++ vector-merge v1 v2 < [start1 end1 start2 end2] -> vector
++ vector-merge! v v1 v2 < [start0 start1 end1 start2 end2] -> unspecific
++
++ The sort procedures sort their data using vector merge sort, which is
++ stable. (The reference implementation is, additionally, a "natural" sort.
++ See below for the properties of this algorithm.)
++
++ The optional START/END arguments provide for sorting of subranges, and
++ default to 0 and the length of the corresponding vector.
++
++ Merge-sorting a vector requires the allocation of a temporary "scratch"
++ work vector for the duration of the sort. This scratch vector can be
++ passed in by the client as the optional TEMP argument; if so, the supplied
++ vector must be of size >= END, and will not be altered outside the range
++ [start,end). If not supplied, the sort routines allocate one themselves.
++
++ The merge operations are stable: an element of V1 will come before an
++ equal-comparing element in V2 in the result vector.
++
++ VECTOR-MERGE-SORT! leaves its result in V[start,end).
++
++ VECTOR-MERGE-SORT returns a vector of length END-START.
++
++ VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
++
++ VECTOR-MERGE! writes its result into vector V, beginning at index START0,
++ for indices less than END0 = START0 + (END1-START1) + (END2-START2). The
++ target subvector
++ V[start0,end0)
++ may not overlap either source subvector
++ V1[start1,end1)
++ V2[start2,end2).
++
++vector-heap-sort-lib - vector heap sort
++ heap-sort v < [start end] -> vector
++ heap-sort! v < [start end] -> unspecific
++
++ These procedures sort their data using heap sort,
++ which is not a stable sorting algorithm.
++
++ HEAP-SORT returns a vector of length END-START.
++ HEAP-SORT! is in-place, leaving its result in V[start,end).
++
++vector-quick-sort-lib - vector quick sort
++ quick-sort v < [start end] -> vector
++ quick-sort! v < [start end] -> unspecific
++
++ These procedures sort their data using quick sort,
++ which is not a stable sorting algorithm.
++
++ QUICK-SORT returns a vector of length END-START.
++ QUICK-SORT! is in-place, leaving its result in V[start,end).
++
++vector-insert-sort-lib - vector insertion sort
++ insert-sort v < [start end] -> vector
++ insert-sort! v < [start end] -> unspecific
++
++ These procedures stably sort their data using insertion sort.
++
++ INSERT-SORT returns a vector of length END-START.
++ INSERT-SORT! is in-place, leaving its result in V[start,end).
++
++delndup-lib - list and vector delete neighbor duplicates
++ list-delete-neighbor-dups lis = -> list
++ list-delete-neighbor-dups! lis = -> list
++
++ vector-delete-neighbor-dups v = [start end] -> vector
++ vector-delete-neighbor-dups! v = [start end] -> end'
++
++ These procedures delete adjacent duplicate elements from a list or
++ a vector, using a given element-equality procedure =. The first/leftmost
++ element of a run of equal elements is the one that survives. The list
++ or vector is not otherwise disordered.
++
++ These procedures are linear time -- much faster than the O(n^2) general
++ duplicate-element deletors that do not assume any "bunching" of elements
++ (such as the ones provided by SRFI-1). If you want to delete duplicate
++ elements from a large list or vector, sort the elements to bring equal
++ items together, then use one of these procedures, for a total time of
++ O(n lg n).
++
++ The comparison function = passed to these procedures is always applied
++ (= x y)
++ where X comes before Y in the containing list or vector.
++
++ LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
++ may share storage with the input list.
++
++ VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
++ rather allocates a fresh vector to hold the result.
++
++ LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
++ mutate its input list in order to construct its answer.
++
++ VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
++ answer, packing its answer into the index range [start,end'), where
++ END' is the non-negative exact integer returned as its value. It
++ returns END' as its result. The vector is not altered outside the range
++ [start,end').
++
++ Examples:
++ (list-delete-neighbor-dups '(1 1 2 7 7 7 0 -2 -2) =)
++ => (1 2 7 0 -2)
++
++ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) =)
++ => #(1 2 7 0 -2)
++
++ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) = 3 7)
++ => #(7 0 -2)
++
++ ;; Result left in v[3,9):
++ (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
++ (cons (vector-delete-neighbor-dups! v = 3)
++ v))
++ => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
++
++
++-------------------------------------------------------------------------------
++* Algorithmic properties
++------------------------
++Different sort and merge algorithms have different properties.
++Choose the algorithm that matches your needs:
++
++Vector insert sort
++ Stable, but only suitable for small vectors -- O(n^2).
++
++Vector quick sort
++ Not stable. Is fast on average -- O(n lg n) -- but has bad worst-case
++ behaviour. Has good memory locality for big vectors (unlike heap sort).
++ A clever pivot-picking trick (median of three samples) helps avoid
++ worst-case behaviour, but pathological cases can still blow up.
++
++Vector heap sort
++ Not stable. Guaranteed fast -- O(n lg n) *worst* case. Poor locality
++ on large vectors. A very reliable workhorse.
++
++Vector merge sort
++ Stable. Not in-place -- requires a temporary buffer of equal size.
++ Fast -- O(n lg n) -- and has good memory locality for large vectors.
++
++ The implementation of vector merge sort provided by this SRFI's reference
++ implementation is, additionally, a "natural" sort, meaning that it
++ exploits existing order in the input data, providing O(n) best case.
++
++Destructive list merge sort
++ Stable, fast and in-place (i.e., allocates no new cons cells). "Fast"
++ means O(n lg n) worse-case, and substantially better if the data
++ is already mostly ordered, all the way down to linear time for
++ a completely-ordered input list (i.e., it is a "natural" sort).
++
++ Note that sorting lists involves chasing pointers through memory, which
++ can be a loser on modern machine architectures because of poor cache &
++ page locality. Pointer *writing*, which is what the SET-CDR!s of a
++ destructive list-sort algorithm do, is even worse, especially if your
++ Scheme has a generational GC -- the writes will thrash the write-barrier.
++ Sorting vectors has inherently better locality.
++
++ This SRFIs destructive list merge and merge sort implementations are
++ opportunistic -- they avoid redundant SET-CDR!s, and try to take long
++ already-ordered runs of list structure as-is when doing the merges.
++
++Pure list merge sort
++ Stable and fast -- O(n lg n) worst-case, and possibly better, depending
++ upon the input list (see above).
++
++
++Algorithm Stable? Worst case Average case In-place
++------------------------------------------------------
++V insert Yes O(n^2) O(n^2) Yes
++V quick No O(n^2) O(n lg n) Yes
++V heap No O(n lg n) O(n lg n) Yes
++V merge Yes O(n lg n) O(n lg n) No
++L merge Yes O(n lg n) O(n lg n) Either
++
++
++-------------------------------------------------------------------------------
++* Topics to be resolved during discussion phase
++-----------------------------------------------
++I particularly solicit comments about the following topics.
++
++- Include VECTOR-BINARY-SEARCH ?
++ Should we include
++ (VECTOR-BINARY-SEARCH v key< elt->key key [start end])
++ in the SRFI? It sort of goes with sorting; it's exactly ten lines of code.
++
++- Comparison function before or after the list/vector argument?
++ Should it be
++ (list-sort < lis)
++ or
++ (list-sort lis <)
++ There is overwhelming consistency among the implementations: data first,
++ < after. Only Chez does it differently.
++
++ I have done it in the backwards-compatible way. But I prefer the < first,
++ data after way.
++
++
++-------------------------------------------------------------------------------
++* Porting and optimisation
++--------------------------
++This package should be trivial to port. There are only four non-R4RS bits
++in the code:
++- Use of multiple-value return, with the R5RS VALUES procedure, and the
++ simple (RECEIVE (var ...) mv-exp body ...) multiple-value binding macro.
++
++- A VECTOR-COPY procedure. This is a tiny little procedure:
++ (vector-copy v [start end])
++
++- Use of the LET-OPTIONALS macro from scsh to parse and default optional
++ arguments to three routines. Again, easy to port the macro or rewrite
++ the code to parse, default, and error check the args by hand.
++
++- Calls to an ERROR function for complaining about bad arguments.
++
++This code is tightly bummed, as far as I can go in portable Scheme.
++
++You could speed up the vector code a lot by error-checking the procedure
++parameters and then shifting over to fixnum-specific arithmetic and
++dangerous vector-indexing and vector-setting primitives. The comments
++in the code indicate where the initial error checks would have to be
++added. There are several (QUOTIENT N 2)'s that could be changed to a
++fixnum right-shift, as well, in both the list and vector code. The code
++is designed to enable this -- each file usually exports one or two "safe"
++procedures that end up calling an internal "dangerous" primitive. The
++little exported cover procedures are where you move the error checks.
++
++This should provide *big* speedups. In fact, all the code bumming I've done
++pretty much disappears in the noise unless you have a good compiler and also
++can dump the vector-index checks and generic arithmetic -- so I've really just
++set things up for you to exploit.
++
++The optional-arg parsing, defaulting, and error checking is done with a
++portable R4RS macro. But if your Scheme has a faster mechanism (e.g., Chez),
++you should definitely port over to it. Note that argument defaulting and
++error-checking are interleaved -- you don't have to error-check defaulted
++START/END args to see if they are fixnums that are legal vector indices for
++the corresponding vector, etc.
++
++
++-------------------------------------------------------------------------------
++* References & Links
++--------------------
++
++This document, in HTML:
++ http://srfi.schemers.org/srfi-32/srfi-32.html
++ [This link may not be valid while the SRFI is in draft form.]
++
++This document, in simple text format:
++ http://srfi.schemers.org/srfi-32/srfi-32.txt
++
++Archive of SRFI-32 discussion-list email:
++ http://srfi.schemers.org/srfi-32/mail-archive/maillist.html
++
++SRFI web site:
++ http://srfi.schemers.org/
++
++[CommonLisp]
++ Common Lisp: the Language
++ Guy L. Steele Jr. (editor).
++ Digital Press, Maynard, Mass., second edition 1990.
++ Available at http://www.elwood.com/alu/table/references.htm#cltl2
++
++ The Common Lisp "HyperSpec," produced by Kent Pitman, is essentially
++ the ANSI spec for Common Lisp:
++ http://www.xanalys.com/software_tools/reference/HyperSpec/
++
++[R5RS]
++ Revised^5 Report on the Algorithmic Language Scheme,
++ R. Kelsey, W. Clinger, J. Rees (editors).
++ Higher-Order and Symbolic Computation, Vol. 11, No. 1, September, 1998.
++ and ACM SIGPLAN Notices, Vol. 33, No. 9, October, 1998.
++
++ Available at http://www.schemers.org/Documents/Standards/
++
++
++-------------------------------------------------------------------------------
++* Acknowledgements
++------------------
++
++I thank the authors of the open source I consulted when designing this
++library, particularly Richard O'Keefe, Donovan Kolby and the MIT Scheme Team.
++
++
++-------------------------------------------------------------------------------
++* Copyright
++-----------
++
++** SRFI text
++============
++This document is copyright (C) Olin Shivers (1998, 1999).
++All Rights Reserved.
++
++Permission is hereby granted, free of charge, to any person obtaining
++a copy of this software and associated documentation files (the
++"Software"), to deal in the Software without restriction, including
++without limitation the rights to use, copy, modify, merge, publish,
++distribute, sublicense, and/or sell copies of the Software, and to
++permit persons to whom the Software is furnished to do so, subject to
++the following conditions:
++
++The above copyright notice and this permission notice shall be
++included in all copies or substantial portions of the Software.
++
++THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
++EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
++MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
++NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
++LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
++OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
++WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
++
++** Reference implementation
++===========================
++Short summary: no restrictions.
++
++While I wrote all of this code myself, I read a lot of code before I began
++writing. However, all such code is, itself, either open source or public
++domain, rendering irrelevant any issue of "copyright taint."
++
++The natural merge sorts (pure list, destructive list, and vector) are not only
++my own code, but are implementations of an algorithm of my own devising. They
++run in O(n lg n) worst case, O(n) best case, and require only a logarithmic
++number of stack frames. And they are stable. And the destructive-list variant
++allocates zero cons cells; it simply rearranges the cells of the input list.
++
++Hence the reference implementation is
++ Copyright (c) 1998 by Olin Shivers.
++and made available under the same copyright as the SRFI text (see above).
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib/srfi/%3a5/let.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5/let.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-#lang s-exp srfi/provider srfi/%3a5
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib/srfi/%3a5.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang s-exp srfi/provider srfi/5
+-
+-;; FIXME: "rest arguments" need to generate mutable lists
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib/srfi/5/let.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/5/let.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,86 +0,0 @@
+-;;;
+-;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
+-;;; Time-stamp: <03/04/08 09:56:06 solsona>
+-;;;
+-;;; Usually, I would add a copyright notice, and the announce that
+-;;; this code is under the LGPL licence. Nevertheless, I only did the
+-;;; port to PLT Scheme v200, and here is the copyright notice,
+-;;; comments, and licence from the original source:
+-;;;
+-;;; Copyright (C) Andy Gaynor (1999-2003)
+-;;;
+-;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
+-;;; Renamed to s:let by Eli Barzilay
+-
+-#lang scheme/base
+-(provide s:let)
+-
+-(define-syntax s:let
+- (syntax-rules ()
+- ;; standard
+- ((s:let () body ...)
+- (let () body ...))
+- ((s:let ((var val) ...) body ...)
+- (let ((var val) ...) body ...))
+-
+- ;; rest style
+- ((s:let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; signature style
+- ((s:let (name bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ;; standard named style
+- ((s:let name (bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ))
+-
+-;; A loop to walk down the list of bindings.
+-
+-(define-syntax let-loop
+- (syntax-rules ()
+-
+- ;; No more bindings - make a LETREC.
+- ((let-loop name () (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ...) . body)))
+- name)
+- vals ...))
+-
+- ;; Rest binding, no name
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (let ((var val) ... (rest-var (list rest-val ...))) . body))
+-
+- ;; Process a (var val) pair.
+- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
+- (let-loop name (more ...) (vars ... var) (vals ... val) body))
+-
+- ;; End with a rest variable - make a LETREC.
+- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ... . rest-var) . body)))
+- name)
+- vals ... rest-vals ...))))
+-
+-;; Four loops - normal and `signature-style', each with and without a rest
+-;; binding.
+-;;
+-;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let fibonacci ((n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib/srfi/5.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/5.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,2 +0,0 @@
+-;; module loader for SRFI-5
+-#lang s-exp srfi/provider srfi/5/let #:unprefix s:
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/info.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1,14 @@
++#lang info
++
++(define collection 'multi)
++
++(define deps '("scheme-lib"
++ "base"
++ "r6rs-lib"
++ "srfi-lib"
++ "compatibility-lib"))
++
++
++(define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1 @@
++#lang s-exp srfi/provider srfi/%3a5
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1,3 @@
++#lang s-exp srfi/provider srfi/5
++
++;; FIXME: "rest arguments" need to generate mutable lists
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1,86 @@
++;;;
++;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
++;;; Time-stamp: <03/04/08 09:56:06 solsona>
++;;;
++;;; Usually, I would add a copyright notice, and the announce that
++;;; this code is under the LGPL licence. Nevertheless, I only did the
++;;; port to PLT Scheme v200, and here is the copyright notice,
++;;; comments, and licence from the original source:
++;;;
++;;; Copyright (C) Andy Gaynor (1999-2003)
++;;;
++;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
++;;; Renamed to s:let by Eli Barzilay
++
++#lang scheme/base
++(provide s:let)
++
++(define-syntax s:let
++ (syntax-rules ()
++ ;; standard
++ ((s:let () body ...)
++ (let () body ...))
++ ((s:let ((var val) ...) body ...)
++ (let ((var val) ...) body ...))
++
++ ;; rest style
++ ((s:let ((var val) . bindings) body ...)
++ (let-loop #f bindings (var) (val) (body ...)))
++
++ ;; signature style
++ ((s:let (name bindings ...) body ...)
++ (let-loop name (bindings ...) () () (body ...)))
++
++ ;; standard named style
++ ((s:let name (bindings ...) body ...)
++ (let-loop name (bindings ...) () () (body ...)))
++
++ ))
++
++;; A loop to walk down the list of bindings.
++
++(define-syntax let-loop
++ (syntax-rules ()
++
++ ;; No more bindings - make a LETREC.
++ ((let-loop name () (vars ...) (vals ...) body)
++ ((letrec ((name (lambda (vars ...) . body)))
++ name)
++ vals ...))
++
++ ;; Rest binding, no name
++ ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
++ (let ((var val) ... (rest-var (list rest-val ...))) . body))
++
++ ;; Process a (var val) pair.
++ ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
++ (let-loop name (more ...) (vars ... var) (vals ... val) body))
++
++ ;; End with a rest variable - make a LETREC.
++ ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
++ ((letrec ((name (lambda (vars ... . rest-var) . body)))
++ name)
++ vals ... rest-vals ...))))
++
++;; Four loops - normal and `signature-style', each with and without a rest
++;; binding.
++;;
++;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
++;; (if (= i n)
++;; f0
++;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
++;;
++;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
++;; (if (= i n)
++;; f0
++;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
++;;
++;;(let fibonacci ((n 10) (i 0) . (f 0 1))
++;; (if (= i n)
++;; (car f)
++;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
++;;
++;;(let (fibonacci (n 10) (i 0) . (f 0 1))
++;; (if (= i n)
++;; (car f)
++;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1,2 @@
++;; module loader for SRFI-5
++#lang s-exp srfi/provider srfi/5/let #:unprefix s:
+diff -urN racket-6.12-orig/share/pkgs/srfi-lite-lib/info.rkt racket-6.12/share/pkgs/srfi-lite-lib/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lite-lib/info.rkt 2018-01-26 16:10:04.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lite-lib/info.rkt 2018-01-31 12:22:58.160501940 -0500
+@@ -1 +1,9 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("base"))) (define pkg-desc "implementation of the most widely used \"srfi\" libraries") (define pkg-authors (quote (mflatt)))))
++#lang info
++
++(define collection 'multi)
++
++(define deps '("base"))
++
++(define pkg-desc "implementation of the most widely used \"srfi\" libraries")
++
++(define pkg-authors '(mflatt))
+Binary files racket-6.12-orig/share/pkgs/srfi-lite-lib/.info.rkt.swp and racket-6.12/share/pkgs/srfi-lite-lib/.info.rkt.swp differ
+diff -urN racket-6.12-orig/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt racket-6.12/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt 2018-01-31 12:22:58.162501933 -0500
+@@ -2,6 +2,7 @@
+
+ (require racket/contract/base
+ racket/file
++ racket/list
+ (only-in racket/runtime-path define-runtime-path)
+ racket/string racket/format
+ syntax/modread
+@@ -113,10 +114,6 @@
+ (lambda ()
+ (with-input-from-file path read))))
+ #t))))
+- (define (rdc ls)
+- (if (null? (cdr ls))
+- '()
+- (cons (car ls) (rdc (cdr ls)))))
+
+ ;;Retrieve a localized template given its package name and a template name
+ (define (localized-template package-name template-name)
+@@ -127,5 +124,5 @@
+ (let ((bundle (hash-ref *localization-bundles* specifier #f)))
+ (cond ((and bundle (assq template-name bundle)) => cdr)
+ ((null? (cdr specifier)) #f)
+- (else (loop (rdc specifier))))))))
++ (else (loop (drop-right specifier 1))))))))
+ )
diff --git a/racket.spec b/racket.spec
new file mode 100644
index 0000000..d3dc907
--- /dev/null
+++ b/racket.spec
@@ -0,0 +1,369 @@
+Name: racket
+Version: 6.12
+Release: 8%{?dist}
+Summary: General purpose programming language
+
+License: GPLv3 and LGPLv3 and MIT
+URL: https://racket-lang.org
+Source0: https://mirror.racket-lang.org/installers/%{version}/%{name}-%{version}-src…
+
+# Update SRFI libraries to include upstream PR 5.
+# See: https://github.com/racket/srfi/pull/5 (merged)
+Patch0: racket-6.12-update-srfi.patch
+
+# Remove SRFI library and docs with restrictive licensing.
+# See: https://github.com/racket/srfi/issues/4 (open)
+# Note: Upstream maintainers have confirmed this
+# is safe, since the removed components are
+# extra elements which nothing else in the
+# package depends on.
+Patch1: racket-6.12-remove-nonfree.patch
+
+# Backport upstream fix of rpaths in racket/racket
+# which break check-buildroot
+# https://github.com/racket/racket/pull/1947 (merged)
+Patch2: racket-6.12-fix-rpaths.patch
+
+# Backport upstream fix of rpaths in racket/web-server
+# which break check-buildroot
+# https://github.com/racket/web-server/issues/36 (closed)
+Patch3: racket-6.12-fix-webserver-rpaths.patch
+
+# Update racket-doc/search to use https://docs.racket-lang.org
+# when no docs exist locally
+# https://github.com/racket/scribble/pull/164 (open)
+Patch4: racket-6.12-fix-doc-open-url.patch
+
+# Issue Building for s390x and armv7hl in koji
+ExcludeArch: s390x armv7hl
+
+# To compile the program
+BuildRequires: gcc
+
+# To fix rpath issue with executables.
+BuildRequires: chrpath
+
+# Racket heavily utilizes the system ffi library.
+BuildRequires: libffi-devel
+
+# For the racket/gui library (via libffi)
+# https://github.com/racket/gui/blob/master/gui-lib/mred/private/wx/gtk/gtk3.…
+BuildRequires: gtk3
+
+# For the racket/draw library (via libffi)
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/cair…
+BuildRequires: cairo
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/pang…
+BuildRequires: pango
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/png.…
+BuildRequires: libpng
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/jpeg…
+BuildRequires: libjpeg-turbo
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/glib…
+BuildRequires: glib2
+
+# To validate desktop file
+BuildRequires: desktop-file-utils
+
+BuildRequires: git
+
+# Require the subpackages
+Requires: racket-minimal%{?_isa} = %{version}-%{release}
+Requires: racket-pkgs = %{version}-%{release}
+
+%description
+Racket is a general-purpose programming language as well as
+the world's first ecosystem for developing and deploying new
+languages. Make your dream language, or use one of the dozens
+already available.
+
+%prep
+%autosetup -v -p1
+# Remove bundled libffi
+rm -r src/foreign/libffi
+
+%build
+cd src
+%configure \
+ --enable-pthread \
+ --enable-shared \
+ --enable-libffi \
+ --disable-strip
+%make_build
+
+%install
+cd src
+%make_install
+
+# Delete mred binaries and replace them with links.
+rm -vf ${RPM_BUILD_ROOT}%{_bindir}/mred
+rm -vf ${RPM_BUILD_ROOT}%{_bindir}/mred-text
+ln -vs %{_bindir}/gracket ${RPM_BUILD_ROOT}%{_bindir}/mred
+ln -vs %{_bindir}/gracket-text ${RPM_BUILD_ROOT}%{_bindir}/mred-text
+
+# Delete static library. Apperently --disable-libs does not stop it.
+rm -vf ${RPM_BUILD_ROOT}%{_libdir}/libracket3m.a
+
+# Delete duplicate license files
+rm -rf %{buildroot}%{_datadir}/racket/COPYING*txt
+
+# Fix the rpath error.
+chrpath --delete ${RPM_BUILD_ROOT}%{_bindir}/racket
+chrpath --delete ${RPM_BUILD_ROOT}%{_libdir}/racket/gracket
+
+# Remove the libtool files.
+rm -f ${RPM_BUILD_ROOT}%{_libdir}/*.la
+
+# Fix paths in the desktop files.
+sed -i "s#${RPM_BUILD_ROOT}##g" \
+ ${RPM_BUILD_ROOT}/%{_datadir}/applications/*.desktop
+
+# Validate desktop files
+desktop-file-validate %{buildroot}/%{_datadir}/applications/*.desktop
+
+# Fix paths in html docs
+DOCS_TO_FIX="
+syntax/module-helpers.html
+rackunit/api.html
+reference/collects.html"
+for i in $DOCS_TO_FIX; do
+ sed -i "s#${RPM_BUILD_ROOT}##g" \
+ ${RPM_BUILD_ROOT}/%{_datadir}/doc/racket/$i
+done
+
+chmod -x ${RPM_BUILD_ROOT}%{_libdir}/racket/starter-sh
+
+%ldconfig_scriptlets
+
+# Equivalent to upstream's minimal-racket release
+%package minimal
+Summary: A minimal Racket installation
+Requires: racket-collects = %{version}-%{release}
+%description minimal
+Racket's core runtime
+
+%package collects
+Summary: Racket's core collections libraries
+BuildArch: noarch
+%description collects
+Libraries providing Racket's core functionality
+
+# Arch independent source and bytecode files
+%package pkgs
+Summary: Racket package collections
+# See BuildRequires section for details on dependencies
+Requires: gtk3
+Requires: cairo
+Requires: pango
+Requires: libpng
+Requires: glib2
+Requires: libjpeg-turbo
+Requires: racket-minimal = %{version}-%{release}
+BuildArch: noarch
+%description pkgs
+Additional packages and libraries for Racket
+
+# Development headers and links
+%package devel
+Summary: Development files for Racket
+Requires: racket-minimal%{?_isa} = %{version}-%{release}
+%description devel
+Files needed to link against Racket.
+
+# HTML documentation
+%package doc
+Summary: Documentation files for Racket
+BuildArch: noarch
+%description doc
+A local installation of the Racket documentation system.
+
+%files
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_bindir}/drracket
+%{_bindir}/gracket
+%{_bindir}/gracket-text
+%{_bindir}/mred-text
+%{_bindir}/mred
+%{_bindir}/mzc
+%{_bindir}/mzpp
+%{_bindir}/mzscheme
+%{_bindir}/mztext
+%{_bindir}/pdf-slatex
+%{_bindir}/plt-games
+%{_bindir}/plt-help
+%{_bindir}/plt-r5rs
+%{_bindir}/plt-r6rs
+%{_bindir}/plt-web-server
+%{_bindir}/scribble
+%{_bindir}/setup-plt
+%{_bindir}/slatex
+%{_bindir}/slideshow
+%{_bindir}/swindle
+%{_datadir}/applications/
+
+%files collects
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_datadir}/racket/collects
+
+%files minimal
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_bindir}/racket
+%{_bindir}/raco
+%{_libdir}/racket
+%{_libdir}/libracket3m-%{version}.so
+%{_datadir}/racket/links.rktd
+%{_datadir}/racket/pkgs/racket-lib
+%{_datadir}/man/man1/racket*
+%{_datadir}/man/man1/raco*
+%dir %{_datadir}/racket
+%dir %{_sysconfdir}/racket/
+%config %{_sysconfdir}/racket/config.rktd
+%exclude %{_libdir}/libracket3m.so
+
+%files pkgs
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_datadir}/racket
+%{_datadir}/man/man1/drracket*
+%{_datadir}/man/man1/gracket*
+%{_datadir}/man/man1/mred*
+%{_datadir}/man/man1/mzc*
+%{_datadir}/man/man1/mzscheme*
+%{_datadir}/man/man1/plt-help*
+%{_datadir}/man/man1/setup-plt*
+%exclude %{_datadir}/racket/links.rktd
+%exclude %dir %{_datadir}/racket/pkgs/racket-lib
+%exclude %dir %{_datadir}/racket/collects
+
+%files devel
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_includedir}/racket
+%{_libdir}/libracket3m.so
+
+%files doc
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_datadir}/doc/racket
+
+%changelog
+* Mon Jul 30 2018 David Benoit <dbenoit(a)redhat.com> 6.12-8
+- Annotate dependencies with links to source code
+- Move dependencies to racket-pkgs, since they are only used
+ by that subpackage
+- Update mred symbolic links
+- Fix ownership of directories
+- Remove executable bit from starter-sh
+
+* Thu Jul 12 2018 David Benoit <dbenoit(a)redhat.com> 6.12-7
+- Remove hardened build since it is enabled by default
+- Add gcc to BuildRequires
+- Remove wildcards from directory listings in files section
+
+* Fri Apr 13 2018 David Benoit <dbenoit(a)redhat.com> 6.12-6
+- Remove license wildcard and add license field to each subpackage
+
+* Fri Apr 6 2018 David Benoit <dbenoit(a)redhat.com> 6.12-5
+- remove update-database post scripts
+- move libracket3m.so link into -devel
+- add ldconfig_scriptlets after install
+- remove disable debug_package and configure
+ with --disable-strip instead
+- add license to files section and update
+ license header field
+- validate desktop files
+- change ownership of /etc/racket
+- update changelog with release info
+- use specific man directory man/man1/*
+- refactor racket into subpackages
+ racket-minimal, racket-collects, and racket-pkgs
+
+* Wed Apr 4 2018 David Benoit <dbenoit(a)redhat.com> 6.12-4
+- noarch -docs subpackage
+
+* Tue Mar 20 2018 David Benoit <dbenoit(a)redhat.com> 6.12-3
+- fix text encoding issue in description section
+- remove doc-open-url scriptlets
+- add scriptlet to fix paths in html docs
+- add patch2 to backport rpaths fix in compiled .zo files
+- add patch3 to backport rpaths fix in web-server-lib
+- add patch4 to configure doc open url dynamically at runtime
+- remove override of __arch_install_post to allow full
+ checking of buildroot.
+
+* Thu Feb 1 2018 David Benoit <dbenoit(a)redhat.com> - 6.12-2
+- Fix duplication of object files
+- Add version to racket-devel requirements
+- Remove base package as a dependency of racket-doc
+- Remove Groups tag
+
+* Wed Jan 31 2018 David Benoit <dbenoit(a)redhat.com> - 6.12-1
+- Update to current stable version
+- Add patch0 to update SRFIs to latest upstream
+- Add patch1 to remove nonfree SRFI components
+
+* Thu Oct 26 2017 David Benoit <dbenoit(a)redhat.com> - 6.10-1
+- Update to current stable version
+- Remove libedit readline patch (fixed upstream)
+- Break docs into separate package
+- Add scriptlets to set doc-open-url based on
+ whether docs are installed
+- Exclude armv7hl and s390x as target arches
+- Update description to match website
+- Change URLs to use https instead of http
+
+
+* Thu Jul 6 2017 David Benoit <dbenoit(a)redhat.com> - 6.9-1
+- Update to current stable version
+- Patch libedit readline error
+- Remove ExclusiveArch to test all builds in koji
+
+* Fri Jan 22 2016 Brandon Thomas <bthomaszx(a)gmail.com> - 6.3-1
+- Update to current stable version.
+- Updated description to match website.
+- Removed build requirement "racket-packaging".
+- Updated to gtk+3.
+- Let Autoprovides determine provides.
+- Debuginfo package is empty and preventing the package from building.
+- Removed uneeded file copies.
+- Remove possible extra static library.
+
+* Sun Dec 14 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.1.6-1
+- Update to current snapshot to fix match hash-table expander.
+
+* Mon Dec 01 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.1-1
+- Update to current stable version.
+
+* Fri Sep 05 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.5-4
+- Use racket-packaging to capture module dependencies.
+
+* Tue Aug 19 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.5-1
+- Updated to 6.1.0.5
+- Merged the -doc package back in.
+
+* Fri Aug 08 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.4-2
+- Do not alter .zo files, prevent check-buildroot from being run instead.
+
+* Thu Aug 07 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.4-1
+- Updated to 6.1.0.4
+- Split-off -doc package.
+
+* Fri Jul 25 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.3-4
+- Updated to 6.1.0.3
+- Dropped the unnecessary static library.
+- Dropped mred programs to enable debug package.
+
+* Sat Jun 22 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.5-1
+- Changed to use 5.3.5 version of Racket.
+- Created static package for developers who may need static libraries.
+- Added RPM optimization options to CFLAGS for build.
+- Added macro to use SMP build options in make.
+
+* Thu May 16 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.4-1
+- Changed to use 5.3.4 version of Racket.
+
+* Tue May 14 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.3-3
+- Moved documentation to /usr/doc directory.
+
+* Mon May 13 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.3-2
+- Remove bundled libffi from racket before building program.
+
+* Thu May 9 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.3-1
+- Initial Revision.
1
0
05 Sep '18
The package rpms/racket.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/racket.git/commit/?id=f241dafce9084…
https://src.fedoraproject.org/cgit/rpms/racket.git/commit/?id=b8569a44d31ac….
Change:
+ExcludeArch: s390x armv7hl
+%ifarch %{ix86}
Thanks.
Full change:
============
commit b8569a44d31acfcec836302dff9ae178d3957c0b
Author: dbenoit <dbenoit(a)redhat.com>
Date: Wed Sep 5 15:37:42 2018 -0400
fix i686 and remove ppc64
diff --git a/racket.spec b/racket.spec
index 5fb06c9..bb31acd 100644
--- a/racket.spec
+++ b/racket.spec
@@ -1,6 +1,6 @@
Name: racket
Version: 7.0
-Release: 1%{?dist}
+Release: 2%{?dist}
Summary: General purpose programming language
License: GPLv3 and LGPLv3 and MIT
@@ -21,7 +21,8 @@ Patch0: racket-7.0-remove-nonfree.patch
Patch1: racket-7.0-fix-doc-open-url.patch
# Issue Building for s390x and armv7hl in koji
-ExcludeArch: s390x armv7hl
+#ExcludeArch: s390x armv7hl
+ExcludeArch: s390x armv7hl ppc64
# To compile the program
BuildRequires: gcc
@@ -70,11 +71,20 @@ rm -r src/foreign/libffi
%build
cd src
+
+# Disable SSE on i686 until fixed upstream
+# https://github.com/racket/racket/issues/2245
+%ifarch %{ix86}
+ %set_build_flags
+ export CFLAGS=$(echo $CFLAGS | sed -e "s/-mfpmath=sse *//")
+%endif
+
%configure \
--enable-pthread \
--enable-shared \
--enable-libffi \
--disable-strip
+
%make_build
%install
@@ -231,7 +241,13 @@ A local installation of the Racket documentation system.
%{_datadir}/doc/racket
%changelog
-* Fri Aug 3 2018 David Benoit <dbenoit(a)redhat.com> 7.0-1
+* Wed Sep 5 2018 David Benoit <dbenoit(a)redhat.com> 7.0-2
+- Disable SSE math on i686 until issue is fixed upstream
+- Exclude ppc due to issue building Racket v7.0 and
+ arch being deprecated in next release
+
+* Fri Aug 17 2018 David Benoit <dbenoit(a)redhat.com> 7.0-1
+- Update sources to Racket v7.0
- Remove 6.12 patches and add update remove nonfree
srfi patch to 7.0
commit ee22cb0e92daba659fd9d9fd4855d952e21cc53f
Author: dbenoit <dbenoit(a)redhat.com>
Date: Fri Aug 31 15:39:51 2018 -0400
update package to v7.0
diff --git a/.gitignore b/.gitignore
index d013960..4fd1ad5 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1 +1 @@
-/racket-6.12-src.tgz
+/racket-7.0-src.tgz
diff --git a/racket-6.12-fix-doc-open-url.patch b/racket-6.12-fix-doc-open-url.patch
deleted file mode 100644
index 7c497f5..0000000
--- a/racket-6.12-fix-doc-open-url.patch
+++ /dev/null
@@ -1,69 +0,0 @@
-diff -urN racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt racket-6.12/share/pkgs/scribble-lib/help/search.rkt
---- racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:14:56.017273013 -0400
-+++ racket-6.12/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:15:37.839113929 -0400
-@@ -17,33 +17,38 @@
- #:notify [notify void])
- (define open-url (get-doc-open-url))
- (cond
-- [open-url
-- (define dest-url (let ([u (string->url open-url)])
-- (combine-url/relative
-- u
-- (string-join
-- (for/list ([s (explode-path sub)])
-- (if (path? s)
-- (path-element->string s)
-- (format "~a" s)))
-- "/"))))
-- (notify (url->string dest-url))
-- (send-url (url->string
-- (struct-copy url dest-url
-- [fragment (or fragment
-- (url-fragment dest-url))]
-- [query (append
-- (url-query dest-url)
-- (if query
-- (url-query
-- (string->url
-- (format "q?~a" query)))
-- null))])))]
-- [else
-- (let* ([path (build-path (find-user-doc-dir) sub)]
-- [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
-- (notify path)
-- (send-url/file path #:fragment fragment #:query query))]))
-+ [open-url
-+ (define dest-url (let ([u (string->url open-url)])
-+ (combine-url/relative
-+ u
-+ (string-join
-+ (for/list ([s (explode-path sub)])
-+ (if (path? s)
-+ (path-element->string s)
-+ (format "~a" s)))
-+ "/"))))
-+ (notify (url->string dest-url))
-+ (send-url (url->string
-+ (struct-copy url dest-url
-+ [fragment (or fragment
-+ (url-fragment dest-url))]
-+ [query (append
-+ (url-query dest-url)
-+ (if query
-+ (url-query
-+ (string->url
-+ (format "q?~a" query)))
-+ null))])))]
-+ [else
-+ (let* ([path (build-path (find-user-doc-dir) sub)]
-+ [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
-+ (notify path)
-+ (if (file-exists? path)
-+ (send-url/file path #:fragment fragment #:query query)
-+ (let ([part (lambda (pfx x) (if x (string-append pfx x) ""))])
-+ (send-url (string-append
-+ "https://docs.racket-lang.org/"
-+ sub (part "#" fragment) (part "?" query))))))]))
-
- ;; This is an example of changing this code to use the online manuals.
- ;; Normally, it's better to set `doc-open-url` in "etc/config.rktd",
diff --git a/racket-6.12-fix-rpaths.patch b/racket-6.12-fix-rpaths.patch
deleted file mode 100644
index f76730e..0000000
--- a/racket-6.12-fix-rpaths.patch
+++ /dev/null
@@ -1,13 +0,0 @@
-diff -urN racket-6.12-f/collects/racket/runtime-path.rkt racket-6.12/collects/racket/runtime-path.rkt
---- racket-6.12-f/collects/racket/runtime-path.rkt 2018-02-06 16:04:29.969878308 -0500
-+++ racket-6.12/collects/racket/runtime-path.rkt 2018-02-06 16:07:10.549176282 -0500
-@@ -186,7 +186,8 @@
- (path-of
- #,(datum->syntax
- #'orig-stx
-- `(,#'this-expression-source-file))))
-+ `(,#'this-expression-source-file)
-+ #'orig-stx)))
- #'void)])
- (apply to-values (resolve-paths (#%variable-reference)
- get-dir
diff --git a/racket-6.12-fix-webserver-rpaths.patch b/racket-6.12-fix-webserver-rpaths.patch
deleted file mode 100644
index 78614cf..0000000
--- a/racket-6.12-fix-webserver-rpaths.patch
+++ /dev/null
@@ -1,35 +0,0 @@
-diff -urN racket-6.12-rp/share/pkgs/web-server-lib/web-server/formlets/lib.rkt racket-6.12-wbsvr/share/pkgs/web-server-lib/web-server/formlets/lib.rkt
---- racket-6.12-rp/share/pkgs/web-server-lib/web-server/formlets/lib.rkt 2018-02-22 18:42:06.131455756 -0500
-+++ racket-6.12-wbsvr/share/pkgs/web-server-lib/web-server/formlets/lib.rkt 2018-03-20 15:07:09.666222028 -0400
-@@ -7,6 +7,7 @@
- racket/function
- racket/serialize
- syntax/location
-+ setup/collects
- (for-syntax racket/base
- syntax/parse))
-
-@@ -70,20 +71,20 @@
- #:range-contracts (map (curry coerce-contract 'formlet/c)
- contracts)))]))
- (define quote-this-module-path
-- (quote-module-path))
-+ (path->collects-relative (quote-module-path)))
- (define-syntax formlet/c
- (syntax-parser
- [(_ range ...)
- #:declare range (expr/c #'contract?
- #:name "range contract argument")
- #'(formlet/c** (-> listof-binding
-- (values (coerce-contract 'formlet/c range.c) ...)))]
-+ (values (coerce-contract 'formlet/c range #;.c) ...)))]
- [name:id
- #`(contract
- (-> contract? (... ...) contract?)
- dynamic-formlet/c
- quote-this-module-path
-- (quote-module-path)
-+ (path->collects-relative (quote-module-path))
- "formlet/c"
- #'name)]))
-
diff --git a/racket-6.12-remove-nonfree.patch b/racket-6.12-remove-nonfree.patch
deleted file mode 100644
index 4e72a3e..0000000
--- a/racket-6.12-remove-nonfree.patch
+++ /dev/null
@@ -1,1124 +0,0 @@
-diff -urN racket-6.12-nf/share/links.rktd racket-6.12/share/links.rktd
---- racket-6.12-nf/share/links.rktd 2018-01-31 13:04:06.757393819 -0500
-+++ racket-6.12/share/links.rktd 2018-01-31 13:04:25.419309681 -0500
-@@ -133,7 +133,6 @@
- (root "pkgs/r5rs-doc")
- (root "pkgs/srfi-lite-lib")
- (root "pkgs/srfi-lib")
-- (root "pkgs/srfi-lib-nonfree")
- (root "pkgs/html-doc")
- (root "pkgs/images-gui-lib")
- (root "pkgs/images-doc")
-@@ -177,7 +176,6 @@
- (root "pkgs/slideshow-exe")
- (root "pkgs/slideshow-plugin")
- (root "pkgs/srfi-doc")
-- (root "pkgs/srfi-doc-nonfree")
- (root "pkgs/syntax-color-doc")
- (root "pkgs/web-server-lib")
- (root "pkgs/unix-socket-lib")
-diff -urN racket-6.12-nf/share/pkgs/srfi/info.rkt racket-6.12/share/pkgs/srfi/info.rkt
---- racket-6.12-nf/share/pkgs/srfi/info.rkt 2018-01-31 13:04:07.216391749 -0500
-+++ racket-6.12/share/pkgs/srfi/info.rkt 2018-01-31 13:05:32.601025227 -0500
-@@ -3,13 +3,9 @@
- (define collection 'multi)
-
- (define deps '("srfi-lib"
-- "srfi-lib-nonfree"
-- "srfi-doc"
-- "srfi-doc-nonfree"))
-+ "srfi-doc"))
- (define implies '("srfi-lib"
-- "srfi-lib-nonfree"
-- "srfi-doc"
-- "srfi-foc-nonfree"))
-+ "srfi-doc"))
-
- (define pkg-desc "Legacy SRFI (Scheme) libraries")
-
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/info.rkt 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,19 +0,0 @@
--#lang info
--
--(define collection 'multi)
--
--(define build-deps '("mzscheme-doc"
-- "scheme-lib"
-- "base"
-- "scribble-lib"
-- "srfi-doc"
-- "srfi-lib-nonfree"
-- "racket-doc"
-- "r5rs-doc"
-- "r6rs-doc"
-- "compatibility-lib"))
--(define update-implies '("srfi-lib-nonfree"))
--
--(define pkg-desc "documentation part of \"srfi nonfree\"")
--
--(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,3 +0,0 @@
--#lang info
--
--(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
-@@ -1,66 +0,0 @@
--#lang scribble/doc
--@(require srfi/scribblings/util
-- scribble/manual
-- scribble/eval
-- scriblib/render-cond
-- scribble/core
-- scribble/html-properties
-- (for-syntax scheme/base)
-- (for-label scheme/base
-- racket/stream))
--
--@; ----------------------------------------------------------------------
--
--@title{SRFI Nonfree Libraries and Documentation}
--
--The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
--Implementation} (a.k.a. @deftech{SRFI}) process allows individual
--members of the Scheme community to propose libraries and extensions to
--be supported by multiple Scheme implementations.
--
--Racket is distributed with implementations of many SRFIs, most of
--which can be implemented as libraries. To import the bindings of SRFI
--@math{n}, use
--
--@racketblock[
--(require @#,elem{@racketidfont{srfi/}@math{n}})
--]
--
--This document lists the SRFIs that are supported by Racket and
--provides a link to the original SRFI specification (which is also
--distributed as part of Racket's documentation).
--
--The following SRFI specification documents are licensed restrictively.
--
--@table-of-contents[]
--
--
--@; ----------------------------------------
--
--@srfi[5]{A compatible let form with signatures and rest arguments}
--
--@redirect[5 '(
-- (let #t "unnamed")
--)]
--
--Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
--
--@; ----------------------------------------
--
--@srfi[29]{Localization}
--
--@redirect[29 '(
-- (current-language #f "current-language")
-- (current-country #f "current-country")
-- (current-locale-details #f "current-locale-details")
-- (declare-bundle! #f "declare-bundle!")
-- (store-bundle #f "store-bundle")
-- (load-bundle! #f "load-bundle!")
-- (localized-template #f "localized-template")
--)]
--
--Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
--
--@; ----------------------------------------
--
--@index-section[]
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
-@@ -1,507 +0,0 @@
--<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
--<html>
-- <head>
-- <meta name="generator" content="HTML Tidy, see www.w3.org">
-- <title>SRFI 29: Localization</title>
-- <meta name="author" content="Scott G. Miller">
-- <meta name="description" content="Localization">
-- </head>
-- <body>
-- <H1>Title</H1>
--
-- SRFI 29: Localization
--
-- <H1>Author</H1>
--
-- Scott G. Miller
--
-- <H1>Abstract</H1>
--
-- This document specifies an interface to retrieving and
-- displaying locale sensitive messages. A Scheme program can
-- register one or more translations of templated messages, and
-- then write Scheme code that can transparently retrieve the
-- appropriate message for the locale under which the Scheme
-- system is running. <br>
--
--
-- <H1>Rationale</H1>
--
-- <p>As any programmer that has ever had to deal with making his
-- or her code readable in more than one locale, the process of
-- sufficiently abstracting program messages from their
-- presentation to the user is non-trivial without help from the
-- programming language. Most modern programming language
-- libraries do provide some mechanism for performing this
-- separation.</p>
--
-- <p>A portable API that allows a piece of code to run without
-- modification in different countries and under different
-- languages is a must for any non-trivial software project.
-- The interface should separate the logic of a program from
-- the myriad of translations that may be necessary.</p>
--
-- <p>The interface described in this document provides such
-- functionality. The underlying implementation is also allowed to
-- use whatever datastructures it likes to provide access to the
-- translations in the most efficient manner possible. In
-- addition, the implementation is provided with standardized
-- functions that programs will use for accessing an external,
-- unspecified repository of translations.</p>
--
-- <p>This interface <i>does not</i> cover all aspects of
-- localization, including support for non-latin characters,
-- number and date formatting, etc. Such functionality is the
-- scope of a future SRFI that may extend this one.</p>
--
-- <H1>Dependencies</H1>
--
-- An SRFI-29 conformant implementation must also implement
-- SRFI-28, Basic Format Strings. Message templates are strings
-- that must be processed by the <tt>format</tt> function
-- specified in that SRFI.
--
-- <H1>Specification</H1>
--
-- <h3>Message Bundles</h3>
--
-- <p>A Message Bundle is a set of message templates and their
-- identifying keys. Each bundle contains one or more such
-- key/value pairs. The bundle itself is associated with a
-- <i>bundle specifier</i> which uniquely identifies the
-- bundle.</p>
--
-- <h3>Bundle Specifiers</h3>
--
-- <p>A Bundle Specifier is a Scheme list that describes, in order
-- of importance, the package and locale of a message bundle.
-- In most cases, a locale specifier will have between one
-- and three elements. The first element is a symbol denoting the
-- package for which this bundle applies. The second and third
-- elements denote a <i>locale</i>. The second element (first
-- element of the locale) if present, is the two letter, ISO 639-1
-- language code for the bundle. The third element, if present, is
-- a two letter ISO 3166-1 country code. In some cases, a
-- fourth element may be present, specifying the encoding used for
-- the bundle. All bundle specifier elements are Scheme
-- symbols.</p>
--
-- <p>If only one translation is provided, it should be designated
-- only by a package name, for example <tt>(mathlib)</tt>. This
-- translation is called the <i>default</i> translation.</p>
--
-- <h3>Bundle Searching</h3>
--
-- <p>When a message template is retrieved from a bundle, the
-- Scheme implementation will provide the locale under which the
-- system is currently running. When the template is retrieved,
-- the package name will be specified. The Scheme system should
-- construct a Bundle Specifier from the provided package name and
-- the active locale. For example, when retrieving a message
-- template for French Canadian, in the <tt>mathlib</tt> package,
-- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
-- program may also retrieve the elements of the current locale
-- using the no-argument procedures:</p>
--
-- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
-- <i>symbol</i></tt><br>
-- <tt><b>current-language</b> <i>symbol</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- When given no arguments, returns the current ISO 639-1
-- language code as a symbol. If provided with an
-- argument, the current language is set to that named by the
-- symbol for the currently executing Scheme thread (or for the
-- entire Scheme system if such a distinction is not possible).
--
-- </blockquote>
--
-- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
-- <i>symbol</i></tt><br>
-- <tt><b>current-country</b> <i>symbol</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- returns the current ISO 3166-1 country code as a symbol.
-- If provided with an argument, the current country is
-- set to that named by the symbol for the currently executing
-- Scheme thread (or for the entire Scheme system if such a
-- distinction is not possible).
-- </blockquote>
--
-- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
-- symbol</i></tt>s<br>
-- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- Returns a list of additional locale details as a list of
-- symbols. This list may contain information about
-- encodings or other more specific information. If
-- provided with an argument, the current locale details are set
-- to those given in the currently executing Scheme thread (or
-- for the entire Scheme system if such a distinction is not
-- possible).
-- </blockquote>
--
-- <p>The Scheme System should first check for a bundle with the
-- exact name provided. If no such bundle is found, the last
-- element from the list is removed and a search is tried for a
-- bundle with that name. If no bundle is then found, the list is
-- shortened by removing the last element again. If no message is
-- found and the bundle specifier is now the empty list, an error
-- should be raised.</p>
--
-- <p>The reason for this search order is to provide the most
-- locale sensitive template possible, but to fall back on more
-- general templates if a translation has not yet been provided
-- for the given locale.</p>
--
-- <h3>Message Templates</h3>
--
-- <p>A message template is a localized message that may or may
-- not contain one of a number of formatting codes. A message
-- template is a Scheme string. The string is of a form that can
-- be processed by the <tt>format</tt> procedure found in many
-- Scheme systems and formally specified in SRFI-28 (Basic Format
-- Strings).</p>
--
-- <p>This SRFI also extends SRFI-28 to provide an additional
-- <tt>format</tt> escape code:</p>
--
-- <blockquote>
-- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
-- follows this code immediately to reference the [N]'th
-- optional value absolutely, rather than the next unconsumed
-- value. The referenced value is <i>not</i> consumed.
-- </blockquote>
-- This extension allows optional values to be positionally
-- referenced, so that message templates can be constructed that
-- can produce the proper word ordering for a language.
--
-- <h3>Preparing Bundles</h3>
-- Before a bundle may be used by the Scheme system to retrieve
-- localized template messages, they must be made available to the
-- Scheme system. This SRFI specifies a way to portably
-- define the bundles, as well as store them in and retrieve them
-- from an unspecified system which may be provided by resources
-- outside the Scheme system.<br>
--
--
-- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
-- association-list</i> -> undefined<br>
-- </tt></p>
--
-- <blockquote>
-- Declares a new bundle named by the given bundle-specifier.
-- The contents of the bundle are defined by the provided
-- association list. The list contains associations
-- between Scheme symbols and the message templates (Scheme
-- strings) they name. If a bundle already exists with the
-- given name, it is overwritten with the newly declared
-- bundle.<br>
-- </blockquote>
-- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
-- boolean</tt><br>
--
--
-- <blockquote>
-- Attempts to store a bundle named by the given bundle
-- specifier, and previously made available using
-- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
-- unspecified mechanism that may be persistent across Scheme
-- system restarts. If successful, a non-false value is
-- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
-- </blockquote>
-- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
-- boolean</tt><br>
--
--
-- <blockquote>
-- Attempts to retrieve a bundle from an unspecified mechanism
-- which stores bundles outside the Scheme system. If the
-- bundle was retrieved successfully, the function returns a
-- non-false value, and the bundle is immediately available to
-- the Scheme system. If the bundle could not be found or loaded
-- successfully, the function returns <tt>#f</tt>, and the
-- Scheme system's bundle registry remains unaffected.<br>
-- </blockquote>
-- A compliant Scheme system may choose not to provide any
-- external mechanism to store localized bundles. If it does
-- not, it must still provide implementations for
-- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
-- case, both functions must return <tt>#f</tt> regardless of the
-- arguments given. Users of this SRFI should recognize that the
-- inability to load or store a localized bundle in an external
-- repository is <i>not</i> a fatal error.<br>
--
--
-- <h3>Retrieving Localized Message Templates</h3>
--
-- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
-- message-template-name</tt></i> <tt>-> <i>string or #f<br>
-- </i></tt></p>
--
-- <blockquote>
-- Retrieves a localized message template for the given package
-- name and the given message template name (both symbols).
-- If no such message could be found, false (#f) is
-- returned.<br>
-- <br>
-- </blockquote>
-- After retrieving a template, the calling program can use
-- <tt>format</tt> to produce a string that can be displayed to
-- the user.<br>
--
--
-- <h2>Examples</h2>
-- The below example makes use of SRFI-29 to display simple,
-- localized messages. It also defines its bundles in such a
-- way that the Scheme system may store and retrieve the bundles
-- from a more efficient system catalog, if available.<br>
--
--<pre>
--(let ((translations
-- '(((en) . ((time . "Its ~a, ~a.")
-- (goodbye . "Goodbye, ~a.")))
-- ((fr) . ((time . "~1@*~a, c'est ~a.")
-- (goodbye . "Au revoir, ~a."))))))
-- (for-each (lambda (translation)
-- (let ((bundle-name (cons 'hello-program (car translation))))
-- (if (not (load-bundle! bundle-name))
-- (begin
-- (declare-bundle! bundle-name (cdr translation))
-- (store-bundle! bundle-name)))))
-- translations))
--
--(define localized-message
-- (lambda (message-name . args)
-- (apply format (cons (localized-template 'hello-program
-- message-name)
-- args))))
--
--(let ((myname "Fred"))
-- (display (localized-message 'time "12:00" myname))
-- (display #\newline)
--
-- (display (localized-message 'goodbye myname))
-- (display #\newline))
--
--;; Displays (English):
--;; Its 12:00, Fred.
--;; Goodbye, Fred.
--;;
--;; French:
--;; Fred, c'est 12:00.
--;; Au revoir, Fred.
--</pre>
--
-- <H1>Implementation</H1>
--
-- <p>The implementation requires that the Scheme system provide a
-- definition for <tt>current-language</tt> and
-- <tt>current-country</tt> capable of distinguishing the correct
-- locale present during a Scheme session. The definitions of
-- those functions in the reference implementation are not capable
-- of that distinction. Their implementation is provided only so
-- that the following code can run in any R4RS scheme system.
-- <br>
-- </p>
--
-- <p>In addition, the below implementation of a compliant
-- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
-- SRFI-23 (Error reporting)</p>
--<pre>
--;; The association list in which bundles will be stored
--(define *localization-bundles* '())
--
--;; The current-language and current-country functions provided
--;; here must be rewritten for each Scheme system to default to the
--;; actual locale of the session
--(define current-language
-- (let ((current-language-value 'en))
-- (lambda args
-- (if (null? args)
-- current-language-value
-- (set! current-language-value (car args))))))
--
--(define current-country
-- (let ((current-country-value 'us))
-- (lambda args
-- (if (null? args)
-- current-country-value
-- (set! current-country-value (car args))))))
--
--;; The load-bundle! and store-bundle! both return #f in this
--;; reference implementation. A compliant implementation need
--;; not rewrite these procedures.
--(define load-bundle!
-- (lambda (bundle-specifier)
-- #f))
--
--(define store-bundle!
-- (lambda (bundle-specifier)
-- #f))
--
--;; Declare a bundle of templates with a given bundle specifier
--(define declare-bundle!
-- (letrec ((remove-old-bundle
-- (lambda (specifier bundle)
-- (cond ((null? bundle) '())
-- ((equal? (caar bundle) specifier)
-- (cdr bundle))
-- (else (cons (car bundle)
-- (remove-old-bundle specifier
-- (cdr bundle))))))))
-- (lambda (bundle-specifier bundle-assoc-list)
-- (set! *localization-bundles*
-- (cons (cons bundle-specifier bundle-assoc-list)
-- (remove-old-bundle bundle-specifier
-- *localization-bundles*))))))
--
--;;Retrieve a localized template given its package name and a template name
--(define localized-template
-- (letrec ((rdc
-- (lambda (ls)
-- (if (null? (cdr ls))
-- '()
-- (cons (car ls) (rdc (cdr ls))))))
-- (find-bundle
-- (lambda (specifier template-name)
-- (cond ((assoc specifier *localization-bundles*) =>
-- (lambda (bundle) bundle))
-- ((null? specifier) #f)
-- (else (find-bundle (rdc specifier)
-- template-name))))))
-- (lambda (package-name template-name)
-- (let loop ((specifier (cons package-name
-- (list (current-language)
-- (current-country)))))
-- (and (not (null? specifier))
-- (let ((bundle (find-bundle specifier template-name)))
-- (and bundle
-- (cond ((assq template-name bundle) => cdr)
-- ((null? (cdr specifier)) #f)
-- (else (loop (rdc specifier)))))))))))
--
--;;An SRFI-28 and SRFI-29 compliant version of format. It requires
--;;SRFI-23 for error reporting.
--(define format
-- (lambda (format-string . objects)
-- (let ((buffer (open-output-string)))
-- (let loop ((format-list (string->list format-string))
-- (objects objects)
-- (object-override #f))
-- (cond ((null? format-list) (get-output-string buffer))
-- ((char=? (car format-list) #\~)
-- (cond ((null? (cdr format-list))
-- (error 'format "Incomplete escape sequence"))
-- ((char-numeric? (cadr format-list))
-- (let posloop ((fl (cddr format-list))
-- (pos (string->number
-- (string (cadr format-list)))))
-- (cond ((null? fl)
-- (error 'format "Incomplete escape sequence"))
-- ((and (eq? (car fl) '#\@)
-- (null? (cdr fl)))
-- (error 'format "Incomplete escape sequence"))
-- ((and (eq? (car fl) '#\@)
-- (eq? (cadr fl) '#\*))
-- (loop (cddr fl) objects (list-ref objects pos)))
-- (else
-- (posloop (cdr fl)
-- (+ (* 10 pos)
-- (string->number
-- (string (car fl)))))))))
-- (else
-- (case (cadr format-list)
-- ((#\a)
-- (cond (object-override
-- (begin
-- (display object-override buffer)
-- (loop (cddr format-list) objects #f)))
-- ((null? objects)
-- (error 'format "No value for escape sequence"))
-- (else
-- (begin
-- (display (car objects) buffer)
-- (loop (cddr format-list)
-- (cdr objects) #f)))))
-- ((#\s)
-- (cond (object-override
-- (begin
-- (display object-override buffer)
-- (loop (cddr format-list) objects #f)))
-- ((null? objects)
-- (error 'format "No value for escape sequence"))
-- (else
-- (begin
-- (write (car objects) buffer)
-- (loop (cddr format-list)
-- (cdr objects) #f)))))
-- ((#\%)
-- (if object-override
-- (error 'format "Escape sequence following positional override does not require a value"))
-- (display #\newline buffer)
-- (loop (cddr format-list) objects #f))
-- ((#\~)
-- (if object-override
-- (error 'format "Escape sequence following positional override does not require a value"))
-- (display #\~ buffer)
-- (loop (cddr format-list) objects #f))
-- (else
-- (error 'format "Unrecognized escape sequence"))))))
-- (else (display (car format-list) buffer)
-- (loop (cdr format-list) objects #f)))))))
--
--</pre>
--
-- <H1>Copyright</H1>
--
-- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
--
-- <p>This document and translations of it may be copied and
-- furnished to others, and derivative works that comment on or
-- otherwise explain it or assist in its implementation may be
-- prepared, copied, published and distributed, in whole or in
-- part, without restriction of any kind, provided that the above
-- copyright notice and this paragraph are included on all such
-- copies and derivative works. However, this document itself may
-- not be modified in any way, such as by removing the copyright
-- notice or references to the Scheme Request For Implementation
-- process or editors, except as needed for the purpose of
-- developing SRFIs in which case the procedures for copyrights
-- defined in the SRFI process must be followed, or as required to
-- translate it into languages other than English.</p>
--
-- <p>The limited permissions granted above are perpetual and will
-- not be revoked by the authors or their successors or
-- assigns.</p>
--
-- <p>This document and the information contained herein is
-- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
-- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
-- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
-- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
-- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
-- <hr>
--
-- <address>
-- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
-- Rush</a>
-- </address>
--
-- <address>
-- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
-- Miller</a>
-- </address>
-- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
-- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
-- Daylight Time 2002 <!-- hhmts end --> <br>
-- </body>
--</html>
--
-diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
---- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-01-31 13:04:07.221391727 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
-@@ -1,345 +0,0 @@
--<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
--<html>
--<head>
-- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
--</head>
--<body>
--
--<H1>Title</H1>
--
--SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
--
--<H1>Author</H1>
--
--Andy Gaynor
--
--<H1>Status</H1>
--
--This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
--You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
--<P><UL>
--<LI>Received: 1999/2/2
--<LI>Draft: 1999/2/10-1999/04/12
--<LI>Final: 1999/4/26
--<LI>Revised reference implementation: 2003/01/27
--</UL>
--
--<H1>Abstract</H1>
--
--The <i>named-let</i> incarnation of the <code>let</code> form has two slight
--inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
--form makes no accommodation for rest arguments, an issue of functionality
--and consistency. As defined, the <code>let</code> form does not accommodate
--signature-style syntax, an issue of aesthetics and consistency. Both
--issues are addressed here in a manner which is compatible with the traditional
--<code>let</code> form but for minor extensions.
--
--<H1>Rationale</H1>
--
--<H2>Signature-style Syntax</H2>
--
--Consider the following two equivalent definitions:
--
--<p><pre>
--(define fibonacci
-- (lambda (n i f0 f1)
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
--
--(define (fibonacci n i f0 f1)
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--</pre>
--
--Although there is a named-let analog for the former form, there is none
--for the latter. To wit, suppose one wished to compute the 10th element
--of the Fibonacci sequence using a named let:
--
--<p>
--<pre>
--(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--Values: 55
--</pre>
--
--As it stands, one cannot equivalently write
--
--<p>
--<pre>
--(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--</pre>
--
--which is consistent with <code>define</code>'s signature-style form.
--<p>Those that favor the signature style may prefer this extension.
--In any case, it may be more appropriate to include all bound names within
--the binding section. As presented, this straightforward extension
--introduces no ambiguity or incompatibility with the existing definition
--of let.
--
--<H2>Rest Arguments</H2>
--
--As it stands, one cannot write a named let with rest arguments, as in
--
--<p>
--<pre>
--(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
-- (if (null? x)
-- 'just-a-silly-contrived-example
-- (begin
-- (write (car x) port)
-- (apply blast port (cdr x)))))
--</pre>
--
--otherwise equivalent to
--
--<p>
--<pre>
--(letrec ((blast (lambda (port . x)
-- (if (null? x)
-- 'just-a-silly-contrived-example
-- (begin
-- (write (car x) port)
-- (apply blast port (cdr x)))))))
-- (blast (current-output-port) (+ 1 2) 4 5))
--</pre>
--
--While this example is rather contrived, the functionality is not.
--There are several times when the author has used this construct in practice.
--Regardless, there is little reason to deny the <code>let</code> form access to
--all the features of lambda functionality.
--
--<H2>Symbols in Binding Sections</H2>
--
--Both the features above rely upon the placement of symbols in <code>let</code>
--binding lists (this statement is intentially simplistic). The only
--other apparent use of such symbol placement is to tersely bind variables
--to unspecified values. For example, one might desire to use
--<code>(let (foo bar baz) ...)</code>
--to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
--unspecified values.
--
--<p>This usage is considered less important in light of the rationales
--presented above, and an alternate syntax is immediately apparent, as
--in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
--be preferable, consistently parenthesizing normal binding clauses.
--
--<H1>Specification</H1>
--
--<H2>Syntax</H2>
--
--<p>
--A formal specification of the syntax follows. Below, body, expression,
--and identifier are free. Each instantiation of binding-name must be
--unique.
--</p>
--
--<p>
--<pre>
-- let = "(" "let" let-bindings body ")"
-- expressions = nothing | expression expressions
-- let-bindings = let-name bindings
-- |"(" let-name "." bindings ")"
-- let-name = identifier
-- bindings = "(" ")"
-- | rest-binding
-- | "(" normal-bindings ["." rest-binding] ")"
--normal-bindings = nothing
-- | normal-binding normal-bindings
--normal-binding = "(" binding-name expression ")"
-- binding-name = identifier
-- rest-binding = "(" binding-name expressions ")"
--</pre>
--
--<p>
--For clarity and convenience, an informal specification follows.
--</p>
--
--<ol>
--<li><a name="unnamed">Unnamed</a>
--
--<p><pre>
--(let ((<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-non-sig">
--Named, non-signature-style, no rest argument</a>
--
--<p><pre>
--(let <name> ((<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-sig">Named, signature-style, no rest argument</a>
--
--<p><pre>
--(let (<name> (<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
--
--<p><pre>
--(let <name> ((<parameter> <argument>)...
--
--. (<rest-parameter> <rest-argument>...))
-- <body>...)
--</pre>
--
--<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
--
--<p><pre>
--(let (<name> (<parameter> <argument>)...
--
--. (<rest-parameter> <rest-argument>...))
-- <body>...)
--</pre>
--</li>
--</ol>
--
--<H2>Semantics</H2>
--
--Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
--and <code>letrec</code> forms, respectively.
--
--<ul>
--<li>For informal syntax 1:
--
--<p><pre>
--(($lambda (<parameter>...) <body>...) <argument>...)
--</pre>
--</li>
--
--<li>For informal syntaxes 2 and 3:
--
--<p>
--<pre>
--($letrec ((<name> ($lambda (<parameter>...) <body>...)))
-- (<name> <argument>...))
--</pre>
--</li>
--
--<li>For informal syntaxes 4 and 5:
--
--<p>
--<pre>
--($letrec ((<name> ($lambda (<parameter>...
--
--. <rest-parameter>) <body>...)))
-- (<name> <argument>... <rest-argument>...))
--</pre>
--</li>
--</ul>
--
--<H1>Implementation</H1>
--
--Here is an implementation using <code>SYNTAX-RULES</code>.
--
--<p>
--<pre>
--;; Use your own standard let.
--;; Or call a lambda.
--;; (define-syntax standard-let
--;;
--;; (syntax-rules ()
--;;
--;; ((let ((var val) ...) body ...)
--;; ((lambda (var ...) body ...) val ...))))
--
--(define-syntax let
--
-- (syntax-rules ()
--
-- ;; No bindings: use standard-let.
-- ((let () body ...)
-- (standard-let () body ...))
-- ;; Or call a lambda.
-- ;; ((lambda () body ...))
--
-- ;; All standard bindings: use standard-let.
-- ((let ((var val) ...) body ...)
-- (standard-let ((var val) ...) body ...))
-- ;; Or call a lambda.
-- ;; ((lambda (var ...) body ...) val ...)
--
-- ;; One standard binding: loop.
-- ;; The all-standard-bindings clause didn't match,
-- ;; so there must be a rest binding.
-- ((let ((var val) . bindings) body ...)
-- (let-loop #f bindings (var) (val) (body ...)))
--
-- ;; Signature-style name: loop.
-- ((let (name binding ...) body ...)
-- (let-loop name (binding ...) () () (body ...)))
--
-- ;; defun-style name: loop.
-- ((let name bindings body ...)
-- (let-loop name bindings () () (body ...)))))
--
--(define-syntax let-loop
--
-- (syntax-rules ()
--
-- ;; Standard binding: destructure and loop.
-- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
-- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
--
-- ;; Rest binding, no name: use standard-let, listing the rest values.
-- ;; Because of let's first clause, there is no "no bindings, no name" clause.
-- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
-- ;; Or call a lambda with a rest parameter on all values.
-- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
-- ;; Or use one of several other reasonable alternatives.
--
-- ;; No bindings, name: call a letrec'ed lambda.
-- ((let-loop name () (var ...) (val ...) body)
-- ((letrec ((name (lambda (var ...) . body)))
-- name)
-- val ...))
--
-- ;; Rest binding, name: call a letrec'ed lambda.
-- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
-- ((letrec ((name (lambda (var ... . rest-var) . body)))
-- name)
-- val ... rest-val ...))))
--</pre>
--
--
--<H1>Copyright</H1>
--
--Copyright (C) Andy Gaynor (1999). All Rights Reserved.
--<p>This document and translations of it may be copied and furnished to
--others, and derivative works that comment on or otherwise explain it or
--assist in its implementation may be prepared, copied, published and distributed,
--in whole or in part, without restriction of any kind, provided that the
--above copyright notice and this paragraph are included on all such copies
--and derivative works. However, this document itself may not be modified
--in any way, such as by removing the copyright notice or references to the
--Scheme Request For Implementation process or editors, except as needed
--for the purpose of developing SRFIs in which case the procedures for copyrights
--defined in the SRFI process must be followed, or as required to translate
--it into languages other than English.
--<p>The limited permissions granted above are perpetual and will not be
--revoked by the authors or their successors or assigns.
--<p>This document and the information contained herein is provided on an
--"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
--EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
--USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
--WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
--
-- <hr>
-- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
--
--</body>
--</html>
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/info.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/info.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,14 +0,0 @@
--#lang info
--
--(define collection 'multi)
--
--(define deps '("scheme-lib"
-- "base"
-- "r6rs-lib"
-- "srfi-lib"
-- "compatibility-lib"))
--
--
--(define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"")
--
--(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1 +0,0 @@
--#lang s-exp srfi/provider srfi/%3a5
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,3 +0,0 @@
--#lang s-exp srfi/provider srfi/5
--
--;; FIXME: "rest arguments" need to generate mutable lists
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,86 +0,0 @@
--;;;
--;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
--;;; Time-stamp: <03/04/08 09:56:06 solsona>
--;;;
--;;; Usually, I would add a copyright notice, and the announce that
--;;; this code is under the LGPL licence. Nevertheless, I only did the
--;;; port to PLT Scheme v200, and here is the copyright notice,
--;;; comments, and licence from the original source:
--;;;
--;;; Copyright (C) Andy Gaynor (1999-2003)
--;;;
--;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
--;;; Renamed to s:let by Eli Barzilay
--
--#lang scheme/base
--(provide s:let)
--
--(define-syntax s:let
-- (syntax-rules ()
-- ;; standard
-- ((s:let () body ...)
-- (let () body ...))
-- ((s:let ((var val) ...) body ...)
-- (let ((var val) ...) body ...))
--
-- ;; rest style
-- ((s:let ((var val) . bindings) body ...)
-- (let-loop #f bindings (var) (val) (body ...)))
--
-- ;; signature style
-- ((s:let (name bindings ...) body ...)
-- (let-loop name (bindings ...) () () (body ...)))
--
-- ;; standard named style
-- ((s:let name (bindings ...) body ...)
-- (let-loop name (bindings ...) () () (body ...)))
--
-- ))
--
--;; A loop to walk down the list of bindings.
--
--(define-syntax let-loop
-- (syntax-rules ()
--
-- ;; No more bindings - make a LETREC.
-- ((let-loop name () (vars ...) (vals ...) body)
-- ((letrec ((name (lambda (vars ...) . body)))
-- name)
-- vals ...))
--
-- ;; Rest binding, no name
-- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-- (let ((var val) ... (rest-var (list rest-val ...))) . body))
--
-- ;; Process a (var val) pair.
-- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
-- (let-loop name (more ...) (vars ... var) (vals ... val) body))
--
-- ;; End with a rest variable - make a LETREC.
-- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
-- ((letrec ((name (lambda (vars ... . rest-var) . body)))
-- name)
-- vals ... rest-vals ...))))
--
--;; Four loops - normal and `signature-style', each with and without a rest
--;; binding.
--;;
--;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
--;; (if (= i n)
--;; f0
--;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
--;;
--;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
--;; (if (= i n)
--;; f0
--;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
--;;
--;;(let fibonacci ((n 10) (i 0) . (f 0 1))
--;; (if (= i n)
--;; (car f)
--;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
--;;
--;;(let (fibonacci (n 10) (i 0) . (f 0 1))
--;; (if (= i n)
--;; (car f)
--;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
-diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
---- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-01-31 13:04:07.236391659 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,2 +0,0 @@
--;; module loader for SRFI-5
--#lang s-exp srfi/provider srfi/5/let #:unprefix s:
diff --git a/racket-6.12-update-srfi.patch b/racket-6.12-update-srfi.patch
deleted file mode 100644
index 1894e32..0000000
--- a/racket-6.12-update-srfi.patch
+++ /dev/null
@@ -1,4369 +0,0 @@
-diff -urN racket-6.12-orig/share/links.rktd racket-6.12/share/links.rktd
---- racket-6.12-orig/share/links.rktd 2018-01-26 16:12:13.000000000 -0500
-+++ racket-6.12/share/links.rktd 2018-01-31 12:22:57.600504036 -0500
-@@ -133,6 +133,7 @@
- (root "pkgs/r5rs-doc")
- (root "pkgs/srfi-lite-lib")
- (root "pkgs/srfi-lib")
-+ (root "pkgs/srfi-lib-nonfree")
- (root "pkgs/html-doc")
- (root "pkgs/images-gui-lib")
- (root "pkgs/images-doc")
-@@ -176,6 +177,7 @@
- (root "pkgs/slideshow-exe")
- (root "pkgs/slideshow-plugin")
- (root "pkgs/srfi-doc")
-+ (root "pkgs/srfi-doc-nonfree")
- (root "pkgs/syntax-color-doc")
- (root "pkgs/web-server-lib")
- (root "pkgs/unix-socket-lib")
-diff -urN racket-6.12-orig/share/pkgs/srfi/info.rkt racket-6.12/share/pkgs/srfi/info.rkt
---- racket-6.12-orig/share/pkgs/srfi/info.rkt 2018-01-26 16:10:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi/info.rkt 2018-01-31 12:22:58.139502019 -0500
-@@ -1 +1,16 @@
--(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("srfi-lib" "srfi-doc"))) (define implies (quote ("srfi-lib" "srfi-doc"))) (define pkg-desc "Legacy SRFI (Scheme) libraries") (define pkg-authors (quote (mflatt noel chongkai jay)))))
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define deps '("srfi-lib"
-+ "srfi-lib-nonfree"
-+ "srfi-doc"
-+ "srfi-doc-nonfree"))
-+(define implies '("srfi-lib"
-+ "srfi-lib-nonfree"
-+ "srfi-doc"
-+ "srfi-foc-nonfree"))
-+
-+(define pkg-desc "Legacy SRFI (Scheme) libraries")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/info.rkt racket-6.12/share/pkgs/srfi-doc/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-doc/info.rkt 2018-01-26 16:10:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/info.rkt 2018-01-31 12:22:58.139502019 -0500
-@@ -1 +1,24 @@
--(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define build-deps (quote ("mzscheme-doc" "scheme-lib" "base" "scribble-lib" "srfi-lib" "racket-doc" "r5rs-doc" "r6rs-doc" "compatibility-lib"))) (define update-implies (quote ("srfi-lib"))) (define pkg-desc "documentation part of \"srfi\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define build-deps '("mzscheme-doc"
-+ "scheme-lib"
-+ "base"
-+ "scribble-lib"
-+ "srfi-lib"
-+ "racket-doc"
-+ "r5rs-doc"
-+ "r6rs-doc"
-+ "compatibility-lib"))
-+
-+(define deps '("scheme-lib"
-+ "base"
-+ "scribble-lib"
-+ "compatibility-lib"))
-+
-+(define update-implies '("srfi-lib"))
-+
-+(define pkg-desc "documentation part of \"srfi\"")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl
---- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl 2018-01-31 12:22:58.144502000 -0500
-@@ -1,5 +1,6 @@
- #lang scribble/doc
--@(require scribble/manual
-+@(require srfi/scribblings/util
-+ scribble/manual
- scribble/eval
- scriblib/render-cond
- scribble/core
-@@ -8,54 +9,6 @@
- (for-label scheme/base
- racket/stream))
-
--@(define-syntax (srfi stx)
-- (syntax-case stx ()
-- [(_ num #:subdir subdir? . title)
-- (with-syntax ([srfi/n (string->symbol (format "srfi/~a" (syntax-e #'num)))])
-- #'(begin
-- (section #:tag (format "srfi-~a" num)
-- #:style 'unnumbered
-- (format "SRFI ~a: " num)
-- . title)
-- (defmodule srfi/n)
-- "Original specification: "
-- (let* ([label (format "SRFI ~a" num)]
-- [sub (if subdir? (format "srfi-~a/" num) "")]
-- [url ( (b) (format "~a/srfi-std/~asrfi-~a.html" b sub num))])
-- (cond-element
-- [(or latex text) @link[(url "http://docs.racket-lang.org") label]]
-- [else @link[(url ".") label]]))))]
-- [(_ num . title) #'(srfi num #:subdir #f . title)]))
--
--@;{ The `lst' argument is a list of
-- (list sym syntactic-form? html-anchor) }
--@(define (redirect n lst #:subdir [subdir? #f])
-- (let ([file (if subdir?
-- (format "srfi-~a/srfi-~a.html" n n)
-- (format "srfi-~a.html" n))]
-- [mod-path (string->symbol (format "srfi/~a" n))])
-- (make-binding-redirect-elements mod-path
-- (map (lambda (b)
-- (list (car b) (cadr b)
-- (build-path "srfi-std" file)
-- (caddr b)))
-- lst))))
--
--@(define in-core
-- (case-lambda
-- [() (in-core ".")]
-- [(k) @elem{This SRFI's bindings are also available in
-- @racketmodname[racket/base]@|k|}]))
--
--@(begin
-- (define-syntax-rule (def-mz mz-if)
-- (begin
-- (require (for-label mzscheme))
-- (define mz-if (racket if))))
-- (def-mz mz-if))
--
--@(define srfi-std (style #f (list (install-resource "srfi-std"))))
--
- @; ----------------------------------------------------------------------
-
- @title{SRFIs: Libraries}
-@@ -245,14 +198,6 @@
-
- @; ----------------------------------------
-
--@srfi[5]{A compatible let form with signatures and rest arguments}
--
--@redirect[5 '(
-- (let #t "unnamed")
--)]
--
--@; ----------------------------------------
--
- @srfi[6]{Basic String Ports}
-
- @redirect[6 '(
-@@ -671,20 +616,6 @@
-
- @; ----------------------------------------
-
--@srfi[29]{Localization}
--
--@redirect[29 '(
-- (current-language #f "current-language")
-- (current-country #f "current-country")
-- (current-locale-details #f "current-locale-details")
-- (declare-bundle! #f "declare-bundle!")
-- (store-bundle #f "store-bundle")
-- (load-bundle! #f "load-bundle!")
-- (localized-template #f "localized-template")
--)]
--
--@; ----------------------------------------
--
- @srfi[30]{Nested Multi-line Comments}
-
- This SRFI's syntax is part of Racket's default reader.
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html
---- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
-@@ -1,507 +0,0 @@
--<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
--<html>
-- <head>
-- <meta name="generator" content="HTML Tidy, see www.w3.org">
-- <title>SRFI 29: Localization</title>
-- <meta name="author" content="Scott G. Miller">
-- <meta name="description" content="Localization">
-- </head>
-- <body>
-- <H1>Title</H1>
--
-- SRFI 29: Localization
--
-- <H1>Author</H1>
--
-- Scott G. Miller
--
-- <H1>Abstract</H1>
--
-- This document specifies an interface to retrieving and
-- displaying locale sensitive messages. A Scheme program can
-- register one or more translations of templated messages, and
-- then write Scheme code that can transparently retrieve the
-- appropriate message for the locale under which the Scheme
-- system is running. <br>
--
--
-- <H1>Rationale</H1>
--
-- <p>As any programmer that has ever had to deal with making his
-- or her code readable in more than one locale, the process of
-- sufficiently abstracting program messages from their
-- presentation to the user is non-trivial without help from the
-- programming language. Most modern programming language
-- libraries do provide some mechanism for performing this
-- separation.</p>
--
-- <p>A portable API that allows a piece of code to run without
-- modification in different countries and under different
-- languages is a must for any non-trivial software project.
-- The interface should separate the logic of a program from
-- the myriad of translations that may be necessary.</p>
--
-- <p>The interface described in this document provides such
-- functionality. The underlying implementation is also allowed to
-- use whatever datastructures it likes to provide access to the
-- translations in the most efficient manner possible. In
-- addition, the implementation is provided with standardized
-- functions that programs will use for accessing an external,
-- unspecified repository of translations.</p>
--
-- <p>This interface <i>does not</i> cover all aspects of
-- localization, including support for non-latin characters,
-- number and date formatting, etc. Such functionality is the
-- scope of a future SRFI that may extend this one.</p>
--
-- <H1>Dependencies</H1>
--
-- An SRFI-29 conformant implementation must also implement
-- SRFI-28, Basic Format Strings. Message templates are strings
-- that must be processed by the <tt>format</tt> function
-- specified in that SRFI.
--
-- <H1>Specification</H1>
--
-- <h3>Message Bundles</h3>
--
-- <p>A Message Bundle is a set of message templates and their
-- identifying keys. Each bundle contains one or more such
-- key/value pairs. The bundle itself is associated with a
-- <i>bundle specifier</i> which uniquely identifies the
-- bundle.</p>
--
-- <h3>Bundle Specifiers</h3>
--
-- <p>A Bundle Specifier is a Scheme list that describes, in order
-- of importance, the package and locale of a message bundle.
-- In most cases, a locale specifier will have between one
-- and three elements. The first element is a symbol denoting the
-- package for which this bundle applies. The second and third
-- elements denote a <i>locale</i>. The second element (first
-- element of the locale) if present, is the two letter, ISO 639-1
-- language code for the bundle. The third element, if present, is
-- a two letter ISO 3166-1 country code. In some cases, a
-- fourth element may be present, specifying the encoding used for
-- the bundle. All bundle specifier elements are Scheme
-- symbols.</p>
--
-- <p>If only one translation is provided, it should be designated
-- only by a package name, for example <tt>(mathlib)</tt>. This
-- translation is called the <i>default</i> translation.</p>
--
-- <h3>Bundle Searching</h3>
--
-- <p>When a message template is retrieved from a bundle, the
-- Scheme implementation will provide the locale under which the
-- system is currently running. When the template is retrieved,
-- the package name will be specified. The Scheme system should
-- construct a Bundle Specifier from the provided package name and
-- the active locale. For example, when retrieving a message
-- template for French Canadian, in the <tt>mathlib</tt> package,
-- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
-- program may also retrieve the elements of the current locale
-- using the no-argument procedures:</p>
--
-- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
-- <i>symbol</i></tt><br>
-- <tt><b>current-language</b> <i>symbol</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- When given no arguments, returns the current ISO 639-1
-- language code as a symbol. If provided with an
-- argument, the current language is set to that named by the
-- symbol for the currently executing Scheme thread (or for the
-- entire Scheme system if such a distinction is not possible).
--
-- </blockquote>
--
-- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
-- <i>symbol</i></tt><br>
-- <tt><b>current-country</b> <i>symbol</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- returns the current ISO 3166-1 country code as a symbol.
-- If provided with an argument, the current country is
-- set to that named by the symbol for the currently executing
-- Scheme thread (or for the entire Scheme system if such a
-- distinction is not possible).
-- </blockquote>
--
-- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
-- symbol</i></tt>s<br>
-- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
-- undefined</tt><br>
-- </p>
--
-- <blockquote>
-- Returns a list of additional locale details as a list of
-- symbols. This list may contain information about
-- encodings or other more specific information. If
-- provided with an argument, the current locale details are set
-- to those given in the currently executing Scheme thread (or
-- for the entire Scheme system if such a distinction is not
-- possible).
-- </blockquote>
--
-- <p>The Scheme System should first check for a bundle with the
-- exact name provided. If no such bundle is found, the last
-- element from the list is removed and a search is tried for a
-- bundle with that name. If no bundle is then found, the list is
-- shortened by removing the last element again. If no message is
-- found and the bundle specifier is now the empty list, an error
-- should be raised.</p>
--
-- <p>The reason for this search order is to provide the most
-- locale sensitive template possible, but to fall back on more
-- general templates if a translation has not yet been provided
-- for the given locale.</p>
--
-- <h3>Message Templates</h3>
--
-- <p>A message template is a localized message that may or may
-- not contain one of a number of formatting codes. A message
-- template is a Scheme string. The string is of a form that can
-- be processed by the <tt>format</tt> procedure found in many
-- Scheme systems and formally specified in SRFI-28 (Basic Format
-- Strings).</p>
--
-- <p>This SRFI also extends SRFI-28 to provide an additional
-- <tt>format</tt> escape code:</p>
--
-- <blockquote>
-- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
-- follows this code immediately to reference the [N]'th
-- optional value absolutely, rather than the next unconsumed
-- value. The referenced value is <i>not</i> consumed.
-- </blockquote>
-- This extension allows optional values to be positionally
-- referenced, so that message templates can be constructed that
-- can produce the proper word ordering for a language.
--
-- <h3>Preparing Bundles</h3>
-- Before a bundle may be used by the Scheme system to retrieve
-- localized template messages, they must be made available to the
-- Scheme system. This SRFI specifies a way to portably
-- define the bundles, as well as store them in and retrieve them
-- from an unspecified system which may be provided by resources
-- outside the Scheme system.<br>
--
--
-- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
-- association-list</i> -> undefined<br>
-- </tt></p>
--
-- <blockquote>
-- Declares a new bundle named by the given bundle-specifier.
-- The contents of the bundle are defined by the provided
-- association list. The list contains associations
-- between Scheme symbols and the message templates (Scheme
-- strings) they name. If a bundle already exists with the
-- given name, it is overwritten with the newly declared
-- bundle.<br>
-- </blockquote>
-- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
-- boolean</tt><br>
--
--
-- <blockquote>
-- Attempts to store a bundle named by the given bundle
-- specifier, and previously made available using
-- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
-- unspecified mechanism that may be persistent across Scheme
-- system restarts. If successful, a non-false value is
-- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
-- </blockquote>
-- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
-- boolean</tt><br>
--
--
-- <blockquote>
-- Attempts to retrieve a bundle from an unspecified mechanism
-- which stores bundles outside the Scheme system. If the
-- bundle was retrieved successfully, the function returns a
-- non-false value, and the bundle is immediately available to
-- the Scheme system. If the bundle could not be found or loaded
-- successfully, the function returns <tt>#f</tt>, and the
-- Scheme system's bundle registry remains unaffected.<br>
-- </blockquote>
-- A compliant Scheme system may choose not to provide any
-- external mechanism to store localized bundles. If it does
-- not, it must still provide implementations for
-- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
-- case, both functions must return <tt>#f</tt> regardless of the
-- arguments given. Users of this SRFI should recognize that the
-- inability to load or store a localized bundle in an external
-- repository is <i>not</i> a fatal error.<br>
--
--
-- <h3>Retrieving Localized Message Templates</h3>
--
-- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
-- message-template-name</tt></i> <tt>-> <i>string or #f<br>
-- </i></tt></p>
--
-- <blockquote>
-- Retrieves a localized message template for the given package
-- name and the given message template name (both symbols).
-- If no such message could be found, false (#f) is
-- returned.<br>
-- <br>
-- </blockquote>
-- After retrieving a template, the calling program can use
-- <tt>format</tt> to produce a string that can be displayed to
-- the user.<br>
--
--
-- <h2>Examples</h2>
-- The below example makes use of SRFI-29 to display simple,
-- localized messages. It also defines its bundles in such a
-- way that the Scheme system may store and retrieve the bundles
-- from a more efficient system catalog, if available.<br>
--
--<pre>
--(let ((translations
-- '(((en) . ((time . "Its ~a, ~a.")
-- (goodbye . "Goodbye, ~a.")))
-- ((fr) . ((time . "~1@*~a, c'est ~a.")
-- (goodbye . "Au revoir, ~a."))))))
-- (for-each (lambda (translation)
-- (let ((bundle-name (cons 'hello-program (car translation))))
-- (if (not (load-bundle! bundle-name))
-- (begin
-- (declare-bundle! bundle-name (cdr translation))
-- (store-bundle! bundle-name)))))
-- translations))
--
--(define localized-message
-- (lambda (message-name . args)
-- (apply format (cons (localized-template 'hello-program
-- message-name)
-- args))))
--
--(let ((myname "Fred"))
-- (display (localized-message 'time "12:00" myname))
-- (display #\newline)
--
-- (display (localized-message 'goodbye myname))
-- (display #\newline))
--
--;; Displays (English):
--;; Its 12:00, Fred.
--;; Goodbye, Fred.
--;;
--;; French:
--;; Fred, c'est 12:00.
--;; Au revoir, Fred.
--</pre>
--
-- <H1>Implementation</H1>
--
-- <p>The implementation requires that the Scheme system provide a
-- definition for <tt>current-language</tt> and
-- <tt>current-country</tt> capable of distinguishing the correct
-- locale present during a Scheme session. The definitions of
-- those functions in the reference implementation are not capable
-- of that distinction. Their implementation is provided only so
-- that the following code can run in any R4RS scheme system.
-- <br>
-- </p>
--
-- <p>In addition, the below implementation of a compliant
-- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
-- SRFI-23 (Error reporting)</p>
--<pre>
--;; The association list in which bundles will be stored
--(define *localization-bundles* '())
--
--;; The current-language and current-country functions provided
--;; here must be rewritten for each Scheme system to default to the
--;; actual locale of the session
--(define current-language
-- (let ((current-language-value 'en))
-- (lambda args
-- (if (null? args)
-- current-language-value
-- (set! current-language-value (car args))))))
--
--(define current-country
-- (let ((current-country-value 'us))
-- (lambda args
-- (if (null? args)
-- current-country-value
-- (set! current-country-value (car args))))))
--
--;; The load-bundle! and store-bundle! both return #f in this
--;; reference implementation. A compliant implementation need
--;; not rewrite these procedures.
--(define load-bundle!
-- (lambda (bundle-specifier)
-- #f))
--
--(define store-bundle!
-- (lambda (bundle-specifier)
-- #f))
--
--;; Declare a bundle of templates with a given bundle specifier
--(define declare-bundle!
-- (letrec ((remove-old-bundle
-- (lambda (specifier bundle)
-- (cond ((null? bundle) '())
-- ((equal? (caar bundle) specifier)
-- (cdr bundle))
-- (else (cons (car bundle)
-- (remove-old-bundle specifier
-- (cdr bundle))))))))
-- (lambda (bundle-specifier bundle-assoc-list)
-- (set! *localization-bundles*
-- (cons (cons bundle-specifier bundle-assoc-list)
-- (remove-old-bundle bundle-specifier
-- *localization-bundles*))))))
--
--;;Retrieve a localized template given its package name and a template name
--(define localized-template
-- (letrec ((rdc
-- (lambda (ls)
-- (if (null? (cdr ls))
-- '()
-- (cons (car ls) (rdc (cdr ls))))))
-- (find-bundle
-- (lambda (specifier template-name)
-- (cond ((assoc specifier *localization-bundles*) =>
-- (lambda (bundle) bundle))
-- ((null? specifier) #f)
-- (else (find-bundle (rdc specifier)
-- template-name))))))
-- (lambda (package-name template-name)
-- (let loop ((specifier (cons package-name
-- (list (current-language)
-- (current-country)))))
-- (and (not (null? specifier))
-- (let ((bundle (find-bundle specifier template-name)))
-- (and bundle
-- (cond ((assq template-name bundle) => cdr)
-- ((null? (cdr specifier)) #f)
-- (else (loop (rdc specifier)))))))))))
--
--;;An SRFI-28 and SRFI-29 compliant version of format. It requires
--;;SRFI-23 for error reporting.
--(define format
-- (lambda (format-string . objects)
-- (let ((buffer (open-output-string)))
-- (let loop ((format-list (string->list format-string))
-- (objects objects)
-- (object-override #f))
-- (cond ((null? format-list) (get-output-string buffer))
-- ((char=? (car format-list) #\~)
-- (cond ((null? (cdr format-list))
-- (error 'format "Incomplete escape sequence"))
-- ((char-numeric? (cadr format-list))
-- (let posloop ((fl (cddr format-list))
-- (pos (string->number
-- (string (cadr format-list)))))
-- (cond ((null? fl)
-- (error 'format "Incomplete escape sequence"))
-- ((and (eq? (car fl) '#\@)
-- (null? (cdr fl)))
-- (error 'format "Incomplete escape sequence"))
-- ((and (eq? (car fl) '#\@)
-- (eq? (cadr fl) '#\*))
-- (loop (cddr fl) objects (list-ref objects pos)))
-- (else
-- (posloop (cdr fl)
-- (+ (* 10 pos)
-- (string->number
-- (string (car fl)))))))))
-- (else
-- (case (cadr format-list)
-- ((#\a)
-- (cond (object-override
-- (begin
-- (display object-override buffer)
-- (loop (cddr format-list) objects #f)))
-- ((null? objects)
-- (error 'format "No value for escape sequence"))
-- (else
-- (begin
-- (display (car objects) buffer)
-- (loop (cddr format-list)
-- (cdr objects) #f)))))
-- ((#\s)
-- (cond (object-override
-- (begin
-- (display object-override buffer)
-- (loop (cddr format-list) objects #f)))
-- ((null? objects)
-- (error 'format "No value for escape sequence"))
-- (else
-- (begin
-- (write (car objects) buffer)
-- (loop (cddr format-list)
-- (cdr objects) #f)))))
-- ((#\%)
-- (if object-override
-- (error 'format "Escape sequence following positional override does not require a value"))
-- (display #\newline buffer)
-- (loop (cddr format-list) objects #f))
-- ((#\~)
-- (if object-override
-- (error 'format "Escape sequence following positional override does not require a value"))
-- (display #\~ buffer)
-- (loop (cddr format-list) objects #f))
-- (else
-- (error 'format "Unrecognized escape sequence"))))))
-- (else (display (car format-list) buffer)
-- (loop (cdr format-list) objects #f)))))))
--
--</pre>
--
-- <H1>Copyright</H1>
--
-- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
--
-- <p>This document and translations of it may be copied and
-- furnished to others, and derivative works that comment on or
-- otherwise explain it or assist in its implementation may be
-- prepared, copied, published and distributed, in whole or in
-- part, without restriction of any kind, provided that the above
-- copyright notice and this paragraph are included on all such
-- copies and derivative works. However, this document itself may
-- not be modified in any way, such as by removing the copyright
-- notice or references to the Scheme Request For Implementation
-- process or editors, except as needed for the purpose of
-- developing SRFIs in which case the procedures for copyrights
-- defined in the SRFI process must be followed, or as required to
-- translate it into languages other than English.</p>
--
-- <p>The limited permissions granted above are perpetual and will
-- not be revoked by the authors or their successors or
-- assigns.</p>
--
-- <p>This document and the information contained herein is
-- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
-- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
-- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
-- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
-- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
-- <hr>
--
-- <address>
-- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
-- Rush</a>
-- </address>
--
-- <address>
-- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
-- Miller</a>
-- </address>
-- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
-- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
-- Daylight Time 2002 <!-- hhmts end --> <br>
-- </body>
--</html>
--
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html
---- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
-@@ -1,345 +0,0 @@
--<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
--<html>
--<head>
-- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
--</head>
--<body>
--
--<H1>Title</H1>
--
--SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
--
--<H1>Author</H1>
--
--Andy Gaynor
--
--<H1>Status</H1>
--
--This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
--You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
--<P><UL>
--<LI>Received: 1999/2/2
--<LI>Draft: 1999/2/10-1999/04/12
--<LI>Final: 1999/4/26
--<LI>Revised reference implementation: 2003/01/27
--</UL>
--
--<H1>Abstract</H1>
--
--The <i>named-let</i> incarnation of the <code>let</code> form has two slight
--inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
--form makes no accommodation for rest arguments, an issue of functionality
--and consistency. As defined, the <code>let</code> form does not accommodate
--signature-style syntax, an issue of aesthetics and consistency. Both
--issues are addressed here in a manner which is compatible with the traditional
--<code>let</code> form but for minor extensions.
--
--<H1>Rationale</H1>
--
--<H2>Signature-style Syntax</H2>
--
--Consider the following two equivalent definitions:
--
--<p><pre>
--(define fibonacci
-- (lambda (n i f0 f1)
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
--
--(define (fibonacci n i f0 f1)
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--</pre>
--
--Although there is a named-let analog for the former form, there is none
--for the latter. To wit, suppose one wished to compute the 10th element
--of the Fibonacci sequence using a named let:
--
--<p>
--<pre>
--(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--Values: 55
--</pre>
--
--As it stands, one cannot equivalently write
--
--<p>
--<pre>
--(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
-- (if (= i n)
-- f0
-- (fibonacci n (+ i 1) f1 (+ f0 f1))))
--</pre>
--
--which is consistent with <code>define</code>'s signature-style form.
--<p>Those that favor the signature style may prefer this extension.
--In any case, it may be more appropriate to include all bound names within
--the binding section. As presented, this straightforward extension
--introduces no ambiguity or incompatibility with the existing definition
--of let.
--
--<H2>Rest Arguments</H2>
--
--As it stands, one cannot write a named let with rest arguments, as in
--
--<p>
--<pre>
--(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
-- (if (null? x)
-- 'just-a-silly-contrived-example
-- (begin
-- (write (car x) port)
-- (apply blast port (cdr x)))))
--</pre>
--
--otherwise equivalent to
--
--<p>
--<pre>
--(letrec ((blast (lambda (port . x)
-- (if (null? x)
-- 'just-a-silly-contrived-example
-- (begin
-- (write (car x) port)
-- (apply blast port (cdr x)))))))
-- (blast (current-output-port) (+ 1 2) 4 5))
--</pre>
--
--While this example is rather contrived, the functionality is not.
--There are several times when the author has used this construct in practice.
--Regardless, there is little reason to deny the <code>let</code> form access to
--all the features of lambda functionality.
--
--<H2>Symbols in Binding Sections</H2>
--
--Both the features above rely upon the placement of symbols in <code>let</code>
--binding lists (this statement is intentially simplistic). The only
--other apparent use of such symbol placement is to tersely bind variables
--to unspecified values. For example, one might desire to use
--<code>(let (foo bar baz) ...)</code>
--to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
--unspecified values.
--
--<p>This usage is considered less important in light of the rationales
--presented above, and an alternate syntax is immediately apparent, as
--in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
--be preferable, consistently parenthesizing normal binding clauses.
--
--<H1>Specification</H1>
--
--<H2>Syntax</H2>
--
--<p>
--A formal specification of the syntax follows. Below, body, expression,
--and identifier are free. Each instantiation of binding-name must be
--unique.
--</p>
--
--<p>
--<pre>
-- let = "(" "let" let-bindings body ")"
-- expressions = nothing | expression expressions
-- let-bindings = let-name bindings
-- |"(" let-name "." bindings ")"
-- let-name = identifier
-- bindings = "(" ")"
-- | rest-binding
-- | "(" normal-bindings ["." rest-binding] ")"
--normal-bindings = nothing
-- | normal-binding normal-bindings
--normal-binding = "(" binding-name expression ")"
-- binding-name = identifier
-- rest-binding = "(" binding-name expressions ")"
--</pre>
--
--<p>
--For clarity and convenience, an informal specification follows.
--</p>
--
--<ol>
--<li><a name="unnamed">Unnamed</a>
--
--<p><pre>
--(let ((<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-non-sig">
--Named, non-signature-style, no rest argument</a>
--
--<p><pre>
--(let <name> ((<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-sig">Named, signature-style, no rest argument</a>
--
--<p><pre>
--(let (<name> (<parameter> <argument>)...)
-- <body>...)
--</pre>
--</li>
--
--<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
--
--<p><pre>
--(let <name> ((<parameter> <argument>)...
--
--. (<rest-parameter> <rest-argument>...))
-- <body>...)
--</pre>
--
--<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
--
--<p><pre>
--(let (<name> (<parameter> <argument>)...
--
--. (<rest-parameter> <rest-argument>...))
-- <body>...)
--</pre>
--</li>
--</ol>
--
--<H2>Semantics</H2>
--
--Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
--and <code>letrec</code> forms, respectively.
--
--<ul>
--<li>For informal syntax 1:
--
--<p><pre>
--(($lambda (<parameter>...) <body>...) <argument>...)
--</pre>
--</li>
--
--<li>For informal syntaxes 2 and 3:
--
--<p>
--<pre>
--($letrec ((<name> ($lambda (<parameter>...) <body>...)))
-- (<name> <argument>...))
--</pre>
--</li>
--
--<li>For informal syntaxes 4 and 5:
--
--<p>
--<pre>
--($letrec ((<name> ($lambda (<parameter>...
--
--. <rest-parameter>) <body>...)))
-- (<name> <argument>... <rest-argument>...))
--</pre>
--</li>
--</ul>
--
--<H1>Implementation</H1>
--
--Here is an implementation using <code>SYNTAX-RULES</code>.
--
--<p>
--<pre>
--;; Use your own standard let.
--;; Or call a lambda.
--;; (define-syntax standard-let
--;;
--;; (syntax-rules ()
--;;
--;; ((let ((var val) ...) body ...)
--;; ((lambda (var ...) body ...) val ...))))
--
--(define-syntax let
--
-- (syntax-rules ()
--
-- ;; No bindings: use standard-let.
-- ((let () body ...)
-- (standard-let () body ...))
-- ;; Or call a lambda.
-- ;; ((lambda () body ...))
--
-- ;; All standard bindings: use standard-let.
-- ((let ((var val) ...) body ...)
-- (standard-let ((var val) ...) body ...))
-- ;; Or call a lambda.
-- ;; ((lambda (var ...) body ...) val ...)
--
-- ;; One standard binding: loop.
-- ;; The all-standard-bindings clause didn't match,
-- ;; so there must be a rest binding.
-- ((let ((var val) . bindings) body ...)
-- (let-loop #f bindings (var) (val) (body ...)))
--
-- ;; Signature-style name: loop.
-- ((let (name binding ...) body ...)
-- (let-loop name (binding ...) () () (body ...)))
--
-- ;; defun-style name: loop.
-- ((let name bindings body ...)
-- (let-loop name bindings () () (body ...)))))
--
--(define-syntax let-loop
--
-- (syntax-rules ()
--
-- ;; Standard binding: destructure and loop.
-- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
-- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
--
-- ;; Rest binding, no name: use standard-let, listing the rest values.
-- ;; Because of let's first clause, there is no "no bindings, no name" clause.
-- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
-- ;; Or call a lambda with a rest parameter on all values.
-- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
-- ;; Or use one of several other reasonable alternatives.
--
-- ;; No bindings, name: call a letrec'ed lambda.
-- ((let-loop name () (var ...) (val ...) body)
-- ((letrec ((name (lambda (var ...) . body)))
-- name)
-- val ...))
--
-- ;; Rest binding, name: call a letrec'ed lambda.
-- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
-- ((letrec ((name (lambda (var ... . rest-var) . body)))
-- name)
-- val ... rest-val ...))))
--</pre>
--
--
--<H1>Copyright</H1>
--
--Copyright (C) Andy Gaynor (1999). All Rights Reserved.
--<p>This document and translations of it may be copied and furnished to
--others, and derivative works that comment on or otherwise explain it or
--assist in its implementation may be prepared, copied, published and distributed,
--in whole or in part, without restriction of any kind, provided that the
--above copyright notice and this paragraph are included on all such copies
--and derivative works. However, this document itself may not be modified
--in any way, such as by removing the copyright notice or references to the
--Scheme Request For Implementation process or editors, except as needed
--for the purpose of developing SRFIs in which case the procedures for copyrights
--defined in the SRFI process must be followed, or as required to translate
--it into languages other than English.
--<p>The limited permissions granted above are perpetual and will not be
--revoked by the authors or their successors or assigns.
--<p>This document and the information contained herein is provided on an
--"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
--EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
--USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
--WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
--
-- <hr>
-- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
--
--</body>
--</html>
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/util.rkt racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/util.rkt
---- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/util.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/util.rkt 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,61 @@
-+#lang scribble/doc
-+
-+@(require scribble/manual
-+ scribble/eval
-+ scriblib/render-cond
-+ scribble/core
-+ scribble/html-properties
-+ (for-syntax scheme/base)
-+ (for-label scheme/base
-+ racket/stream))
-+
-+@(provide (all-defined-out))
-+
-+@(define-syntax (srfi stx)
-+ (syntax-case stx ()
-+ [(_ num #:subdir subdir? . title)
-+ (with-syntax ([srfi/n (string->symbol (format "srfi/~a" (syntax-e #'num)))])
-+ #'(begin
-+ (section #:tag (format "srfi-~a" num)
-+ #:style 'unnumbered
-+ (format "SRFI ~a: " num)
-+ . title)
-+ (defmodule srfi/n)
-+ "Original specification: "
-+ (let* ([label (format "SRFI ~a" num)]
-+ [sub (if subdir? (format "srfi-~a/" num) "")]
-+ [url ( (b) (format "~a/srfi-std/~asrfi-~a.html" b sub num))])
-+ (cond-element
-+ [(or latex text) @link[(url "http://docs.racket-lang.org") label]]
-+ [else @link[(url ".") label]]))))]
-+ [(_ num . title) #'(srfi num #:subdir #f . title)]))
-+
-+@;{ The `lst' argument is a list of
-+ (list sym syntactic-form? html-anchor) }
-+@(define (redirect n lst #:subdir [subdir? #f])
-+ (let ([file (if subdir?
-+ (format "srfi-~a/srfi-~a.html" n n)
-+ (format "srfi-~a.html" n))]
-+ [mod-path (string->symbol (format "srfi/~a" n))])
-+ (make-binding-redirect-elements mod-path
-+ (map (lambda (b)
-+ (list (car b) (cadr b)
-+ (build-path "srfi-std" file)
-+ (caddr b)))
-+ lst))))
-+
-+@(define in-core
-+ (case-lambda
-+ [() (in-core ".")]
-+ [(k) @elem{This SRFI's bindings are also available in
-+ @racketmodname[racket/base]@|k|}]))
-+
-+@(begin
-+ (define-syntax-rule (def-mz mz-if)
-+ (begin
-+ (require (for-label mzscheme))
-+ (define mz-if (racket if))))
-+ (def-mz mz-if))
-+
-+@(define srfi-std (style #f (list (install-resource "srfi-std"))))
-+
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,19 @@
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define build-deps '("mzscheme-doc"
-+ "scheme-lib"
-+ "base"
-+ "scribble-lib"
-+ "srfi-doc"
-+ "srfi-lib-nonfree"
-+ "racket-doc"
-+ "r5rs-doc"
-+ "r6rs-doc"
-+ "compatibility-lib"))
-+(define update-implies '("srfi-lib-nonfree"))
-+
-+(define pkg-desc "documentation part of \"srfi nonfree\"")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,3 @@
-+#lang info
-+
-+(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,66 @@
-+#lang scribble/doc
-+@(require srfi/scribblings/util
-+ scribble/manual
-+ scribble/eval
-+ scriblib/render-cond
-+ scribble/core
-+ scribble/html-properties
-+ (for-syntax scheme/base)
-+ (for-label scheme/base
-+ racket/stream))
-+
-+@; ----------------------------------------------------------------------
-+
-+@title{SRFI Nonfree Libraries and Documentation}
-+
-+The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
-+Implementation} (a.k.a. @deftech{SRFI}) process allows individual
-+members of the Scheme community to propose libraries and extensions to
-+be supported by multiple Scheme implementations.
-+
-+Racket is distributed with implementations of many SRFIs, most of
-+which can be implemented as libraries. To import the bindings of SRFI
-+@math{n}, use
-+
-+@racketblock[
-+(require @#,elem{@racketidfont{srfi/}@math{n}})
-+]
-+
-+This document lists the SRFIs that are supported by Racket and
-+provides a link to the original SRFI specification (which is also
-+distributed as part of Racket's documentation).
-+
-+The following SRFI specification documents are licensed restrictively.
-+
-+@table-of-contents[]
-+
-+
-+@; ----------------------------------------
-+
-+@srfi[5]{A compatible let form with signatures and rest arguments}
-+
-+@redirect[5 '(
-+ (let #t "unnamed")
-+)]
-+
-+Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
-+
-+@; ----------------------------------------
-+
-+@srfi[29]{Localization}
-+
-+@redirect[29 '(
-+ (current-language #f "current-language")
-+ (current-country #f "current-country")
-+ (current-locale-details #f "current-locale-details")
-+ (declare-bundle! #f "declare-bundle!")
-+ (store-bundle #f "store-bundle")
-+ (load-bundle! #f "load-bundle!")
-+ (localized-template #f "localized-template")
-+)]
-+
-+Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
-+
-+@; ----------------------------------------
-+
-+@index-section[]
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,507 @@
-+<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
-+<html>
-+ <head>
-+ <meta name="generator" content="HTML Tidy, see www.w3.org">
-+ <title>SRFI 29: Localization</title>
-+ <meta name="author" content="Scott G. Miller">
-+ <meta name="description" content="Localization">
-+ </head>
-+ <body>
-+ <H1>Title</H1>
-+
-+ SRFI 29: Localization
-+
-+ <H1>Author</H1>
-+
-+ Scott G. Miller
-+
-+ <H1>Abstract</H1>
-+
-+ This document specifies an interface to retrieving and
-+ displaying locale sensitive messages. A Scheme program can
-+ register one or more translations of templated messages, and
-+ then write Scheme code that can transparently retrieve the
-+ appropriate message for the locale under which the Scheme
-+ system is running. <br>
-+
-+
-+ <H1>Rationale</H1>
-+
-+ <p>As any programmer that has ever had to deal with making his
-+ or her code readable in more than one locale, the process of
-+ sufficiently abstracting program messages from their
-+ presentation to the user is non-trivial without help from the
-+ programming language. Most modern programming language
-+ libraries do provide some mechanism for performing this
-+ separation.</p>
-+
-+ <p>A portable API that allows a piece of code to run without
-+ modification in different countries and under different
-+ languages is a must for any non-trivial software project.
-+ The interface should separate the logic of a program from
-+ the myriad of translations that may be necessary.</p>
-+
-+ <p>The interface described in this document provides such
-+ functionality. The underlying implementation is also allowed to
-+ use whatever datastructures it likes to provide access to the
-+ translations in the most efficient manner possible. In
-+ addition, the implementation is provided with standardized
-+ functions that programs will use for accessing an external,
-+ unspecified repository of translations.</p>
-+
-+ <p>This interface <i>does not</i> cover all aspects of
-+ localization, including support for non-latin characters,
-+ number and date formatting, etc. Such functionality is the
-+ scope of a future SRFI that may extend this one.</p>
-+
-+ <H1>Dependencies</H1>
-+
-+ An SRFI-29 conformant implementation must also implement
-+ SRFI-28, Basic Format Strings. Message templates are strings
-+ that must be processed by the <tt>format</tt> function
-+ specified in that SRFI.
-+
-+ <H1>Specification</H1>
-+
-+ <h3>Message Bundles</h3>
-+
-+ <p>A Message Bundle is a set of message templates and their
-+ identifying keys. Each bundle contains one or more such
-+ key/value pairs. The bundle itself is associated with a
-+ <i>bundle specifier</i> which uniquely identifies the
-+ bundle.</p>
-+
-+ <h3>Bundle Specifiers</h3>
-+
-+ <p>A Bundle Specifier is a Scheme list that describes, in order
-+ of importance, the package and locale of a message bundle.
-+ In most cases, a locale specifier will have between one
-+ and three elements. The first element is a symbol denoting the
-+ package for which this bundle applies. The second and third
-+ elements denote a <i>locale</i>. The second element (first
-+ element of the locale) if present, is the two letter, ISO 639-1
-+ language code for the bundle. The third element, if present, is
-+ a two letter ISO 3166-1 country code. In some cases, a
-+ fourth element may be present, specifying the encoding used for
-+ the bundle. All bundle specifier elements are Scheme
-+ symbols.</p>
-+
-+ <p>If only one translation is provided, it should be designated
-+ only by a package name, for example <tt>(mathlib)</tt>. This
-+ translation is called the <i>default</i> translation.</p>
-+
-+ <h3>Bundle Searching</h3>
-+
-+ <p>When a message template is retrieved from a bundle, the
-+ Scheme implementation will provide the locale under which the
-+ system is currently running. When the template is retrieved,
-+ the package name will be specified. The Scheme system should
-+ construct a Bundle Specifier from the provided package name and
-+ the active locale. For example, when retrieving a message
-+ template for French Canadian, in the <tt>mathlib</tt> package,
-+ the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
-+ program may also retrieve the elements of the current locale
-+ using the no-argument procedures:</p>
-+
-+ <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
-+ <i>symbol</i></tt><br>
-+ <tt><b>current-language</b> <i>symbol</i> ->
-+ undefined</tt><br>
-+ </p>
-+
-+ <blockquote>
-+ When given no arguments, returns the current ISO 639-1
-+ language code as a symbol. If provided with an
-+ argument, the current language is set to that named by the
-+ symbol for the currently executing Scheme thread (or for the
-+ entire Scheme system if such a distinction is not possible).
-+
-+ </blockquote>
-+
-+ <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
-+ <i>symbol</i></tt><br>
-+ <tt><b>current-country</b> <i>symbol</i> ->
-+ undefined</tt><br>
-+ </p>
-+
-+ <blockquote>
-+ returns the current ISO 3166-1 country code as a symbol.
-+ If provided with an argument, the current country is
-+ set to that named by the symbol for the currently executing
-+ Scheme thread (or for the entire Scheme system if such a
-+ distinction is not possible).
-+ </blockquote>
-+
-+ <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
-+ symbol</i></tt>s<br>
-+ <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
-+ undefined</tt><br>
-+ </p>
-+
-+ <blockquote>
-+ Returns a list of additional locale details as a list of
-+ symbols. This list may contain information about
-+ encodings or other more specific information. If
-+ provided with an argument, the current locale details are set
-+ to those given in the currently executing Scheme thread (or
-+ for the entire Scheme system if such a distinction is not
-+ possible).
-+ </blockquote>
-+
-+ <p>The Scheme System should first check for a bundle with the
-+ exact name provided. If no such bundle is found, the last
-+ element from the list is removed and a search is tried for a
-+ bundle with that name. If no bundle is then found, the list is
-+ shortened by removing the last element again. If no message is
-+ found and the bundle specifier is now the empty list, an error
-+ should be raised.</p>
-+
-+ <p>The reason for this search order is to provide the most
-+ locale sensitive template possible, but to fall back on more
-+ general templates if a translation has not yet been provided
-+ for the given locale.</p>
-+
-+ <h3>Message Templates</h3>
-+
-+ <p>A message template is a localized message that may or may
-+ not contain one of a number of formatting codes. A message
-+ template is a Scheme string. The string is of a form that can
-+ be processed by the <tt>format</tt> procedure found in many
-+ Scheme systems and formally specified in SRFI-28 (Basic Format
-+ Strings).</p>
-+
-+ <p>This SRFI also extends SRFI-28 to provide an additional
-+ <tt>format</tt> escape code:</p>
-+
-+ <blockquote>
-+ <tt>~[n]@*</tt> - Causes a value-requiring escape code that
-+ follows this code immediately to reference the [N]'th
-+ optional value absolutely, rather than the next unconsumed
-+ value. The referenced value is <i>not</i> consumed.
-+ </blockquote>
-+ This extension allows optional values to be positionally
-+ referenced, so that message templates can be constructed that
-+ can produce the proper word ordering for a language.
-+
-+ <h3>Preparing Bundles</h3>
-+ Before a bundle may be used by the Scheme system to retrieve
-+ localized template messages, they must be made available to the
-+ Scheme system. This SRFI specifies a way to portably
-+ define the bundles, as well as store them in and retrieve them
-+ from an unspecified system which may be provided by resources
-+ outside the Scheme system.<br>
-+
-+
-+ <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
-+ association-list</i> -> undefined<br>
-+ </tt></p>
-+
-+ <blockquote>
-+ Declares a new bundle named by the given bundle-specifier.
-+ The contents of the bundle are defined by the provided
-+ association list. The list contains associations
-+ between Scheme symbols and the message templates (Scheme
-+ strings) they name. If a bundle already exists with the
-+ given name, it is overwritten with the newly declared
-+ bundle.<br>
-+ </blockquote>
-+ <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
-+ boolean</tt><br>
-+
-+
-+ <blockquote>
-+ Attempts to store a bundle named by the given bundle
-+ specifier, and previously made available using
-+ <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
-+ unspecified mechanism that may be persistent across Scheme
-+ system restarts. If successful, a non-false value is
-+ returned. If unsuccessful, <tt>#f</tt> is returned.<br>
-+ </blockquote>
-+ <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
-+ boolean</tt><br>
-+
-+
-+ <blockquote>
-+ Attempts to retrieve a bundle from an unspecified mechanism
-+ which stores bundles outside the Scheme system. If the
-+ bundle was retrieved successfully, the function returns a
-+ non-false value, and the bundle is immediately available to
-+ the Scheme system. If the bundle could not be found or loaded
-+ successfully, the function returns <tt>#f</tt>, and the
-+ Scheme system's bundle registry remains unaffected.<br>
-+ </blockquote>
-+ A compliant Scheme system may choose not to provide any
-+ external mechanism to store localized bundles. If it does
-+ not, it must still provide implementations for
-+ <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
-+ case, both functions must return <tt>#f</tt> regardless of the
-+ arguments given. Users of this SRFI should recognize that the
-+ inability to load or store a localized bundle in an external
-+ repository is <i>not</i> a fatal error.<br>
-+
-+
-+ <h3>Retrieving Localized Message Templates</h3>
-+
-+ <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
-+ message-template-name</tt></i> <tt>-> <i>string or #f<br>
-+ </i></tt></p>
-+
-+ <blockquote>
-+ Retrieves a localized message template for the given package
-+ name and the given message template name (both symbols).
-+ If no such message could be found, false (#f) is
-+ returned.<br>
-+ <br>
-+ </blockquote>
-+ After retrieving a template, the calling program can use
-+ <tt>format</tt> to produce a string that can be displayed to
-+ the user.<br>
-+
-+
-+ <h2>Examples</h2>
-+ The below example makes use of SRFI-29 to display simple,
-+ localized messages. It also defines its bundles in such a
-+ way that the Scheme system may store and retrieve the bundles
-+ from a more efficient system catalog, if available.<br>
-+
-+<pre>
-+(let ((translations
-+ '(((en) . ((time . "Its ~a, ~a.")
-+ (goodbye . "Goodbye, ~a.")))
-+ ((fr) . ((time . "~1@*~a, c'est ~a.")
-+ (goodbye . "Au revoir, ~a."))))))
-+ (for-each (lambda (translation)
-+ (let ((bundle-name (cons 'hello-program (car translation))))
-+ (if (not (load-bundle! bundle-name))
-+ (begin
-+ (declare-bundle! bundle-name (cdr translation))
-+ (store-bundle! bundle-name)))))
-+ translations))
-+
-+(define localized-message
-+ (lambda (message-name . args)
-+ (apply format (cons (localized-template 'hello-program
-+ message-name)
-+ args))))
-+
-+(let ((myname "Fred"))
-+ (display (localized-message 'time "12:00" myname))
-+ (display #\newline)
-+
-+ (display (localized-message 'goodbye myname))
-+ (display #\newline))
-+
-+;; Displays (English):
-+;; Its 12:00, Fred.
-+;; Goodbye, Fred.
-+;;
-+;; French:
-+;; Fred, c'est 12:00.
-+;; Au revoir, Fred.
-+</pre>
-+
-+ <H1>Implementation</H1>
-+
-+ <p>The implementation requires that the Scheme system provide a
-+ definition for <tt>current-language</tt> and
-+ <tt>current-country</tt> capable of distinguishing the correct
-+ locale present during a Scheme session. The definitions of
-+ those functions in the reference implementation are not capable
-+ of that distinction. Their implementation is provided only so
-+ that the following code can run in any R4RS scheme system.
-+ <br>
-+ </p>
-+
-+ <p>In addition, the below implementation of a compliant
-+ <tt>format</tt> requires SRFI-6 (Basic String Ports) and
-+ SRFI-23 (Error reporting)</p>
-+<pre>
-+;; The association list in which bundles will be stored
-+(define *localization-bundles* '())
-+
-+;; The current-language and current-country functions provided
-+;; here must be rewritten for each Scheme system to default to the
-+;; actual locale of the session
-+(define current-language
-+ (let ((current-language-value 'en))
-+ (lambda args
-+ (if (null? args)
-+ current-language-value
-+ (set! current-language-value (car args))))))
-+
-+(define current-country
-+ (let ((current-country-value 'us))
-+ (lambda args
-+ (if (null? args)
-+ current-country-value
-+ (set! current-country-value (car args))))))
-+
-+;; The load-bundle! and store-bundle! both return #f in this
-+;; reference implementation. A compliant implementation need
-+;; not rewrite these procedures.
-+(define load-bundle!
-+ (lambda (bundle-specifier)
-+ #f))
-+
-+(define store-bundle!
-+ (lambda (bundle-specifier)
-+ #f))
-+
-+;; Declare a bundle of templates with a given bundle specifier
-+(define declare-bundle!
-+ (letrec ((remove-old-bundle
-+ (lambda (specifier bundle)
-+ (cond ((null? bundle) '())
-+ ((equal? (caar bundle) specifier)
-+ (cdr bundle))
-+ (else (cons (car bundle)
-+ (remove-old-bundle specifier
-+ (cdr bundle))))))))
-+ (lambda (bundle-specifier bundle-assoc-list)
-+ (set! *localization-bundles*
-+ (cons (cons bundle-specifier bundle-assoc-list)
-+ (remove-old-bundle bundle-specifier
-+ *localization-bundles*))))))
-+
-+;;Retrieve a localized template given its package name and a template name
-+(define localized-template
-+ (letrec ((rdc
-+ (lambda (ls)
-+ (if (null? (cdr ls))
-+ '()
-+ (cons (car ls) (rdc (cdr ls))))))
-+ (find-bundle
-+ (lambda (specifier template-name)
-+ (cond ((assoc specifier *localization-bundles*) =>
-+ (lambda (bundle) bundle))
-+ ((null? specifier) #f)
-+ (else (find-bundle (rdc specifier)
-+ template-name))))))
-+ (lambda (package-name template-name)
-+ (let loop ((specifier (cons package-name
-+ (list (current-language)
-+ (current-country)))))
-+ (and (not (null? specifier))
-+ (let ((bundle (find-bundle specifier template-name)))
-+ (and bundle
-+ (cond ((assq template-name bundle) => cdr)
-+ ((null? (cdr specifier)) #f)
-+ (else (loop (rdc specifier)))))))))))
-+
-+;;An SRFI-28 and SRFI-29 compliant version of format. It requires
-+;;SRFI-23 for error reporting.
-+(define format
-+ (lambda (format-string . objects)
-+ (let ((buffer (open-output-string)))
-+ (let loop ((format-list (string->list format-string))
-+ (objects objects)
-+ (object-override #f))
-+ (cond ((null? format-list) (get-output-string buffer))
-+ ((char=? (car format-list) #\~)
-+ (cond ((null? (cdr format-list))
-+ (error 'format "Incomplete escape sequence"))
-+ ((char-numeric? (cadr format-list))
-+ (let posloop ((fl (cddr format-list))
-+ (pos (string->number
-+ (string (cadr format-list)))))
-+ (cond ((null? fl)
-+ (error 'format "Incomplete escape sequence"))
-+ ((and (eq? (car fl) '#\@)
-+ (null? (cdr fl)))
-+ (error 'format "Incomplete escape sequence"))
-+ ((and (eq? (car fl) '#\@)
-+ (eq? (cadr fl) '#\*))
-+ (loop (cddr fl) objects (list-ref objects pos)))
-+ (else
-+ (posloop (cdr fl)
-+ (+ (* 10 pos)
-+ (string->number
-+ (string (car fl)))))))))
-+ (else
-+ (case (cadr format-list)
-+ ((#\a)
-+ (cond (object-override
-+ (begin
-+ (display object-override buffer)
-+ (loop (cddr format-list) objects #f)))
-+ ((null? objects)
-+ (error 'format "No value for escape sequence"))
-+ (else
-+ (begin
-+ (display (car objects) buffer)
-+ (loop (cddr format-list)
-+ (cdr objects) #f)))))
-+ ((#\s)
-+ (cond (object-override
-+ (begin
-+ (display object-override buffer)
-+ (loop (cddr format-list) objects #f)))
-+ ((null? objects)
-+ (error 'format "No value for escape sequence"))
-+ (else
-+ (begin
-+ (write (car objects) buffer)
-+ (loop (cddr format-list)
-+ (cdr objects) #f)))))
-+ ((#\%)
-+ (if object-override
-+ (error 'format "Escape sequence following positional override does not require a value"))
-+ (display #\newline buffer)
-+ (loop (cddr format-list) objects #f))
-+ ((#\~)
-+ (if object-override
-+ (error 'format "Escape sequence following positional override does not require a value"))
-+ (display #\~ buffer)
-+ (loop (cddr format-list) objects #f))
-+ (else
-+ (error 'format "Unrecognized escape sequence"))))))
-+ (else (display (car format-list) buffer)
-+ (loop (cdr format-list) objects #f)))))))
-+
-+</pre>
-+
-+ <H1>Copyright</H1>
-+
-+ Copyright (C) Scott G. Miller (2002). All Rights Reserved.
-+
-+ <p>This document and translations of it may be copied and
-+ furnished to others, and derivative works that comment on or
-+ otherwise explain it or assist in its implementation may be
-+ prepared, copied, published and distributed, in whole or in
-+ part, without restriction of any kind, provided that the above
-+ copyright notice and this paragraph are included on all such
-+ copies and derivative works. However, this document itself may
-+ not be modified in any way, such as by removing the copyright
-+ notice or references to the Scheme Request For Implementation
-+ process or editors, except as needed for the purpose of
-+ developing SRFIs in which case the procedures for copyrights
-+ defined in the SRFI process must be followed, or as required to
-+ translate it into languages other than English.</p>
-+
-+ <p>The limited permissions granted above are perpetual and will
-+ not be revoked by the authors or their successors or
-+ assigns.</p>
-+
-+ <p>This document and the information contained herein is
-+ provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
-+ EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
-+ BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
-+ HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
-+ OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
-+ <hr>
-+
-+ <address>
-+ Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
-+ Rush</a>
-+ </address>
-+
-+ <address>
-+ Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
-+ Miller</a>
-+ </address>
-+ <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
-+ <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
-+ Daylight Time 2002 <!-- hhmts end --> <br>
-+ </body>
-+</html>
-+
-diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
---- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-01-31 12:22:58.144502000 -0500
-@@ -0,0 +1,345 @@
-+<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
-+<html>
-+<head>
-+ <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
-+</head>
-+<body>
-+
-+<H1>Title</H1>
-+
-+SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
-+
-+<H1>Author</H1>
-+
-+Andy Gaynor
-+
-+<H1>Status</H1>
-+
-+This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
-+You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
-+<P><UL>
-+<LI>Received: 1999/2/2
-+<LI>Draft: 1999/2/10-1999/04/12
-+<LI>Final: 1999/4/26
-+<LI>Revised reference implementation: 2003/01/27
-+</UL>
-+
-+<H1>Abstract</H1>
-+
-+The <i>named-let</i> incarnation of the <code>let</code> form has two slight
-+inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
-+form makes no accommodation for rest arguments, an issue of functionality
-+and consistency. As defined, the <code>let</code> form does not accommodate
-+signature-style syntax, an issue of aesthetics and consistency. Both
-+issues are addressed here in a manner which is compatible with the traditional
-+<code>let</code> form but for minor extensions.
-+
-+<H1>Rationale</H1>
-+
-+<H2>Signature-style Syntax</H2>
-+
-+Consider the following two equivalent definitions:
-+
-+<p><pre>
-+(define fibonacci
-+ (lambda (n i f0 f1)
-+ (if (= i n)
-+ f0
-+ (fibonacci n (+ i 1) f1 (+ f0 f1)))))
-+
-+(define (fibonacci n i f0 f1)
-+ (if (= i n)
-+ f0
-+ (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+</pre>
-+
-+Although there is a named-let analog for the former form, there is none
-+for the latter. To wit, suppose one wished to compute the 10th element
-+of the Fibonacci sequence using a named let:
-+
-+<p>
-+<pre>
-+(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
-+ (if (= i n)
-+ f0
-+ (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+Values: 55
-+</pre>
-+
-+As it stands, one cannot equivalently write
-+
-+<p>
-+<pre>
-+(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
-+ (if (= i n)
-+ f0
-+ (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+</pre>
-+
-+which is consistent with <code>define</code>'s signature-style form.
-+<p>Those that favor the signature style may prefer this extension.
-+In any case, it may be more appropriate to include all bound names within
-+the binding section. As presented, this straightforward extension
-+introduces no ambiguity or incompatibility with the existing definition
-+of let.
-+
-+<H2>Rest Arguments</H2>
-+
-+As it stands, one cannot write a named let with rest arguments, as in
-+
-+<p>
-+<pre>
-+(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
-+ (if (null? x)
-+ 'just-a-silly-contrived-example
-+ (begin
-+ (write (car x) port)
-+ (apply blast port (cdr x)))))
-+</pre>
-+
-+otherwise equivalent to
-+
-+<p>
-+<pre>
-+(letrec ((blast (lambda (port . x)
-+ (if (null? x)
-+ 'just-a-silly-contrived-example
-+ (begin
-+ (write (car x) port)
-+ (apply blast port (cdr x)))))))
-+ (blast (current-output-port) (+ 1 2) 4 5))
-+</pre>
-+
-+While this example is rather contrived, the functionality is not.
-+There are several times when the author has used this construct in practice.
-+Regardless, there is little reason to deny the <code>let</code> form access to
-+all the features of lambda functionality.
-+
-+<H2>Symbols in Binding Sections</H2>
-+
-+Both the features above rely upon the placement of symbols in <code>let</code>
-+binding lists (this statement is intentially simplistic). The only
-+other apparent use of such symbol placement is to tersely bind variables
-+to unspecified values. For example, one might desire to use
-+<code>(let (foo bar baz) ...)</code>
-+to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
-+unspecified values.
-+
-+<p>This usage is considered less important in light of the rationales
-+presented above, and an alternate syntax is immediately apparent, as
-+in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
-+be preferable, consistently parenthesizing normal binding clauses.
-+
-+<H1>Specification</H1>
-+
-+<H2>Syntax</H2>
-+
-+<p>
-+A formal specification of the syntax follows. Below, body, expression,
-+and identifier are free. Each instantiation of binding-name must be
-+unique.
-+</p>
-+
-+<p>
-+<pre>
-+ let = "(" "let" let-bindings body ")"
-+ expressions = nothing | expression expressions
-+ let-bindings = let-name bindings
-+ |"(" let-name "." bindings ")"
-+ let-name = identifier
-+ bindings = "(" ")"
-+ | rest-binding
-+ | "(" normal-bindings ["." rest-binding] ")"
-+normal-bindings = nothing
-+ | normal-binding normal-bindings
-+normal-binding = "(" binding-name expression ")"
-+ binding-name = identifier
-+ rest-binding = "(" binding-name expressions ")"
-+</pre>
-+
-+<p>
-+For clarity and convenience, an informal specification follows.
-+</p>
-+
-+<ol>
-+<li><a name="unnamed">Unnamed</a>
-+
-+<p><pre>
-+(let ((<parameter> <argument>)...)
-+ <body>...)
-+</pre>
-+</li>
-+
-+<li><a name="named-non-sig">
-+Named, non-signature-style, no rest argument</a>
-+
-+<p><pre>
-+(let <name> ((<parameter> <argument>)...)
-+ <body>...)
-+</pre>
-+</li>
-+
-+<li><a name="named-sig">Named, signature-style, no rest argument</a>
-+
-+<p><pre>
-+(let (<name> (<parameter> <argument>)...)
-+ <body>...)
-+</pre>
-+</li>
-+
-+<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
-+
-+<p><pre>
-+(let <name> ((<parameter> <argument>)...
-+
-+. (<rest-parameter> <rest-argument>...))
-+ <body>...)
-+</pre>
-+
-+<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
-+
-+<p><pre>
-+(let (<name> (<parameter> <argument>)...
-+
-+. (<rest-parameter> <rest-argument>...))
-+ <body>...)
-+</pre>
-+</li>
-+</ol>
-+
-+<H2>Semantics</H2>
-+
-+Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
-+and <code>letrec</code> forms, respectively.
-+
-+<ul>
-+<li>For informal syntax 1:
-+
-+<p><pre>
-+(($lambda (<parameter>...) <body>...) <argument>...)
-+</pre>
-+</li>
-+
-+<li>For informal syntaxes 2 and 3:
-+
-+<p>
-+<pre>
-+($letrec ((<name> ($lambda (<parameter>...) <body>...)))
-+ (<name> <argument>...))
-+</pre>
-+</li>
-+
-+<li>For informal syntaxes 4 and 5:
-+
-+<p>
-+<pre>
-+($letrec ((<name> ($lambda (<parameter>...
-+
-+. <rest-parameter>) <body>...)))
-+ (<name> <argument>... <rest-argument>...))
-+</pre>
-+</li>
-+</ul>
-+
-+<H1>Implementation</H1>
-+
-+Here is an implementation using <code>SYNTAX-RULES</code>.
-+
-+<p>
-+<pre>
-+;; Use your own standard let.
-+;; Or call a lambda.
-+;; (define-syntax standard-let
-+;;
-+;; (syntax-rules ()
-+;;
-+;; ((let ((var val) ...) body ...)
-+;; ((lambda (var ...) body ...) val ...))))
-+
-+(define-syntax let
-+
-+ (syntax-rules ()
-+
-+ ;; No bindings: use standard-let.
-+ ((let () body ...)
-+ (standard-let () body ...))
-+ ;; Or call a lambda.
-+ ;; ((lambda () body ...))
-+
-+ ;; All standard bindings: use standard-let.
-+ ((let ((var val) ...) body ...)
-+ (standard-let ((var val) ...) body ...))
-+ ;; Or call a lambda.
-+ ;; ((lambda (var ...) body ...) val ...)
-+
-+ ;; One standard binding: loop.
-+ ;; The all-standard-bindings clause didn't match,
-+ ;; so there must be a rest binding.
-+ ((let ((var val) . bindings) body ...)
-+ (let-loop #f bindings (var) (val) (body ...)))
-+
-+ ;; Signature-style name: loop.
-+ ((let (name binding ...) body ...)
-+ (let-loop name (binding ...) () () (body ...)))
-+
-+ ;; defun-style name: loop.
-+ ((let name bindings body ...)
-+ (let-loop name bindings () () (body ...)))))
-+
-+(define-syntax let-loop
-+
-+ (syntax-rules ()
-+
-+ ;; Standard binding: destructure and loop.
-+ ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
-+ (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
-+
-+ ;; Rest binding, no name: use standard-let, listing the rest values.
-+ ;; Because of let's first clause, there is no "no bindings, no name" clause.
-+ ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-+ (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
-+ ;; Or call a lambda with a rest parameter on all values.
-+ ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
-+ ;; Or use one of several other reasonable alternatives.
-+
-+ ;; No bindings, name: call a letrec'ed lambda.
-+ ((let-loop name () (var ...) (val ...) body)
-+ ((letrec ((name (lambda (var ...) . body)))
-+ name)
-+ val ...))
-+
-+ ;; Rest binding, name: call a letrec'ed lambda.
-+ ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
-+ ((letrec ((name (lambda (var ... . rest-var) . body)))
-+ name)
-+ val ... rest-val ...))))
-+</pre>
-+
-+
-+<H1>Copyright</H1>
-+
-+Copyright (C) Andy Gaynor (1999). All Rights Reserved.
-+<p>This document and translations of it may be copied and furnished to
-+others, and derivative works that comment on or otherwise explain it or
-+assist in its implementation may be prepared, copied, published and distributed,
-+in whole or in part, without restriction of any kind, provided that the
-+above copyright notice and this paragraph are included on all such copies
-+and derivative works. However, this document itself may not be modified
-+in any way, such as by removing the copyright notice or references to the
-+Scheme Request For Implementation process or editors, except as needed
-+for the purpose of developing SRFIs in which case the procedures for copyrights
-+defined in the SRFI process must be followed, or as required to translate
-+it into languages other than English.
-+<p>The limited permissions granted above are perpetual and will not be
-+revoked by the authors or their successors or assigns.
-+<p>This document and the information contained herein is provided on an
-+"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
-+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
-+USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
-+WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
-+
-+ <hr>
-+ <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
-+
-+</body>
-+</html>
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/info.rkt racket-6.12/share/pkgs/srfi-lib/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/info.rkt 2018-01-26 16:10:02.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/info.rkt 2018-01-31 12:22:58.144502000 -0500
-@@ -1 +1,15 @@
--(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("scheme-lib" "base" "srfi-lite-lib" "r6rs-lib" "compatibility-lib"))) (define implies (quote ("srfi-lite-lib"))) (define pkg-desc "implementation (no documentation) part of \"srfi\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define deps '("scheme-lib"
-+ "base"
-+ "srfi-lite-lib"
-+ "r6rs-lib"
-+ "compatibility-lib"))
-+
-+(define implies '("srfi-lite-lib"))
-+
-+(define pkg-desc "implementation (no documentation) part of \"srfi\"")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/sort.txt racket-6.12/share/pkgs/srfi-lib/srfi/32/sort.txt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/sort.txt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/32/sort.txt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,1069 +0,0 @@
--The SRFI-32 sort libraries -*- outline -*-
--Olin Shivers
--First draft: 1998/10/19
--Last update: 2002/7/21
--
--[Todo: del-list-neighbor-dups!
-- vector-copy -> subvector
-- use srfi-23 for reporting errors
-- use srfi-16 for n-aries?
--
--Emacs should display this document in outline mode. Say c-h m for
--instructions on how to move through it by sections (e.g., c-c c-n, c-c c-p).
--
--* Table of contents
---------------------
--Abstract
--Procedure index
--Introduction
--What's wrong with the current state of affairs?
--Design rules
-- What vs. how
-- Consistency across function signatures
-- Data parameter first, less-than parameter after
-- Ordering, comparison functions & stability
-- All vector operations accept optional subrange parameters
-- Required vs. allowed side-effects
--Procedure specification
-- Procedure naming and functionality
-- Types of parameters and return values
-- sort-lib - general sorting package
-- Algorithm-specific sorting packages
--Algorithmic properties
--Topics to be resolved during discussion phase
--Porting and optimisation
--References & Links
--Acknowledgements
--Copyright
--
--
--* Abstract
------------
--Current Scheme sorting packages are, every one of them, surprisingly bad. I've
--designed the API for a full-featured sort toolkit, which I propose as a SRFI.
--
--The spec comes with 1200 lines of high-quality reference code: tightly
--written, highly commented, portable code, available for free. Implementors
--want this code. It's better than what you have.
--
---------------------------------------------------------------------------------
--* Procedure index
-------------------
--list-sorted? vector-sorted?
--
--list-merge vector-merge
--list-sort vector-sort
--list-stable-sort vector-stable-sort
--list-delete-neighbor-dups vector-delete-neighbor-dups
--
--list-merge! vector-merge!
--list-sort! vector-sort!
--list-stable-sort! vector-stable-sort!
--list-delete-neighbor-dups! vector-delete-neighbor-dups!
--
--quick-sort heap-sort insert-sort list-merge-sort vector-merge-sort
--quick-sort! heap-sort! insert-sort! list-merge-sort! vector-merge-sort!
--quick-sort3!
--
--vector-binary-search
--vector-binary-search3
--
---------------------------------------------------------------------------------
--* Introduction
----------------
--As I'll detail below, I wasn't very happy with the state of the Scheme
--world for sorting and merging lists and vectors. So I have designed and
--written a fairly comprehensive sorting & merging toolkit. It is
--
-- - very portable,
--
-- - much better code than what is currently in Elk, Gambit, Bigloo,
-- Scheme->C, MzScheme, RScheme, Scheme48, MIT Scheme, or slib, and
--
-- - priced to move: free code.
--
--The package includes
-- - Vector insert sort (stable)
-- - Vector heap sort
-- - Vector quick sort (with median-of-3 pivot picking)
-- - Vector merge sort (stable)
-- - Pure and destructive list merge sort (stable)
-- - Stable vector and list merge
-- - Miscellaneous sort-related procedures: Vector and list merging,
-- sorted? predicates, vector binary search, vector and list
-- delete-equal-neighbor procedures.
-- - A general, non-algorithmic set of procedure names for general sorting
-- and merging.
--
--Scheme programmers may want to adopt this package. I'd like Scheme
--implementors to adopt this code and its API -- in fact, the code is a bribe to
--make it easy for implementors to converge on the suggested API. I mean, you'd
--really have to be a boor to take this free code I wrote and mutate its
--interface over to your incompatible, unportable API, wouldn't you? But you
--could, of course -- it's freely available. More in the spirit of the offering,
--you could make this API available, and then also write a little module
--providing your old interface that is defined in terms of this API. "Scheme
--implementors," in this context, includes slib, which is not a standalone
--implementation of Scheme, but rather an influential collection of API's and
--code.
--
--The code is tightly bummed. It is clearly written, and commented in my usual
--voluminous style. This includes notes on porting and implementation-specific
--optimisations.
--
--
---------------------------------------------------------------------------------
--* What's wrong with the current state of affairs?
---------------------------------------------------
--
--It's just amazing to me that in 2002, sorting and merging hasn't been
--completely put to bed. These are well-understood algorithms, each of them well
--under a page of code. The straightforward algorithms are basic, core stuff --
--sophomore-level. But if you tour the major Scheme implementations out there on
--the Net, you find badly written code that provides extremely spotty coverage
--of the algorithm space. One implementation even has a buggy implementation
--that has been in use for about 20 years. Another has an O(n^2) algorithm...
--implemented in C for speed.
--
--Open source-code is a wonderful thing. In a couple of hours, I was able to
--download and check the sources of 9 Scheme systems. Here are my notes from the
--systems I checked. You can skip to the next section if you aren't morbidly
--curious.
--
--slib
-- sorted? vector-or-list <
-- merge list1 list2 <
-- merge! list1 list2 <
-- sort vector-or-list <
-- sort! vector-or-list <
--
-- Richard O'Keefe's stable list merge sort is right idea, but implemented
-- using gratuitous variable side effects. It also does redundant SET-CDR!s.
-- The vector sort converts to list, merge sorts, then reconverts
-- to vector. This is a bad idea -- non-local pointer chasing bad; vector
-- shuffling good. If you must allocate temp storage, might as well allocate
-- a temp vector and use vector merge sort.
--
--MIT Scheme
-- sort! vector <
-- merge-sort! vector <
-- quick-sort! vector <
--
-- sort vector-or-list <
-- merge-sort vector-or-list <
-- quick-sort vector-or-list <
--
-- Naive vector quicksort: loser, for worst-case performance reasons.
-- List sort by "list->vector; quicksort; vector->list," hence also loser.
-- A clever stable vector merge sort, albeit not very bummed.
--
--Scheme 48 & T
-- sort-list list <
-- sort-list! list <
-- list-merge! list1 list2 <
--
-- Bob Nix's implementation of online merge-sort, written in the early 80's.
-- Conses unnecessary bookkeeping structure, which isn't necessary with a
-- proper recursive formulation. Also, does redundant SET-CDR!s. No vector
-- sort. Also, has a bug -- is claimed to be a stable sort, but isn't! To see
-- this, get the S48 code, and try
-- (define (my< x y) (< (abs x) (abs y)))
-- (list-merge! (list 0 2) (list -2) my<) ; -> (0 2 -2)
-- (list-merge! (list 2) (list 0 -2) my<) ; -> (0 -2 2)
-- This could be fixed very easily, but it isn't worth it given the
-- other problems with the algorithm.
--
--RScheme
-- vector-sort! vector <
-- sort collection <
--
-- Good basic implementation of vector heapsort, which has O(n lg n)
-- worst-case time. Code ugly, needs tuning. List sort by "list->vector;
-- sort; vector->list." Nothing for stable sorting.
--
--MzScheme
-- quicksort lis <
-- mergesort alox <
--
-- Sorts lists with (list->vector; quicksort; vector->list) -- but the core
-- quicksort is not available for vector sorting. Nothing for stable sorting.
-- Quicksort picks pivot naively, inducing O(n^2) worse-case behaviour on a
-- fairly common case: an already-sorted list.
--
--Bigloo, STK
-- sort vector-or-list <
-- Uses an O(n^2) algorithm... implemented in C for speed. Hmm.
-- (See runtime/Ieee/vector.scm and runtime/Clib/cvector.c)
--
--Gambit
-- sort-list list <
-- Nothing for vectors. Simple, slow, unstable merge sort for lists.
--
--Elk
-- Another naive quicksort. Lists handled by converting to vector.
-- sort vector-or-list <
-- sort! vector-or-list <
--
--Chez Scheme
-- merge < list1 list2
-- merge! < list1 list2
-- sort < list
-- sort! < list
--
-- These are stable. I have not seen the source code.
--
--Common Lisp
-- sort sequence < [key]
-- stable-sort sequence < [key]
-- merge result-type sequence1 sequence2 < [key]
--
-- The sort procedures are allowed, but not required, to be destructive.
--
--SML/NJ
-- sort: ('a*'a -> bool) -> 'a list -> 'a list
-- "Smooth applicative merge sort," which is stable.
-- There is also a highly bummed quicksort for vectors.
--
--The right solution: Implement a full toolbox of carefully written standard sort
--routines.
--
--Having the source of all these above-cited Schemes available for study made
--life a lot easier writing this code. I appreciate the authors making their
--source available under such open terms.
--
--
---------------------------------------------------------------------------------
--* Design rules
----------------
--
--** What vs. how
--===============
--There are two different interfaces: "what" (simple) & "how" (detailed).
--
-- - Simple: you specify semantics: datatype (list or vector),
-- mutability, and stability.
--
-- - Detailed: you specify the actual algorithm (quick, heap,
-- insert, merge). Different algorithms have different properties,
-- both semantic & pragmatic, so these exports are necessary.
--
-- It is necessarily the case that the specifications of these procedures
-- make statements about execution "pragmatics." For example, the sole
-- distinction between heap sort and quick sort -- both of which are
-- provided by this library -- is one of execution time, which is not a
-- "semantic" distinction. Similar resource-use statements are made about
-- "iterative" procedures, meaning that they can execute on input of
-- arbitrary size in a constant number of stack frames.
--
--** Consistency across function signatures
--=========================================
--The two interfaces share common function signatures wherever
--possible, to facilitate switching a given call from one procedure
--to another.
--
--** Less-than parameter first, data parameter after
--==================================================
--These procedures uniformly observe the following parameter order:
--the data to be sorted comes after the comparison function.
--That is, we write
-- (sort < lis)
--not
-- (sort lis <).
--
--With the sole exception of Chez Scheme, this is the exact opposite of
--every sort function out there in current use in the Scheme world. (See
--the summary of related APIs above.) However, it is consistent with common
--practice across Scheme libraries in general to put the ordering function
--first -- the "operation currying" convention. (E.g., consider FOR-EACH or
--MAP or FIND.)
--
--The original draft of this SRFI used the data-first/comparison-last convention
--for backwards compatibility -- a decision I made with internal misgivings.
--Happily, however, the overwhelming response from the discussion phase
--supported "cleaning up" this issue and re-converging the parameter order with
--the general Scheme "op currying" convention. So the original decision was
--inverted in favor of the comparison-first/data-last convention.
--
--** Ordering, comparison functions & stability
--=============================================
--These routines take a < comparison function, not a <= comparison
--function, and they sort into increasing order. The difference between
--a < spec and a <= spec comes up in three places:
-- - the definition of an ordered or sorted data set,
-- - the definition of a stable sorting algorithm, and
-- - correctness of quicksort.
--
--+ We say that a data set (a list or vector) is *sorted* or *ordered*
-- if it contains no adjacent pair of values ... X Y ... such that Y < X.
--
-- In other words, scanning across the data never takes a "downwards" step.
--
-- If you use a <= procedure where these algorithms expect a <
-- procedure, you may not get the answers you expect. For example,
-- the LIST-SORTED? function will return false if you pass it a <= comparison
-- function and an ordered list containing adjacent equal elements.
--
--+ A "stable" sort is one that preserves the pre-existing order of equal
-- elements. Suppose, for example, that we sort a list of numbers by
-- comparing their absolute values, i.e., using comparison function
-- (lambda (x y) (< (abs x) (abs y)))
-- If we sort a list that contains both 3 and -3:
-- ... 3 ... -3 ...
-- then a stable sort is an algorithm that will not swap the order
-- of these two elements, that is, the answer is guaranteed to to look like
-- ... 3 -3 ...
-- not
-- ... -3 3 ...
--
-- Choosing < for the comparison function instead of <= affects how stability
-- is coded. Given an adjacent pair X Y, (< y x) means "Y should be moved in
-- front of X" -- otherwise, leave things as they are. So using a <= function
-- where a < function is expected will *invert* stability.
--
-- This is due to the definition of equality, given a < comparator:
-- (and (not (< x y))
-- (not (< y x)))
-- The definition is rather different, given a <= comparator:
-- (and (<= x y)
-- (<= y x))
--
--+ A "stable" merge is one that reliably favors one of its data sets
-- when equal items appear in both data sets. *All merge operations in
-- this library are stable*, breaking ties between data sets in favor
-- of the first data set -- elements of the first list come before equal
-- elements in the second list.
--
-- So, if we are merging two lists of numbers ordered by absolute value,
-- the stable merge operation LIST-MERGE
-- (list-merge (lambda (x y) (< (abs x) (abs y)))
-- '(0 -2 4 8 -10) '(-1 3 -4 7))
-- reliably places the 4 of the first list before the equal-comparing -4
-- of the second list:
-- (0 -1 -2 4 -4 7 8 -10)
--
--+ Some sort algorithms will *not work correctly* if given a <= when they
-- expect a < comparison (or vice-versa). For example, violating quicksort's
-- spec may cause it to produce wrong answers, diverge, raise an error, or do
-- some fourth thing. To see why, consider the left-scan part of the standard
-- quicksort partition step:
-- (let ((i (let scan ((i i)) (if (elt< (vector-ref v i) pivot)
-- (scan (+ i 1))
-- i))))
-- ...)
-- Consider applying this loop to a vector of all zeroes (hence, PIVOT, as
-- well, is zero), but erroneously using <= for the ELT< function. The loop
-- will scan right off the end of the vector, producing a vector-index error.
-- The guarantee that the scan loop will terminate before running off the end
-- of the vector depends critically upon ELT< performing as a true, irreflexive
-- < relation. Running off the end of the vector is only one of a variety of
-- possibly ways to lose -- other, variant implementations of quicksort can,
-- instead, loop forever on some data sets if ELT< is a <= predicate.
--
--In short, if your comparison function F answers true to (F x x), then
-- - using a stable sorting or merging algorithm will not give you a
-- stable sort or merge,
-- - LIST-SORTED? may surprise you, and
-- - quicksort may fail in a variety of possible ways.
--Note that you can synthesize a < function from a <= function with
-- (lambda (x y) (not (<= y x)))
--if need be.
--
--Precise definitions give sharp edges to tools, but require care in use.
--"Measure twice, cut once."
--
--I have adopted the choice of < from Common Lisp. One would assume the definers
--of Common Lisp had a good reason for adopting < instead of <=, but canvassing
--several of the principal actors in the definition process has turned up no
--better reason than "an arbitrary but consistent choice." At minimum, then,
--this SRFI extends the coverage of that consistent choice.
--
--** All vector operations accept optional subrange parameters
--============================================================
--The vector operations specified below all take optional START/END arguments
--indicating a selected subrange of a vector's elements. If a START parameter or
--START/END parameter pair is given to such a procedure, they must be exact,
--non-negative integers, such that
-- 0 <= START <= END <= (VECTOR-LENGTH V)
--where V is the related vector parameter. If not specified, they default to 0
--and the length of the vector, respectively. They are interpreted to select the
--range [START,END), that is, all elements from index START (inclusive) up to,
--but not including, index END.
--
--** Required vs. allowed side-effects
--====================================
--LIST-SORT! and LIST-STABLE-SORT! are allowed, but not required,
--to alter their arguments' cons cells to construct the result list. This is
--consistent with the what-not-how character of the group of procedures
--to which they belong (the "sort-lib" package).
--
--The LIST-DELETE-NEIGHBOR-DUPS!, LIST-MERGE! and LIST-MERGE-SORT! procedures,
--on the other hand, provide specific algorithms, and, as such, explicitly
--commit to the use of side-effects on their input lists in order to guarantee
--their key algorithmic properties (e.g., linear-time operation, constant-space
--stack use).
--
---------------------------------------------------------------------------------
--* Procedure specification
---------------------------
--The procedures are split into several packages. In a Scheme system that has a
--module or package system, these procedures should be contained in modules
--named as follows:
-- Package name Functionality
-- ------------ -------------
-- sort-lib General sorting for lists & vectors
-- sorted?-lib Sorted predicates for lists & vectors
-- list-merge-sort-lib List merge sort
-- vector-merge-sort-lib Vector merge sort
-- vector-heap-sort-lib Vector heap sort
-- vector-quick-sort-lib Vector quick sort
-- vector-insert-sort-lib Vector insertion sort
-- delndup-lib List and vector delete neighbor duplicates
-- binsearch-lib Vector binary search
--
--A Scheme system without a module system should provide all of the bindings
--defined in all of these modules as components of the "SRFI-32" package.
--
--Note that there is no "list insert sort" package, as you might as well always
--use list merge sort. The reference implementation's destructive list merge
--sort will do fewer SET-CDR!s than a destructive insert sort.
--
--** Procedure naming and functionality
--=====================================
--Almost all of the procedures described below are variants of two basic
--operations: sorting and merging. These procedures are consistently named
--by composing a set of basic lexemes to indicate what they do.
--
--Lexeme Meaning
-------- -------
--"sort" The procedure sorts its input data set by some < comparison function.
--
--"merge" The procedure merges two ordered data sets into a single ordered
-- result.
--
--"stable" This lexeme indicates that the sort is a stable one.
--
--"vector" The procedure operates upon vectors.
--
--"list" The procedure operates upon lists.
--
--"!" Procedures that end in "!" are allowed, and sometimes required,
-- to reuse their input storage to construct their answer.
--
--** Types of parameters and return values
--========================================
--In the procedures specified below,
-- - A LIS parameter is a list;
--
-- - A V parameter is a vector;
--
-- - A < or = parameter is a procedure accepting two arguments taken from the
-- specified procedure's data set(s), and returning a boolean;
--
-- - START and END parameters are exact, non-negative integers that
-- serve as vector indices selecting a subrange of some associated vector.
-- When specified, they must satisfy the relation
-- 0 <= start <= end <= (vector-length v)
-- where V is the associated vector.
--
--Passing values to procedures with these parameters that do not satisfy these
--types is an error.
--
--If a procedure is said to return "unspecified," this means that nothing at all
--is said about what the procedure returns, not even the number of return
--values. Such a procedure is not even required to be consistent from call to
--call in the nature or number of its return values. It is simply required to
--return a value (or values) that may be passed to a command continuation, e.g.
--as the value of an expression appearing as a non-terminal subform of a BEGIN
--expression. Note that in R5RS, this restricts such a procedure to returning a
--single value; non-R5RS systems may not even provide this restriction.
--
--** sort-lib - general sorting package
--=====================================
--This library provides basic sorting and merging functionality suitable for
--general programming. The procedures are named by their semantic properties,
--i.e., what they do to the data (sort, stable sort, merge, and so forth).
--
-- Procedure Suggested algorithm
-- -------------------------------------------------------------------------
-- list-sorted? < lis -> boolean
-- list-merge < lis1 lis2 -> list
-- list-merge! < lis1 lis2 -> list
-- list-sort < lis -> list (vector heap or quick)
-- list-sort! < lis -> list (list merge sort)
-- list-stable-sort < lis -> list (vector merge sort)
-- list-stable-sort! < lis -> list (list merge sort)
-- list-delete-neighbor-dups = lis -> list
-- list-delete-neighbor-dups! = lis -> list
--
-- vector-sorted? < v [start end] -> boolean
-- vector-merge < v1 v2 [start1 end1 start2 end2] -> vector
-- vector-merge! < v v1 v2 [start start1 end1 start2 end2] -> unspecified
-- vector-sort < v [start end] -> vector (heap or quick sort)
-- vector-sort! < v [start end] -> unspecified (heap or quick sort)
-- vector-stable-sort < v [start end] -> vector (vector merge sort)
-- vector-stable-sort! < v [start end] -> unspecified (vector merge sort)
-- vector-delete-neighbor-dups = v [start end] -> vector
-- vector-delete-neighbor-dups! = target source [t-start s-start s-end] -> t-end
--
-- LIST-SORTED? and VECTOR-SORTED? return true if their input list or vector
-- is in sorted order, as determined by their < comparison parameter.
--
-- All four merge operations are stable: an element of the initial list LIS1
-- or vector V1 will come before an equal-comparing element in the second
-- list LIS2 or vector V2 in the result.
--
-- The procedures
-- LIST-MERGE
-- LIST-SORT
-- LIST-STABLE-SORT
-- LIST-DELETE-NEIGHBOR-DUPS
-- do not alter their inputs and are allowed to return a value that shares
-- a common tail with a list argument.
--
-- The procedures
-- LIST-SORT!
-- LIST-STABLE-SORT!
-- are "linear update" operators -- they are allowed, but not required, to
-- alter the cons cells of their arguments to produce their results.
--
-- On the other hand, the procedures
-- LIST-DELETE-NEIGHBOR-DUPS!
-- LIST-MERGE!
-- make only a single, iterative, linear-time pass over their argument lists,
-- using SET-CDR!s to rearrange the cells of the lists into the final result
-- -- they work "in place." Hence, any cons cell appearing in the result must
-- have originally appeared in an input. The intent of this
-- iterative-algorithm commitment is to allow the programmer to be sure that
-- if, for example, LIST-MERGE! is asked to merge two ten-million-element
-- lists, the operation will complete without performing some extremely
-- (possibly twenty-million) deep recursion.
--
-- The vector procedures
-- VECTOR-SORT
-- VECTOR-STABLE-SORT
-- VECTOR-DELETE-NEIGHBOR-DUPS
-- do not alter their inputs, but allocate a fresh vector for their result,
-- of length END - START.
--
-- The vector procedures
-- VECTOR-SORT!
-- VECTOR-STABLE-SORT!
-- sort their data in-place. (But note that VECTOR-STABLE-SORT! may
-- allocate temporary storage proportional to the size of the input --
-- I am not aware of O(n lg n) stable vector-sorting algorithms that
-- run in constant space.)
--
-- VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
--
-- VECTOR-MERGE! writes its result into vector V, beginning at index START,
-- for indices less than END = START + (END1-START1) + (END2-START2). The
-- target subvector
-- V[start,end)
-- may not overlap either source subvector
-- V1[start1,end1)
-- V2[start2,end2).
--
-- The ...-DELETE-NEIGHBOR-DUPS-... procedures:
-- These procedures delete adjacent duplicate elements from a list or a
-- vector, using a given element-equality procedure. The first/leftmost
-- element of a run of equal elements is the one that survives. The list or
-- vector is not otherwise disordered.
--
-- These procedures are linear time -- much faster than the O(n^2) general
-- duplicate-element deletors that do not assume any "bunching" of elements
-- (such as the ones provided by SRFI-1). If you want to delete duplicate
-- elements from a large list or vector, you can sort the elements to bring
-- equal items together, then use one of these procedures, for a total time
-- of O(n lg n).
--
-- The comparison function = passed to these procedures is always applied
-- (= x y)
-- where X comes before Y in the containing list or vector.
--
-- - LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
-- may share storage with the input list.
--
-- - VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
-- rather allocates a fresh vector to hold the result.
--
-- - LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
-- mutate its input list in order to construct its answer.
--
-- - VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
-- answer, packing its answer into the index range [start,end'), where
-- END' is the non-negative exact integer returned as its value. It
-- returns END' as its result. The vector is not altered outside the range
-- [start,end').
--
-- - VECTOR-DELETE-NEIGHBOR-DUPS! scans vector SOURCE in range
-- [S-START,S-END), writing its result to vector TARGET beginning at index
-- T-START. It returns exact, non-negative integer T-END, which indicates
-- that the results of the operation are found in index range
-- [T-START,T-END) of TARGET; elements of TARGET outside this range
-- are unaltered.
--
-- It is an error for memory cell TARGET[T-START] to be a memory cell in
-- the region SOURCE[1 + S-START, S-END). In a Scheme implementation
-- that does not allow distinct vectors to share storage, this means
-- that one of the following must be true:
-- 1. (not (eq? source target))
-- 2. t-start not-in [s-start + 1, s-end)
--
-- - Examples:
-- (list-delete-neighbor-dups = '(1 1 2 7 7 7 0 -2 -2))
-- => (1 2 7 0 -2)
--
-- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2))
-- => #(1 2 7 0 -2)
--
-- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2) 3 7)
-- => #(7 0 -2)
--
-- ;; Result left in v[3,9):
-- (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
-- (cons (vector-delete-neighbor-dups! = v 3)
-- v))
-- => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
--
--
--** Algorithm-specific sorting packages
--======================================
--These packages provide more specific sorting functionality, that is,
--specific commitment to particular algorithms that have particular
--pragmatic consequences (such as memory locality, asymptotic running time)
--beyond their semantic behaviour (sorting, stable sorting, merging, etc.).
--Programmers that need a particular algorithm can use one of these packages.
--
--sorted?-lib - sorted predicates
-- list-sorted? < lis -> boolean
-- vector-sorted? < v [start end] -> boolean
--
-- Return #f iff there is an adjacent pair ... X Y ... in the input
-- list or vector such that Y < X. The optional START/END range
-- arguments restrict VECTOR-SORTED? to the indicated subvector.
--
--list-merge-sort-lib - list merge sort
-- list-merge-sort < lis -> list
-- list-merge-sort! < lis -> list
-- list-merge lis1 < lis2 -> list
-- list-merge! lis1 < lis2 -> list
--
-- The sort procedures sort their data using a list merge sort, which is
-- stable. (The reference implementation is, additionally, a "natural" sort.
-- See below for the properties of this algorithm.)
--
-- The ! procedures are destructive -- they use SET-CDR!s to rearrange the
-- cells of the lists into the proper order. As such, they do not allocate
-- any extra cons cells -- they are "in place" sorts. Additionally,
-- LIST-MERGE! is iterative -- it can operate on arguments of arbitrary size
-- with a constant number of stack frames.
--
-- The merge operations are stable: an element of LIS1 will come before an
-- equal-comparing element in LIS2 in the result list.
--
--vector-merge-sort-lib - vector merge sort
-- vector-merge-sort < v [start end temp] -> vector
-- vector-merge-sort! < v [start end temp] -> unspecified
-- vector-merge < v1 v2 [start1 end1 start2 end2] -> vector
-- vector-merge! < v v1 v2 [start start1 end1 start2 end2] -> unspecified
--
-- The sort procedures sort their data using vector merge sort, which is
-- stable. (The reference implementation is, additionally, a "natural" sort.
-- See below for the properties of this algorithm.)
--
-- The optional START/END arguments provide for sorting of subranges, and
-- default to 0 and the length of the corresponding vector.
--
-- Merge-sorting a vector requires the allocation of a temporary "scratch"
-- work vector for the duration of the sort. This scratch vector can be
-- passed in by the client as the optional TEMP argument; if so, the supplied
-- vector must be of size >= END, and will not be altered outside the range
-- [start,end). If not supplied, the sort routines allocate one themselves.
--
-- The merge operations are stable: an element of V1 will come before an
-- equal-comparing element in V2 in the result vector.
--
-- VECTOR-MERGE-SORT! leaves its result in V[start,end).
--
-- VECTOR-MERGE-SORT returns a vector of length END-START.
--
-- VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
--
-- VECTOR-MERGE! writes its result into vector V, beginning at index START,
-- for indices less than END = START + (END1-START1) + (END2-START2). The
-- target subvector
-- V[start,end)
-- may not overlap either source subvector
-- V1[start1,end1)
-- V2[start2,end2).
--
--vector-heap-sort-lib - vector heap sort
-- heap-sort < v [start end] -> vector
-- heap-sort! < v [start end] -> unspecified
--
-- These procedures sort their data using heap sort,
-- which is not a stable sorting algorithm.
--
-- HEAP-SORT returns a vector of length END-START.
-- HEAP-SORT! is in-place, leaving its result in V[start,end).
--
--vector-quick-sort-lib - vector quick sort
-- quick-sort < v [start end] -> vector
-- quick-sort! < v [start end] -> unspecified
-- quick-sort3! c v [start end] -> unspecified
--
-- These procedures sort their data using quick sort,
-- which is not a stable sorting algorithm.
--
-- QUICK-SORT returns a vector of length END-START.
-- QUICK-SORT! is in-place, leaving its result in V[start,end).
--
-- QUICK-SORT3! is a variant of quick-sort that takes a three-way
-- comparison function C. C compares a pair of elements and returns
-- an exact integer whose sign indicates their relationship:
-- (c x y) < 0 => x<y
-- (c x y) = 0 => x=y
-- (c x y) > 0 => x>y
-- To help remember the relationship between the sign of the result and
-- the relation, use the function - as the model for C: (- x y) < 0
-- means that x < y; (- x y) > 0 means that x > y.
--
-- The extra discrimination provided by the three-way comparison can
-- provide significant speedups when sorting data sets with many duplicates,
-- especially when the comparison function is relatively expensive (e.g.,
-- comparing long strings).
--
-- WARNING: Some sort algorithms, such as insertion sort or heap sort,
-- can tolerate being passed a <= comparison function when they expect a <
-- function -- insertion and merge sort may simply invert stability; and
-- heap sort will run a bit slower, but otherwise produce a correct answer.
--
-- Quicksort, however, is much more critically sensitive to the distinction
-- between a < and a <= comparison. If QUICK-SORT or QUICK-SORT! expect a <
-- comparison function, and are erroneously given a <= function, they may,
-- depending on implementation, produce an unsorted result, go into an
-- infinite loop, cause a run-time error, occasionally produce a correct
-- result, or do some fifth thing.
--
-- Implementors may wish to write QUICKSORT3! so that it (a) tests the
-- comparison function (by checking that (c v[start] v[start]) produces
-- false), or (b) is tolerant of an erroneous <= function, or (c) both.
-- Clients of this function, however, should not count on this.
--
--vector-insert-sort-lib - vector insertion sort
-- insert-sort < v [start end] -> vector
-- insert-sort! < v [start end] -> unspecified
--
-- These procedures stably sort their data using insertion sort.
--
-- INSERT-SORT returns a vector of length END-START.
-- INSERT-SORT! is in-place, leaving its result in V[start,end).
--
--delndup-lib - list and vector delete neighbor duplicates
-- list-delete-neighbor-dups = lis -> list
-- list-delete-neighbor-dups! = lis -> list
--
-- vector-delete-neighbor-dups = v [start end] -> vector
-- vector-delete-neighbor-dups! = v [start end] -> end'
--
-- These procedures delete adjacent duplicate elements from a list or
-- a vector, using a given element-equality procedure =. The first/leftmost
-- element of a run of equal elements is the one that survives. The list
-- or vector is not otherwise disordered.
--
-- These procedures are linear time -- much faster than the O(n^2) general
-- duplicate-element deletors that do not assume any "bunching" of elements
-- (such as the ones provided by SRFI-1). If you want to delete duplicate
-- elements from a large list or vector, you can sort the elements to bring
-- equal items together, then use one of these procedures, for a total time
-- of O(n lg n).
--
-- The comparison function = passed to these procedures is always applied
-- (= x y)
-- where X comes before Y in the containing list or vector.
--
-- LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
-- may share storage with the input list.
--
-- VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
-- rather allocates a fresh vector to hold the result.
--
-- LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
-- mutate its input list in order to construct its answer.
--
-- VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
-- answer, packing its answer into the index range [start,end'), where
-- END' is the non-negative exact integer returned as its value. It
-- returns END' as its result. The vector is not altered outside the range
-- [start,end').
--
-- Examples:
-- (list-delete-neighbor-dups = '(1 1 2 7 7 7 0 -2 -2))
-- => (1 2 7 0 -2)
--
-- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2))
-- => #(1 2 7 0 -2)
--
-- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2) 3 7)
-- => #(7 0 -2)
--
-- ;; Result left in v[3,9):
-- (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
-- (cons (vector-delete-neighbor-dups! = v 3)
-- v))
-- => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
--
--binsearch-lib - vector binary search lib
-- vector-binary-search elt< elt->key key v [start end] -> integer-or-false
-- vector-binary-search3 c v [start end] -> integer-or-false
--
-- VECTOR-BINARY-SEARCH searches vector V in range [START,END) (which
-- default to 0 and the length of V, respectively) for an element whose
-- associated key is equal to KEY. The procedure ELT->KEY is used to map
-- an element to its associated key. The elements of the vector are assumed
-- to be ordered by the ELT< relation on these keys. That is,
-- (vector-sorted? (lambda (x y) (elt< (elt->key x) (elt->key y)))
-- v start end) => true
-- An element E of V is a match for KEY if it's neither less nor greater
-- than the key:
-- (and (not (elt< (elt->key e) key))
-- (not (elt< key (elt->key e))))
-- If there is such an element, the procedure returns its index in the
-- vector as an exact integer. If there is no such element in the searched
-- range, the procedure returns false.
--
-- (vector-binary-search < car 4 '#((1 . one) (3 . three)
-- (4 . four) (25 . twenty-five)))
-- => 2
--
-- (vector-binary-search < car 7 '#((1 . one) (3 . three)
-- (4 . four) (25 . twenty-five)))
-- => #f
--
-- VECTOR-BINARY-SEARCH3 is a variant that uses a three-way comparison
-- function C. C compares its parameter to the search key, and returns an
-- exact integer whose sign indicates its relationship to the search key.
-- (c x) < 0 => x < search-key
-- (c x) = 0 => x = search-key
-- (c x) > 0 => x > search-key
--
-- (vector-binary-search3 (lambda (elt) (- (car elt) 4))
-- '#((1 . one) (3 . three)
-- (4 . four) (25 . twenty-five)))
-- => 2
--
-- Rationale:
-- - Why isn't VECTOR-BINARY-SEARCH's ELT->KEY computation simply absorbed
-- into the < function? It is separated out because the < function is
-- applied twice inside the binary-search inner loop, once with the search
-- key for the first argument and the element key for the second argument,
-- and once, with the reverse argument order. This is not necessary for
-- VECTOR-BINARY-SEARCH3.
--
-- - When a comparison operation is able to produce a three-way
-- discrimination, the inner loop of the binary search can trim the number
-- of per-iteration comparisons from an average of 1.5 to a guaranteed
-- single comparison per iteration. This can be a significant savings when
-- searching with an expensive comparison operation (e.g., one that
-- uses string compare, sends email, references a database, or queries
-- a network service such as a web server).
--
-- - Failure is signaled by false (rather than, say, -1) so that searches
-- can be used in conditional forms such as
-- (or (vector-binary-search ...) ...)
-- or
-- (cond ((vector-binary-search ...) => index-consumer)
-- ...)
--
---------------------------------------------------------------------------------
--* Algorithmic properties
--------------------------
--Different sort and merge algorithms have different properties.
--Choose the algorithm that matches your needs:
--
--Vector insert sort
-- Stable, but only suitable for small vectors -- O(n^2).
--
--Vector quick sort
-- Not stable. Is fast on average -- O(n lg n) -- but has bad worst-case
-- behaviour. Has good memory locality for big vectors (unlike heap sort).
-- A clever pivot-picking trick (median of three samples) helps avoid
-- worst-case behaviour, but pathological cases can still blow up.
--
--Vector heap sort
-- Not stable. Guaranteed fast -- O(n lg n) *worst* case. Poor locality
-- on large vectors. A very reliable workhorse.
--
--Vector merge sort
-- Stable. Not in-place -- requires a temporary buffer of equal size.
-- Fast -- O(n lg n) -- and has good memory locality for large vectors.
--
-- The implementation of vector merge sort provided by this SRFI's reference
-- implementation is, additionally, a "natural" sort, meaning that it
-- exploits existing order in the input data, providing O(n) best case.
--
--Destructive list merge sort
-- Stable, fast and in-place (i.e., allocates no new cons cells). "Fast"
-- means O(n lg n) worse-case, and substantially better if the data
-- is already mostly ordered, all the way down to linear time for
-- a completely-ordered input list (i.e., it is a "natural" sort).
--
-- Note that sorting lists involves chasing pointers through memory, which
-- can be a loser on modern machine architectures because of poor cache &
-- page locality. Pointer *writing*, which is what the SET-CDR!s of a
-- destructive list-sort algorithm do, is even worse, especially if your
-- Scheme has a generational GC -- the writes will thrash the write-barrier.
-- Sorting vectors has inherently better locality.
--
-- This SRFI's destructive list merge and merge sort implementations are
-- opportunistic -- they avoid redundant SET-CDR!s, and try to take long
-- already-ordered runs of list structure as-is when doing the merges.
--
--Pure list merge sort
-- Stable and fast -- O(n lg n) worst-case, and possibly O(n), depending
-- upon the input list (see discussion above).
--
--
--Algorithm Stable? Worst case Average case In-place
--------------------------------------------------------
--Vector insert Yes O(n^2) O(n^2) Yes
--Vector quick No O(n^2) O(n lg n) Yes
--Vector heap No O(n lg n) O(n lg n) Yes
--Vector merge Yes O(n lg n) O(n lg n) No
--List merge Yes O(n lg n) O(n lg n) Either
--
--
---------------------------------------------------------------------------------
--* Porting and optimisation
----------------------------
--This package should be trivial to port. There are only four non-R4RS bits
--in the code:
--- Use of multiple-value return, with the R5RS VALUES procedure, and the
-- simple (RECEIVE (var ...) mv-exp body ...) multiple-value binding macro
-- of SRFI-8.
--
--- A VECTOR-COPY procedure. This is a tiny little procedure:
-- (vector-copy v [start end])
--
--- Use of the LET-OPTIONALS macro from scsh to parse and default optional
-- arguments to three routines. Again, easy to port the macro or rewrite
-- the code to parse, default, and error check the args by hand.
--
--- Calls to an ERROR function for complaining about bad arguments.
--
--This code is tightly bummed, as far as I can go in portable Scheme.
--
--You could speed up the vector code a lot by error-checking the procedure
--parameters and then shifting over to fixnum-specific arithmetic and dangerous
--vector-indexing and vector-setting primitives. The comments in the code
--indicate where the initial error checks would have to be added. There are
--several (QUOTIENT N 2)'s that could be changed to a fixnum right-shift, as
--well, in both the list and vector code (SRFI 33 provides such an operator).
--The code is designed to enable this -- each file usually exports one or two
--"safe" procedures that end up calling an internal "dangerous" primitive. The
--little exported cover procedures are where you move the error checks.
--
--This should provide *big* speedups. In fact, all the code bumming I've done
--pretty much disappears in the noise unless you have a good compiler and also
--can dump the vector-index checks and generic arithmetic -- so I've really just
--set things up for you to exploit.
--
--The optional-arg parsing, defaulting, and error checking is done with a
--portable R4RS macro. But if your Scheme has a faster mechanism (e.g., Chez),
--you should definitely port over to it. Note that argument defaulting and
--error-checking are interleaved -- you don't have to error-check defaulted
--START/END args to see if they are fixnums that are legal vector indices for
--the corresponding vector, etc.
--
--
---------------------------------------------------------------------------------
--* References & Links
----------------------
--
--This document, in HTML:
-- http://srfi.schemers.org/srfi-32/srfi-32.html
-- [This link may not be valid while the SRFI is in draft form.]
--
--This document, in simple text format:
-- http://srfi.schemers.org/srfi-32/srfi-32.txt
--
--Archive of SRFI-32 discussion-list email:
-- http://srfi.schemers.org/srfi-32/mail-archive/maillist.html
--
--SRFI web site:
-- http://srfi.schemers.org/
--
--[CommonLisp]
-- Common Lisp: the Language
-- Guy L. Steele Jr. (editor).
-- Digital Press, Maynard, Mass., second edition 1990.
-- Available at http://www.elwood.com/alu/table/references.htm#cltl2
--
-- The Common Lisp "HyperSpec," produced by Kent Pitman, is essentially
-- the ANSI spec for Common Lisp:
-- http://www.xanalys.com/software_tools/reference/HyperSpec/
--
--[R5RS]
-- Revised^5 Report on the Algorithmic Language Scheme,
-- R. Kelsey, W. Clinger, J. Rees (editors).
-- Higher-Order and Symbolic Computation, Vol. 11, No. 1, September, 1998.
-- and ACM SIGPLAN Notices, Vol. 33, No. 9, October, 1998.
--
-- Available at http://www.schemers.org/Documents/Standards/
--
--
---------------------------------------------------------------------------------
--* Acknowledgements
--------------------
--
--I thank the authors of the open source I consulted when designing this
--library, particularly Richard O'Keefe, Donovan Kolby and the MIT Scheme Team.
--
--
---------------------------------------------------------------------------------
--* Copyright
-------------
--
--** SRFI text
--============
--This document is copyright (C) Olin Shivers (1998, 1999).
--All Rights Reserved.
--
--This document and translations of it may be copied and furnished to others,
--and derivative works that comment on or otherwise explain it or assist in its
--implementation may be prepared, copied, published and distributed, in whole or
--in part, without restriction of any kind, provided that the above copyright
--notice and this paragraph are included on all such copies and derivative
--works. However, this document itself may not be modified in any way, such as
--by removing the copyright notice or references to the Scheme Request For
--Implementation process or editors, except as needed for the purpose of
--developing SRFIs in which case the procedures for copyrights defined in the
--SRFI process must be followed, or as required to translate it into languages
--other than English.
--
--The limited permissions granted above are perpetual and will not be revoked by
--the authors or their successors or assigns.
--
--This document and the information contained herein is provided on an "AS IS"
--basis and THE AUTHORS AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR
--IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
--INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
--MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
--
--** Reference implementation
--===========================
--Short summary: no restrictions.
--
--While I wrote all of this code myself, I read a lot of code before I began
--writing. However, all such code is, itself, either open source or public
--domain, rendering irrelevant any issue of "copyright taint."
--
--The natural merge sorts (pure list, destructive list, and vector) are not only
--my own code, but are implementations of an algorithm of my own devising. They
--run in O(n lg n) worst case, O(n) best case, and require only a logarithmic
--number of stack frames. And they are stable. And the destructive-list variant
--allocates zero cons cells; it simply rearranges the cells of the input list.
--
--Hence the reference implementation is
-- Copyright (c) 1998 by Olin Shivers.
--and made available under the same copyright as the SRFI text (see above).
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/srfi-32.txt racket-6.12/share/pkgs/srfi-lib/srfi/32/srfi-32.txt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/srfi-32.txt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/32/srfi-32.txt 2018-01-31 12:22:58.152501970 -0500
-@@ -0,0 +1,938 @@
-+The SRFI-32 sort libraries -*- outline -*-
-+Olin Shivers
-+First draft: 1998/10/19
-+Last update: 2002/7/21
-+
-+Emacs should display this document in outline mode. Say c-h m for
-+instructions on how to move through it by sections (e.g., c-c c-n, c-c c-p).
-+
-+* Table of contents
-+-------------------
-+Abstract
-+Procedure index
-+Introduction
-+What's wrong with the current state of affairs?
-+Design rules
-+ What vs. how
-+ Consistency across function signatures
-+ Data parameter first, less-than parameter after
-+ Ordering, comparison functions & stability
-+ All vector operations accept optional subrange parameters
-+ Required vs. allowed side-effects
-+Procedure specification
-+ Procedure naming and functionality
-+ Types of parameters and return values
-+ sort-lib - general sorting package
-+ Algorithm-specific sorting packages
-+Algorithmic properties
-+Topics to be resolved during discussion phase
-+Porting and optimisation
-+References & Links
-+Acknowledgements
-+Copyright
-+
-+
-+* Abstract
-+----------
-+Current Scheme sorting packages are, every one of them, surprisingly bad. I've
-+designed the API for a full-featured sort toolkit, which I propose as an SRFI.
-+The spec comes with 1200 lines of high-quality reference code: tightly
-+written, highly commented, portable code, available for free. Implementors
-+want this code. It's better than what you have.
-+
-+-------------------------------------------------------------------------------
-+* Procedure index
-+-----------------
-+list-sorted? vector-sorted?
-+
-+list-merge vector-merge
-+list-sort vector-sort
-+list-stable-sort vector-stable-sort
-+list-delete-neighbor-dups vector-delete-neighbor-dups
-+
-+list-merge! vector-merge!
-+list-sort! vector-sort!
-+list-stable-sort! vector-stable-sort!
-+list-delete-neighbor-dups! vector-delete-neighbor-dups!
-+
-+heap-sort quick-sort insert-sort list-merge-sort vector-merge-sort
-+heap-sort! quick-sort! insert-sort! list-merge-sort! vector-merge-sort!
-+
-+-------------------------------------------------------------------------------
-+* Introduction
-+--------------
-+As I'll detail bewlow, I wasn't very happy with the state of the Scheme
-+world for sorting and merging lists and vectors. So I have designed and
-+written a fairly comprehensive sorting & merging toolkit. It is
-+
-+ - very portable,
-+
-+ - much better code than what is currently in Elk, Gambit, Bigloo,
-+ Scheme->C, MzScheme, RScheme, Scheme48, MIT Scheme, or slib, and
-+
-+ - priced to move: free code.
-+
-+The package includes
-+ - Vector insert sort (stable)
-+ - Vector heap sort
-+ - Vector quick sort (with median-of-3 pivot picking)
-+ - Vector merge sort (stable)
-+ - Pure and destructive list merge sort (stable)
-+ - Stable vector and list merge
-+ - Miscellaneous sort-related procedures: Vector and list merging,
-+ sorted? predicates, vector binary search, vector and list
-+ delete-equal-neighbor procedures.
-+ - A general, non-algorithmic set of procedure names for general sorting
-+ and merging.
-+
-+Scheme programmers may want to adopt this package. I'd like Scheme
-+implementors to adopt this code and its API -- in fact, the code is a bribe to
-+make it easy for implementors to converge on the suggested API. I mean, you'd
-+really have to be a boor to take this free code I wrote and mutate its
-+interface over to your incompatible, unportable API, wouldn't you? But you
-+could, of course -- it's freely available. More in the spirit of the offering,
-+you could make this API available, and then also write a little module
-+providing your old interface that is defined in terms of this API. "Scheme
-+implementors," in this context, includes slib, which isn't really a standalone
-+implementation of Scheme, but is an influential collection of API's and code.
-+
-+The code is tightly bummed. It is clearly written, and commented in my usual
-+voluminous style. This includes notes on porting and implementation-specific
-+optimisations.
-+
-+
-+-------------------------------------------------------------------------------
-+* What's wrong with the current state of affairs?
-+-------------------------------------------------
-+
-+It's just amazing to me that in 2002, sorting and merging hasn't been
-+completely put to bed. These are well-understood algorithms, each of them well
-+under a page of code. The straightforward algorithms are basic, core stuff --
-+sophomore-level. But if you tour the major Scheme implementations out there on
-+the Net, you find badly written code that provides extremely spotty coverage
-+of the algorithm space. One implementation even has a buggy implementation
-+that has been in use for about 20 years!
-+
-+Open source-code is a wonderful thing. In a couple of hours, I was able to
-+download and check the sources of 9 Scheme systems. Here are my notes from the
-+systems I checked. You can skip to the next section if you aren't morbidly
-+curious.
-+
-+slib
-+ sorted? vector-or-list <
-+ merge list1 list2 <
-+ merge! list1 list2 <
-+ sort vector-or-list <
-+ sort! vector-or-list <
-+
-+ Richard O'Keefe's stable list merge sort is right idea, but implemented
-+ using gratuitous variable side effects. It also does redundant SET-CDR!s.
-+ The vector sort converts to list, merge sorts, then reconverts
-+ to vector. This is a bad idea -- non-local pointer chasing bad; vector
-+ shuffling good.
-+
-+MIT Scheme
-+ sort! vector <
-+ merge-sort! vector <
-+ quick-sort! vector <
-+
-+ sort vector-or-list <
-+ merge-sort vector-or-list <
-+ quick-sort vector-or-list <
-+
-+ Naive vector quicksort: loser, for worst-case performance reasons.
-+ List sort by "list->vector; quicksort; vector->list," hence also loser.
-+ A clever stable vector merge sort, albeit not very bummed.
-+
-+Scheme 48 & T
-+ sort-list list <
-+ sort-list! list <
-+ list-merge! list1 list2 <
-+
-+ Bob Nix's implementation of online merge-sort, written in the early 80's.
-+ Conses unnecessary bookkeeping structure, which isn't necessary with a
-+ proper recursive formulation. Also, does redundant SET-CDR!s. No vector
-+ sort. Also, has a bug -- is claimed to be a stable sort, but isn't! To see
-+ this, get the S48 code, and try
-+ (define (my< x y) (< (quotient x 2) (quotient y 2)))
-+ (list-merge! (list 0 2) (list 3) my<) ; -> (0 2 3)
-+ (list-merge! (list 2) (list 0 3) my<) ; -> (0 3 2)
-+ This could be fixed very easily, but it isn't worth it given the
-+ other problems with the algorithm.
-+
-+RScheme
-+ vector-sort! vector <
-+ sort collection <
-+
-+ Good basic implementation of vector heapsort, which has O(n lg n)
-+ worst-case time. Code ugly, needs tuning. List sort by "list->vector;
-+ sort; vector->list", which allocates unneeded temp storage. Nothing
-+ for stable sorting.
-+
-+MzScheme
-+ Naive quicksort -- but not available for vector sorting, even
-+ though it internally uses a vector. Nothing for stable sorting,
-+ and naive quicksort has bad worst-case behaviour.
-+
-+Bigloo, Scheme->C
-+ Couldn't find anything -- but maybe I didn't search for the right
-+ thing, since the Bigloo names are French. (I invite correction from
-+ the Bigloo implementors.)
-+
-+Gambit
-+ sort-list list <
-+ Nothing for vectors. Simple, slow, unstable merge sort for lists.
-+
-+Elk
-+ Another naive quicksort. Lists handled by converting to vector.
-+ sort vector-or-list <
-+ sort! vector-or-list <
-+
-+Chez Scheme
-+ merge < list1 list2
-+ merge! < list1 list2
-+ sort < list
-+ sort! < list
-+
-+ These are stable. I have not seen the source code.
-+
-+Common Lisp
-+ sort sequence < [key]
-+ stable-sort sequence < [key]
-+ merge result-type sequence1 sequence2 < [key]
-+
-+ The sort procedures are allowed, but not required, to be destructive.
-+
-+SML/NJ
-+ sort: ('a*'a -> bool) -> 'a list -> 'a list
-+ "Smooth applicative merge sort," which is stable.
-+ There is also a highly bummed quicksort for vectors.
-+
-+The right solution: Implement a full toolbox of carefully written standard sort
-+routines.
-+
-+Having the source available for all of these above-cited Schemes made
-+life a lot easier writing this code. I appreciate the authors making their
-+source available under such open terms.
-+
-+
-+-------------------------------------------------------------------------------
-+* Design rules
-+--------------
-+
-+** What vs. how
-+===============
-+There are two different interfaces: "what" (simple) & "how" (detailed).
-+
-+ - Simple: you specify semantics: datatype (list or vector),
-+ mutability, and stability.
-+
-+ - Detailed: you specify the actual algorithm (quick, heap,
-+ insert, merge). Different algorithms have different properties,
-+ both semantic & pragmatic, so these exports are necessary.
-+
-+ It is necessarily the case that the specifications of these procedures
-+ make statements about execution "pragmatics." For example, the sole
-+ distinction between heap sort and quick sort -- both of which are
-+ provided by this library -- is one of execution time, which is not a
-+ "semantic" distinction. Similar resource-use statements are made about
-+ "iterative" procedures, meaning that they can execute on input of
-+ arbitrary size without needing to allocate an unbounded number of stack
-+ frames.
-+
-+** Consistency across function signatures
-+=========================================
-+The two interfaces share common function signatures wherever
-+possible, to facilitate switching a given call from one procedure
-+to another.
-+
-+** Data parameter first, less-than parameter after
-+==================================================
-+These procedures uniformly observe the following parameter order:
-+the data to be sorted come before the the comparison function.
-+That is, we write
-+ (sort lis <)
-+not
-+ (sort < lis).
-+This is consistent with every single implementation out there, with
-+the sole exception of Chez Scheme.
-+
-+In my opinion, it would be more consistent with other Scheme libraries
-+to put the ordering function first -- the "operation currying" convention.
-+(E.g., consider FOR-EACH or MAP or FIND.) I decided to leave things as they
-+are in favor of near-total backwards compatibility with existing practice.
-+
-+[Perhaps this should be discussed.]
-+
-+** Ordering, comparison functions & stability
-+=============================================
-+These routines take a < comparison function, not a <= comparison
-+function, and they sort into increasing order. The difference between
-+a < spec and a <= spec comes up in two places:
-+ - the definition of an ordered or sorted data set, and
-+ - the definition of a stable sorting algorithm.
-+
-++ We say that a data set (a list or vector) is *sorted* or *ordered*
-+ if it contains no adjacent pair of values ... X Y ... such that Y < X.
-+
-+ In other words, scanning across the data never takes a "downwards" step.
-+
-+ If you use a <= procedure where these algorithms expect a <
-+ procedure, you may not get the answers you expect. For example,
-+ the LIST-SORTED? function will return false if you pass it a <= comparison
-+ function and an ordered list containing adjacent equal elements.
-+
-++ A "stable" sort is one that preserves the pre-existing order of equal
-+ elements. Suppose, for example, that we sort a list of numbers by
-+ comparing their absolute values, i.e., using comparison function
-+ (lambda (x y) (< (abs x) (abs y)))
-+ If we sort a list that contains both 3 and -3:
-+ ... 3 ... -3 ...
-+ then a stable sort is an algorithm that will not swap the order
-+ of these two elements, that is, the answer will look like
-+ ... 3 -3 ...
-+ not
-+ ... -3 3 ...
-+
-+ Choosing < for the comparison function instead of <= affects how stability
-+ is coded. Given an adjacent pair X Y, (< y x) means "Y should be moved in
-+ front of X" -- otherwise, leave things as they are. So using a <= function
-+ where a < function is expected will *invert* stability.
-+
-+ This is due to the definition of equality, given a < comparator:
-+ (and (not (< x y))
-+ (not (< y x)))
-+ The definition is rather different, given a <= comparator:
-+ (and (<= x y)
-+ (<= x y))
-+
-++ A "stable" merge is one that reliably favors one of its data sets
-+ when equal items appear in both data sets. *All merge operations in
-+ this library are stable*, breaking ties between data sets in favor
-+ of the first data set -- elements of the first list come before equal
-+ elements in the second list.
-+
-+ So, if we are merging two lists of numbers ordered by absolute value
-+ using the stable merge operation LIST-MERGE
-+ (list-merge '(0 -2 4 8 -10) '(-1 3 -4 7)
-+ (lambda (x y) (< (abs x) (abs y))))
-+ reliably places the 4 of the first list before the equal-comparing -4
-+ of the second list:
-+ (0 -1 -2 4 -4 7 8 -10)
-+
-+In short, if your comparison function F answers true to (F x x), then
-+using a stable sorting or merging algorithm will not give you a stable sort
-+or merge, and LIST-SORTED? may surprise you. Note that you can synthesize a <
-+function from a <= function with
-+ (lambda (x y) (not (<= y x)))
-+if need be.
-+
-+Precise definitions give sharp edges to tools, but require care
-+in use. "Measure twice, cut once."
-+
-+I have adopted the choice of < from Common Lisp. I assume they
-+had a good reason for adopting < instead of <=. I'd love to know
-+what this reason is; send me email if you can explain it, please.
-+
-+** All vector operations accept optional subrange parameters
-+============================================================
-+The vector operations specified below all take optional START/END arguments
-+indicating a selected subrange of a vector's elements. If a START parameter or
-+START/END parameter pair is given to such a procedure, they must be exact,
-+non-negative integers, such that
-+ 0 <= START <= END <= (VECTOR-LENGTH V)
-+where V is the related vector parameter. If not specified, they default to 0
-+and the length of the vector, respectively. They are interpreted to select the
-+range [START,END), that is, all elements from index START (inclusive) up to,
-+but not including, index END.
-+
-+** Required vs. allowed side-effects
-+====================================
-+LIST-SORT! and LIST-STABLE-SORT! are allowed, but not required,
-+to alter their arguments' cons cells to construct the result list. This is
-+consistent with the what-not-how character of the group of procedures
-+to which they belong (the "sort-lib" package).
-+
-+The LIST-DELETE-NEIGHBOR-DUPS!, LIST-MERGE! and LIST-MERGE-SORT! procedures,
-+on the other hand, provide specific algorithms, and, as such, explicitly
-+commit to the use of side-effects on their input lists in order to guarantee
-+their key algorithmic properties (e.g., linear-time operation, constant-space
-+stack use).
-+
-+-------------------------------------------------------------------------------
-+* Procedure specification
-+-------------------------
-+The procedures are split into several packages. In a Scheme system that has a
-+module or package system, these procedures should be contained in modules
-+named as follows:
-+ Package name Functionality
-+ ------------ -------------
-+ sort-lib General sorting for lists & vectors
-+ sorted?-lib Sorted predicates for lists & vectors
-+ list-merge-sort-lib List merge sort
-+ vector-merge-sort-lib Vector merge sort
-+ vector-heap-sort-lib Vector heap sort
-+ vector-quick-sort-lib Vector quick sort
-+ vector-insert-sort-lib Vector insertion sort
-+ delndup-lib List and vector delete neighbor duplicates
-+
-+A Scheme system without a module system should provide all of the bindings
-+defined in all of these modules as components of the "SRFI-32" package.
-+
-+Note that there is no list insert sort package, as you might as well always
-+use list merge sort. The reference implementation's destructive list merge
-+sort will do fewer SET-CDR!s than a destructive insert sort.
-+
-+** Procedure naming and functionality
-+=====================================
-+Almost all of the procedures described below are variants of two basic
-+operations: sorting and merging. These procedures are consistently named
-+by composing a set of basic lexemes to indicate what they do.
-+
-+Lexeme Meaning
-+------ -------
-+"sort" The procedure sorts its input data set by some < comparison function.
-+
-+"merge" The procedure merges two ordered data sets into a single ordered
-+ result.
-+
-+"stable" This lexeme indicates that the sort is a stable one.
-+
-+"vector" The procedure operates upon vectors.
-+
-+"list" The procedure operates upon lists.
-+
-+"!" Procedures that end in "!" are allowed, and sometimes required,
-+ to reuse their input storage to construct their answer.
-+
-+** Types of parameters and return values
-+========================================
-+In the procedures specified below,
-+ - A LIS parameter is a list;
-+
-+ - A V parameter is a vector;
-+
-+ - A < or = parameter is a procedure accepting two arguments taken from the
-+ specified procedure's data set(s), and returning a boolean;
-+
-+ - START and END parameters are exact, non-negative integers that
-+ serve as vector indices selecting a subrange of some associated vector.
-+ When specified, they must satisfy the relation
-+ 0 <= start <= end <= (vector-length v)
-+ where V is the associated vector.
-+
-+Passing values to procedures with these parameters that do not satisfy these
-+types is an error.
-+
-+If a procedure is said to return "unspecified," this means that nothing at all
-+is said about what the procedure returns, not even the number of return
-+values. Such a procedure is not even required to be consistent from call to
-+call in the nature or number of its return values. It is simply required to
-+return a value (or values) that may be passed to a command continuation, e.g.
-+as the value of an expression appearing as a non-terminal subform of a BEGIN
-+expression. Note that in R5RS, this restricts such a procedure to returning a
-+single value; non-R5RS systems may not even provide this restriction.
-+
-+** sort-lib - general sorting package
-+=====================================
-+This library provides basic sorting and merging functionality suitable for
-+general programming. The procedures are named by their semantic properties,
-+i.e., what they do to the data (sort, stable sort, merge, and so forth).
-+
-+ Procedure Suggested algorithm
-+ -------------------------------------------------------------------------
-+ list-sorted? lis < -> boolean
-+ list-merge lis1 lis2 < -> list
-+ list-merge! lis1 lis2 < -> list
-+ list-sort lis < -> list (vector heap or quick)
-+ list-sort! lis < -> list (list merge sort)
-+ list-stable-sort lis < -> list (vector merge sort)
-+ list-stable-sort! lis < -> list (list merge sort)
-+ list-delete-neighbor-dups lis = -> list
-+ list-delete-neighbor-dups! lis = -> list
-+
-+ vector-sorted? v < [start end] -> boolean
-+ vector-merge v1 v2 < [start1 end1 start2 end2] -> vector
-+ vector-merge! v v1 v2 < [start start1 end1 start2 end2] -> unspecific
-+ vector-sort v < [start end] -> vector (heap or quick sort)
-+ vector-sort! v < [start end] -> unspecific (heap or quick sort)
-+ vector-stable-sort v < [start end] -> vector (vector merge sort)
-+ vector-stable-sort! v < [start end] -> unspecific (vector merge sort)
-+ vector-delete-neighbor-dups v = [start end] -> vector
-+ vector-delete-neighbor-dups! v = [start end] -> end'
-+
-+ LIST-SORTED? and VECTOR-SORTED? return true if their input list or vector
-+ is in sorted order, as determined by their < comparison parameter.
-+
-+ All four merge operations are stable: an element of the initial list LIS1
-+ or vector V1 will come before an equal-comparing element in the second
-+ list LIS2 or vector V2 in the result.
-+
-+ The procedures
-+ LIST-MERGE
-+ LIST-SORT
-+ LIST-STABLE-SORT
-+ LIST-DELETE-NEIGHBOR-DUPS
-+ do not alter their inputs and are allowed to return a value that shares
-+ a common tail with a list argument.
-+
-+ The procedures
-+ LIST-SORT!
-+ LIST-STABLE-SORT!
-+ are "linear update" operators -- they are allowed, but not required, to
-+ alter the cons cells of their arguments to produce their results.
-+
-+ On the other hand, the procedures
-+ LIST-DELETE-NEIGHBOR-DUPS!
-+ LIST-MERGE!
-+ make only a single, iterative, linear-time pass over their argument lists,
-+ using SET-CDR!s to rearrange the cells of the lists into the final result
-+ -- they work "in place." Hence, any cons cell appearing in the result must
-+ have originally appeared in an input. The intent of this
-+ iterative-algorithm commitment is to allow the programmer to be sure that
-+ if, for example, LIST-MERGE! is asked to merge two ten-million-element
-+ lists, the operation will complete without performing some extremely
-+ (possibly twenty-million) deep recursion.
-+
-+ The vector procedures
-+ VECTOR-SORT
-+ VECTOR-STABLE-SORT
-+ VECTOR-DELETE-NEIGHBOR-DUPS
-+ do not alter their inputs, but allocate a fresh vector for their result,
-+ of length END-START.
-+
-+ The vector procedures
-+ VECTOR-SORT!
-+ VECTOR-STABLE-SORT!
-+ sort their data in-place. (But note that VECTOR-STABLE-SORT! may
-+ allocate temporary storage proportional to the size of the input --
-+ I am not aware of O(n lg n) stable vector sorting algorithms that
-+ run in constant space.)
-+
-+ VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
-+
-+ VECTOR-MERGE! writes its result into vector V, beginning at index START0,
-+ for indices less than END0 = START0 + (END1-START1) + (END2-START2). The
-+ target subvector
-+ V[start0,end0)
-+ may not overlap either source subvector
-+ V1[start1,end1)
-+ V2[start2,end2).
-+
-+ The DELETE-NEIGHBOR-DUP-... procedures:
-+ These procedures delete adjacent duplicate elements from a list or a
-+ vector, using a given element-equality procedure. The first/leftmost
-+ element of a run of equal elements is the one that survives. The list or
-+ vector is not otherwise disordered.
-+
-+ These procedures are linear time -- much faster than the O(n^2) general
-+ duplicate-element deletors that do not assume any "bunching" of elements
-+ (such as the ones provided by SRFI-1). If you want to delete duplicate
-+ elements from a large list or vector, sort the elements to bring equal
-+ items together, then use one of these procedures, for a total time of
-+ O(n lg n).
-+
-+ The comparison function = passed to these procedures is always applied
-+ (= x y)
-+ where X comes before Y in the containing list or vector.
-+
-+ - LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
-+ may share storage with the input list.
-+
-+ - VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
-+ rather allocates a fresh vector to hold the result.
-+
-+ - LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
-+ mutate its input list in order to construct its answer.
-+
-+ - VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
-+ answer, packing its answer into the index range [start,end'), where
-+ END' is the non-negative exact integer returned as its value. It
-+ returns END' as its result. The vector is not altered outside the range
-+ [start,end').
-+
-+ [Maybe this procedure should take a "target" vector to write?]
-+
-+ - Examples:
-+ (list-delete-neighbor-dups '(1 1 2 7 7 7 0 -2 -2) =)
-+ => (1 2 7 0 -2)
-+
-+ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) =)
-+ => #(1 2 7 0 -2)
-+
-+ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) = 3 7)
-+ => #(7 0 -2)
-+
-+ ;; Result left in v[3,9):
-+ (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
-+ (cons (vector-delete-neighbor-dups! v = 3)
-+ v))
-+ => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
-+
-+
-+** Algorithm-specific sorting packages
-+======================================
-+These packages provide more specific sorting functionality, that is,
-+specific committment to particular algorithms that have particular
-+pragmatic consequences (such as memory locality, asymptotic running time)
-+beyond their semantic behaviour (sorting, stable sorting, merging, etc.).
-+Programmers that need a particular algorithm can use one of these packages.
-+
-+sorted?-lib - sorted predicates
-+ list-sorted? lis < -> boolean
-+ vector-sorted? v < [start end] -> boolean
-+
-+ Return #f iff there is an adjacent pair ... X Y ... in the input
-+ list or vector such that Y < X. The optional START/END range
-+ arguments restrict VECTOR-SORTED? to the indicated subvector.
-+
-+list-merge-sort-lib - list merge sort
-+ list-merge-sort lis < -> list
-+ list-merge-sort! lis < -> list
-+ list-merge lis1 lis2 < -> list
-+ list-merge! lis1 lis2 < -> list
-+
-+ The sort procedures sort their data using a list merge sort, which is
-+ stable. (The reference implementation is, additionally, a "natural" sort.
-+ See below for the properties of this algorithm.)
-+
-+ The ! procedures are destructive -- they use SET-CDR!s to rearrange the
-+ cells of the lists into the proper order. As such, they do not allocate
-+ any extra cons cells -- they are "in place" sorts. Additionally,
-+ LIST-MERGE! is iterative, not recursive -- it can operate on arguments of
-+ arbitrary size without requiring an unbounded amount of stack space.
-+
-+ The merge operations are stable: an element of LIS1 will come before an
-+ equal-comparing element in LIS2 in the result list.
-+
-+vector-merge-sort-lib - vector merge sort
-+ vector-merge-sort v < [start end temp] -> vector
-+ vector-merge-sort! v < [start end temp] -> unspecific
-+ vector-merge v1 v2 < [start1 end1 start2 end2] -> vector
-+ vector-merge! v v1 v2 < [start0 start1 end1 start2 end2] -> unspecific
-+
-+ The sort procedures sort their data using vector merge sort, which is
-+ stable. (The reference implementation is, additionally, a "natural" sort.
-+ See below for the properties of this algorithm.)
-+
-+ The optional START/END arguments provide for sorting of subranges, and
-+ default to 0 and the length of the corresponding vector.
-+
-+ Merge-sorting a vector requires the allocation of a temporary "scratch"
-+ work vector for the duration of the sort. This scratch vector can be
-+ passed in by the client as the optional TEMP argument; if so, the supplied
-+ vector must be of size >= END, and will not be altered outside the range
-+ [start,end). If not supplied, the sort routines allocate one themselves.
-+
-+ The merge operations are stable: an element of V1 will come before an
-+ equal-comparing element in V2 in the result vector.
-+
-+ VECTOR-MERGE-SORT! leaves its result in V[start,end).
-+
-+ VECTOR-MERGE-SORT returns a vector of length END-START.
-+
-+ VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
-+
-+ VECTOR-MERGE! writes its result into vector V, beginning at index START0,
-+ for indices less than END0 = START0 + (END1-START1) + (END2-START2). The
-+ target subvector
-+ V[start0,end0)
-+ may not overlap either source subvector
-+ V1[start1,end1)
-+ V2[start2,end2).
-+
-+vector-heap-sort-lib - vector heap sort
-+ heap-sort v < [start end] -> vector
-+ heap-sort! v < [start end] -> unspecific
-+
-+ These procedures sort their data using heap sort,
-+ which is not a stable sorting algorithm.
-+
-+ HEAP-SORT returns a vector of length END-START.
-+ HEAP-SORT! is in-place, leaving its result in V[start,end).
-+
-+vector-quick-sort-lib - vector quick sort
-+ quick-sort v < [start end] -> vector
-+ quick-sort! v < [start end] -> unspecific
-+
-+ These procedures sort their data using quick sort,
-+ which is not a stable sorting algorithm.
-+
-+ QUICK-SORT returns a vector of length END-START.
-+ QUICK-SORT! is in-place, leaving its result in V[start,end).
-+
-+vector-insert-sort-lib - vector insertion sort
-+ insert-sort v < [start end] -> vector
-+ insert-sort! v < [start end] -> unspecific
-+
-+ These procedures stably sort their data using insertion sort.
-+
-+ INSERT-SORT returns a vector of length END-START.
-+ INSERT-SORT! is in-place, leaving its result in V[start,end).
-+
-+delndup-lib - list and vector delete neighbor duplicates
-+ list-delete-neighbor-dups lis = -> list
-+ list-delete-neighbor-dups! lis = -> list
-+
-+ vector-delete-neighbor-dups v = [start end] -> vector
-+ vector-delete-neighbor-dups! v = [start end] -> end'
-+
-+ These procedures delete adjacent duplicate elements from a list or
-+ a vector, using a given element-equality procedure =. The first/leftmost
-+ element of a run of equal elements is the one that survives. The list
-+ or vector is not otherwise disordered.
-+
-+ These procedures are linear time -- much faster than the O(n^2) general
-+ duplicate-element deletors that do not assume any "bunching" of elements
-+ (such as the ones provided by SRFI-1). If you want to delete duplicate
-+ elements from a large list or vector, sort the elements to bring equal
-+ items together, then use one of these procedures, for a total time of
-+ O(n lg n).
-+
-+ The comparison function = passed to these procedures is always applied
-+ (= x y)
-+ where X comes before Y in the containing list or vector.
-+
-+ LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
-+ may share storage with the input list.
-+
-+ VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
-+ rather allocates a fresh vector to hold the result.
-+
-+ LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
-+ mutate its input list in order to construct its answer.
-+
-+ VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
-+ answer, packing its answer into the index range [start,end'), where
-+ END' is the non-negative exact integer returned as its value. It
-+ returns END' as its result. The vector is not altered outside the range
-+ [start,end').
-+
-+ Examples:
-+ (list-delete-neighbor-dups '(1 1 2 7 7 7 0 -2 -2) =)
-+ => (1 2 7 0 -2)
-+
-+ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) =)
-+ => #(1 2 7 0 -2)
-+
-+ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) = 3 7)
-+ => #(7 0 -2)
-+
-+ ;; Result left in v[3,9):
-+ (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
-+ (cons (vector-delete-neighbor-dups! v = 3)
-+ v))
-+ => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
-+
-+
-+-------------------------------------------------------------------------------
-+* Algorithmic properties
-+------------------------
-+Different sort and merge algorithms have different properties.
-+Choose the algorithm that matches your needs:
-+
-+Vector insert sort
-+ Stable, but only suitable for small vectors -- O(n^2).
-+
-+Vector quick sort
-+ Not stable. Is fast on average -- O(n lg n) -- but has bad worst-case
-+ behaviour. Has good memory locality for big vectors (unlike heap sort).
-+ A clever pivot-picking trick (median of three samples) helps avoid
-+ worst-case behaviour, but pathological cases can still blow up.
-+
-+Vector heap sort
-+ Not stable. Guaranteed fast -- O(n lg n) *worst* case. Poor locality
-+ on large vectors. A very reliable workhorse.
-+
-+Vector merge sort
-+ Stable. Not in-place -- requires a temporary buffer of equal size.
-+ Fast -- O(n lg n) -- and has good memory locality for large vectors.
-+
-+ The implementation of vector merge sort provided by this SRFI's reference
-+ implementation is, additionally, a "natural" sort, meaning that it
-+ exploits existing order in the input data, providing O(n) best case.
-+
-+Destructive list merge sort
-+ Stable, fast and in-place (i.e., allocates no new cons cells). "Fast"
-+ means O(n lg n) worse-case, and substantially better if the data
-+ is already mostly ordered, all the way down to linear time for
-+ a completely-ordered input list (i.e., it is a "natural" sort).
-+
-+ Note that sorting lists involves chasing pointers through memory, which
-+ can be a loser on modern machine architectures because of poor cache &
-+ page locality. Pointer *writing*, which is what the SET-CDR!s of a
-+ destructive list-sort algorithm do, is even worse, especially if your
-+ Scheme has a generational GC -- the writes will thrash the write-barrier.
-+ Sorting vectors has inherently better locality.
-+
-+ This SRFIs destructive list merge and merge sort implementations are
-+ opportunistic -- they avoid redundant SET-CDR!s, and try to take long
-+ already-ordered runs of list structure as-is when doing the merges.
-+
-+Pure list merge sort
-+ Stable and fast -- O(n lg n) worst-case, and possibly better, depending
-+ upon the input list (see above).
-+
-+
-+Algorithm Stable? Worst case Average case In-place
-+------------------------------------------------------
-+V insert Yes O(n^2) O(n^2) Yes
-+V quick No O(n^2) O(n lg n) Yes
-+V heap No O(n lg n) O(n lg n) Yes
-+V merge Yes O(n lg n) O(n lg n) No
-+L merge Yes O(n lg n) O(n lg n) Either
-+
-+
-+-------------------------------------------------------------------------------
-+* Topics to be resolved during discussion phase
-+-----------------------------------------------
-+I particularly solicit comments about the following topics.
-+
-+- Include VECTOR-BINARY-SEARCH ?
-+ Should we include
-+ (VECTOR-BINARY-SEARCH v key< elt->key key [start end])
-+ in the SRFI? It sort of goes with sorting; it's exactly ten lines of code.
-+
-+- Comparison function before or after the list/vector argument?
-+ Should it be
-+ (list-sort < lis)
-+ or
-+ (list-sort lis <)
-+ There is overwhelming consistency among the implementations: data first,
-+ < after. Only Chez does it differently.
-+
-+ I have done it in the backwards-compatible way. But I prefer the < first,
-+ data after way.
-+
-+
-+-------------------------------------------------------------------------------
-+* Porting and optimisation
-+--------------------------
-+This package should be trivial to port. There are only four non-R4RS bits
-+in the code:
-+- Use of multiple-value return, with the R5RS VALUES procedure, and the
-+ simple (RECEIVE (var ...) mv-exp body ...) multiple-value binding macro.
-+
-+- A VECTOR-COPY procedure. This is a tiny little procedure:
-+ (vector-copy v [start end])
-+
-+- Use of the LET-OPTIONALS macro from scsh to parse and default optional
-+ arguments to three routines. Again, easy to port the macro or rewrite
-+ the code to parse, default, and error check the args by hand.
-+
-+- Calls to an ERROR function for complaining about bad arguments.
-+
-+This code is tightly bummed, as far as I can go in portable Scheme.
-+
-+You could speed up the vector code a lot by error-checking the procedure
-+parameters and then shifting over to fixnum-specific arithmetic and
-+dangerous vector-indexing and vector-setting primitives. The comments
-+in the code indicate where the initial error checks would have to be
-+added. There are several (QUOTIENT N 2)'s that could be changed to a
-+fixnum right-shift, as well, in both the list and vector code. The code
-+is designed to enable this -- each file usually exports one or two "safe"
-+procedures that end up calling an internal "dangerous" primitive. The
-+little exported cover procedures are where you move the error checks.
-+
-+This should provide *big* speedups. In fact, all the code bumming I've done
-+pretty much disappears in the noise unless you have a good compiler and also
-+can dump the vector-index checks and generic arithmetic -- so I've really just
-+set things up for you to exploit.
-+
-+The optional-arg parsing, defaulting, and error checking is done with a
-+portable R4RS macro. But if your Scheme has a faster mechanism (e.g., Chez),
-+you should definitely port over to it. Note that argument defaulting and
-+error-checking are interleaved -- you don't have to error-check defaulted
-+START/END args to see if they are fixnums that are legal vector indices for
-+the corresponding vector, etc.
-+
-+
-+-------------------------------------------------------------------------------
-+* References & Links
-+--------------------
-+
-+This document, in HTML:
-+ http://srfi.schemers.org/srfi-32/srfi-32.html
-+ [This link may not be valid while the SRFI is in draft form.]
-+
-+This document, in simple text format:
-+ http://srfi.schemers.org/srfi-32/srfi-32.txt
-+
-+Archive of SRFI-32 discussion-list email:
-+ http://srfi.schemers.org/srfi-32/mail-archive/maillist.html
-+
-+SRFI web site:
-+ http://srfi.schemers.org/
-+
-+[CommonLisp]
-+ Common Lisp: the Language
-+ Guy L. Steele Jr. (editor).
-+ Digital Press, Maynard, Mass., second edition 1990.
-+ Available at http://www.elwood.com/alu/table/references.htm#cltl2
-+
-+ The Common Lisp "HyperSpec," produced by Kent Pitman, is essentially
-+ the ANSI spec for Common Lisp:
-+ http://www.xanalys.com/software_tools/reference/HyperSpec/
-+
-+[R5RS]
-+ Revised^5 Report on the Algorithmic Language Scheme,
-+ R. Kelsey, W. Clinger, J. Rees (editors).
-+ Higher-Order and Symbolic Computation, Vol. 11, No. 1, September, 1998.
-+ and ACM SIGPLAN Notices, Vol. 33, No. 9, October, 1998.
-+
-+ Available at http://www.schemers.org/Documents/Standards/
-+
-+
-+-------------------------------------------------------------------------------
-+* Acknowledgements
-+------------------
-+
-+I thank the authors of the open source I consulted when designing this
-+library, particularly Richard O'Keefe, Donovan Kolby and the MIT Scheme Team.
-+
-+
-+-------------------------------------------------------------------------------
-+* Copyright
-+-----------
-+
-+** SRFI text
-+============
-+This document is copyright (C) Olin Shivers (1998, 1999).
-+All Rights Reserved.
-+
-+Permission is hereby granted, free of charge, to any person obtaining
-+a copy of this software and associated documentation files (the
-+"Software"), to deal in the Software without restriction, including
-+without limitation the rights to use, copy, modify, merge, publish,
-+distribute, sublicense, and/or sell copies of the Software, and to
-+permit persons to whom the Software is furnished to do so, subject to
-+the following conditions:
-+
-+The above copyright notice and this permission notice shall be
-+included in all copies or substantial portions of the Software.
-+
-+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
-+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
-+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
-+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
-+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
-+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
-+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
-+
-+** Reference implementation
-+===========================
-+Short summary: no restrictions.
-+
-+While I wrote all of this code myself, I read a lot of code before I began
-+writing. However, all such code is, itself, either open source or public
-+domain, rendering irrelevant any issue of "copyright taint."
-+
-+The natural merge sorts (pure list, destructive list, and vector) are not only
-+my own code, but are implementations of an algorithm of my own devising. They
-+run in O(n lg n) worst case, O(n) best case, and require only a logarithmic
-+number of stack frames. And they are stable. And the destructive-list variant
-+allocates zero cons cells; it simply rearranges the cells of the input list.
-+
-+Hence the reference implementation is
-+ Copyright (c) 1998 by Olin Shivers.
-+and made available under the same copyright as the SRFI text (see above).
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib/srfi/%3a5/let.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5/let.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1 +0,0 @@
--#lang s-exp srfi/provider srfi/%3a5
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib/srfi/%3a5.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,3 +0,0 @@
--#lang s-exp srfi/provider srfi/5
--
--;; FIXME: "rest arguments" need to generate mutable lists
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib/srfi/5/let.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/5/let.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,86 +0,0 @@
--;;;
--;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
--;;; Time-stamp: <03/04/08 09:56:06 solsona>
--;;;
--;;; Usually, I would add a copyright notice, and the announce that
--;;; this code is under the LGPL licence. Nevertheless, I only did the
--;;; port to PLT Scheme v200, and here is the copyright notice,
--;;; comments, and licence from the original source:
--;;;
--;;; Copyright (C) Andy Gaynor (1999-2003)
--;;;
--;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
--;;; Renamed to s:let by Eli Barzilay
--
--#lang scheme/base
--(provide s:let)
--
--(define-syntax s:let
-- (syntax-rules ()
-- ;; standard
-- ((s:let () body ...)
-- (let () body ...))
-- ((s:let ((var val) ...) body ...)
-- (let ((var val) ...) body ...))
--
-- ;; rest style
-- ((s:let ((var val) . bindings) body ...)
-- (let-loop #f bindings (var) (val) (body ...)))
--
-- ;; signature style
-- ((s:let (name bindings ...) body ...)
-- (let-loop name (bindings ...) () () (body ...)))
--
-- ;; standard named style
-- ((s:let name (bindings ...) body ...)
-- (let-loop name (bindings ...) () () (body ...)))
--
-- ))
--
--;; A loop to walk down the list of bindings.
--
--(define-syntax let-loop
-- (syntax-rules ()
--
-- ;; No more bindings - make a LETREC.
-- ((let-loop name () (vars ...) (vals ...) body)
-- ((letrec ((name (lambda (vars ...) . body)))
-- name)
-- vals ...))
--
-- ;; Rest binding, no name
-- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-- (let ((var val) ... (rest-var (list rest-val ...))) . body))
--
-- ;; Process a (var val) pair.
-- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
-- (let-loop name (more ...) (vars ... var) (vals ... val) body))
--
-- ;; End with a rest variable - make a LETREC.
-- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
-- ((letrec ((name (lambda (vars ... . rest-var) . body)))
-- name)
-- vals ... rest-vals ...))))
--
--;; Four loops - normal and `signature-style', each with and without a rest
--;; binding.
--;;
--;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
--;; (if (= i n)
--;; f0
--;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
--;;
--;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
--;; (if (= i n)
--;; f0
--;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
--;;
--;;(let fibonacci ((n 10) (i 0) . (f 0 1))
--;; (if (= i n)
--;; (car f)
--;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
--;;
--;;(let (fibonacci (n 10) (i 0) . (f 0 1))
--;; (if (= i n)
--;; (car f)
--;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib/srfi/5.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib/srfi/5.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
-@@ -1,2 +0,0 @@
--;; module loader for SRFI-5
--#lang s-exp srfi/provider srfi/5/let #:unprefix s:
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/info.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1,14 @@
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define deps '("scheme-lib"
-+ "base"
-+ "r6rs-lib"
-+ "srfi-lib"
-+ "compatibility-lib"))
-+
-+
-+(define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"")
-+
-+(define pkg-authors '(mflatt noel chongkai jay))
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1 @@
-+#lang s-exp srfi/provider srfi/%3a5
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1,3 @@
-+#lang s-exp srfi/provider srfi/5
-+
-+;; FIXME: "rest arguments" need to generate mutable lists
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1,86 @@
-+;;;
-+;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
-+;;; Time-stamp: <03/04/08 09:56:06 solsona>
-+;;;
-+;;; Usually, I would add a copyright notice, and the announce that
-+;;; this code is under the LGPL licence. Nevertheless, I only did the
-+;;; port to PLT Scheme v200, and here is the copyright notice,
-+;;; comments, and licence from the original source:
-+;;;
-+;;; Copyright (C) Andy Gaynor (1999-2003)
-+;;;
-+;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
-+;;; Renamed to s:let by Eli Barzilay
-+
-+#lang scheme/base
-+(provide s:let)
-+
-+(define-syntax s:let
-+ (syntax-rules ()
-+ ;; standard
-+ ((s:let () body ...)
-+ (let () body ...))
-+ ((s:let ((var val) ...) body ...)
-+ (let ((var val) ...) body ...))
-+
-+ ;; rest style
-+ ((s:let ((var val) . bindings) body ...)
-+ (let-loop #f bindings (var) (val) (body ...)))
-+
-+ ;; signature style
-+ ((s:let (name bindings ...) body ...)
-+ (let-loop name (bindings ...) () () (body ...)))
-+
-+ ;; standard named style
-+ ((s:let name (bindings ...) body ...)
-+ (let-loop name (bindings ...) () () (body ...)))
-+
-+ ))
-+
-+;; A loop to walk down the list of bindings.
-+
-+(define-syntax let-loop
-+ (syntax-rules ()
-+
-+ ;; No more bindings - make a LETREC.
-+ ((let-loop name () (vars ...) (vals ...) body)
-+ ((letrec ((name (lambda (vars ...) . body)))
-+ name)
-+ vals ...))
-+
-+ ;; Rest binding, no name
-+ ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
-+ (let ((var val) ... (rest-var (list rest-val ...))) . body))
-+
-+ ;; Process a (var val) pair.
-+ ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
-+ (let-loop name (more ...) (vars ... var) (vals ... val) body))
-+
-+ ;; End with a rest variable - make a LETREC.
-+ ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
-+ ((letrec ((name (lambda (vars ... . rest-var) . body)))
-+ name)
-+ vals ... rest-vals ...))))
-+
-+;; Four loops - normal and `signature-style', each with and without a rest
-+;; binding.
-+;;
-+;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
-+;; (if (= i n)
-+;; f0
-+;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+;;
-+;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
-+;; (if (= i n)
-+;; f0
-+;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
-+;;
-+;;(let fibonacci ((n 10) (i 0) . (f 0 1))
-+;; (if (= i n)
-+;; (car f)
-+;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
-+;;
-+;;(let (fibonacci (n 10) (i 0) . (f 0 1))
-+;; (if (= i n)
-+;; (car f)
-+;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
-diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
---- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-01-31 12:22:58.159501944 -0500
-@@ -0,0 +1,2 @@
-+;; module loader for SRFI-5
-+#lang s-exp srfi/provider srfi/5/let #:unprefix s:
-diff -urN racket-6.12-orig/share/pkgs/srfi-lite-lib/info.rkt racket-6.12/share/pkgs/srfi-lite-lib/info.rkt
---- racket-6.12-orig/share/pkgs/srfi-lite-lib/info.rkt 2018-01-26 16:10:04.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lite-lib/info.rkt 2018-01-31 12:22:58.160501940 -0500
-@@ -1 +1,9 @@
--(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("base"))) (define pkg-desc "implementation of the most widely used \"srfi\" libraries") (define pkg-authors (quote (mflatt)))))
-+#lang info
-+
-+(define collection 'multi)
-+
-+(define deps '("base"))
-+
-+(define pkg-desc "implementation of the most widely used \"srfi\" libraries")
-+
-+(define pkg-authors '(mflatt))
-Binary files racket-6.12-orig/share/pkgs/srfi-lite-lib/.info.rkt.swp and racket-6.12/share/pkgs/srfi-lite-lib/.info.rkt.swp differ
-diff -urN racket-6.12-orig/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt racket-6.12/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt
---- racket-6.12-orig/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt 2018-01-26 15:34:52.000000000 -0500
-+++ racket-6.12/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt 2018-01-31 12:22:58.162501933 -0500
-@@ -2,6 +2,7 @@
-
- (require racket/contract/base
- racket/file
-+ racket/list
- (only-in racket/runtime-path define-runtime-path)
- racket/string racket/format
- syntax/modread
-@@ -113,10 +114,6 @@
- (lambda ()
- (with-input-from-file path read))))
- #t))))
-- (define (rdc ls)
-- (if (null? (cdr ls))
-- '()
-- (cons (car ls) (rdc (cdr ls)))))
-
- ;;Retrieve a localized template given its package name and a template name
- (define (localized-template package-name template-name)
-@@ -127,5 +124,5 @@
- (let ((bundle (hash-ref *localization-bundles* specifier #f)))
- (cond ((and bundle (assq template-name bundle)) => cdr)
- ((null? (cdr specifier)) #f)
-- (else (loop (rdc specifier))))))))
-+ (else (loop (drop-right specifier 1))))))))
- )
diff --git a/racket-7.0-fix-doc-open-url.patch b/racket-7.0-fix-doc-open-url.patch
new file mode 100644
index 0000000..7c497f5
--- /dev/null
+++ b/racket-7.0-fix-doc-open-url.patch
@@ -0,0 +1,69 @@
+diff -urN racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt racket-6.12/share/pkgs/scribble-lib/help/search.rkt
+--- racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:14:56.017273013 -0400
++++ racket-6.12/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:15:37.839113929 -0400
+@@ -17,33 +17,38 @@
+ #:notify [notify void])
+ (define open-url (get-doc-open-url))
+ (cond
+- [open-url
+- (define dest-url (let ([u (string->url open-url)])
+- (combine-url/relative
+- u
+- (string-join
+- (for/list ([s (explode-path sub)])
+- (if (path? s)
+- (path-element->string s)
+- (format "~a" s)))
+- "/"))))
+- (notify (url->string dest-url))
+- (send-url (url->string
+- (struct-copy url dest-url
+- [fragment (or fragment
+- (url-fragment dest-url))]
+- [query (append
+- (url-query dest-url)
+- (if query
+- (url-query
+- (string->url
+- (format "q?~a" query)))
+- null))])))]
+- [else
+- (let* ([path (build-path (find-user-doc-dir) sub)]
+- [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
+- (notify path)
+- (send-url/file path #:fragment fragment #:query query))]))
++ [open-url
++ (define dest-url (let ([u (string->url open-url)])
++ (combine-url/relative
++ u
++ (string-join
++ (for/list ([s (explode-path sub)])
++ (if (path? s)
++ (path-element->string s)
++ (format "~a" s)))
++ "/"))))
++ (notify (url->string dest-url))
++ (send-url (url->string
++ (struct-copy url dest-url
++ [fragment (or fragment
++ (url-fragment dest-url))]
++ [query (append
++ (url-query dest-url)
++ (if query
++ (url-query
++ (string->url
++ (format "q?~a" query)))
++ null))])))]
++ [else
++ (let* ([path (build-path (find-user-doc-dir) sub)]
++ [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
++ (notify path)
++ (if (file-exists? path)
++ (send-url/file path #:fragment fragment #:query query)
++ (let ([part (lambda (pfx x) (if x (string-append pfx x) ""))])
++ (send-url (string-append
++ "https://docs.racket-lang.org/"
++ sub (part "#" fragment) (part "?" query))))))]))
+
+ ;; This is an example of changing this code to use the online manuals.
+ ;; Normally, it's better to set `doc-open-url` in "etc/config.rktd",
diff --git a/racket-7.0-remove-nonfree.patch b/racket-7.0-remove-nonfree.patch
new file mode 100644
index 0000000..778001a
--- /dev/null
+++ b/racket-7.0-remove-nonfree.patch
@@ -0,0 +1,1074 @@
+diff -urN racket-7.0.orig/share/links.rktd racket-7.0/share/links.rktd
+--- racket-7.0.orig/share/links.rktd 2018-08-03 15:23:48.604561727 -0400
++++ racket-7.0/share/links.rktd 2018-08-03 15:23:45.481540939 -0400
+@@ -175,9 +175,7 @@
+ ("sasl" "pkgs/sasl-doc")
+ (root "pkgs/slideshow-exe")
+ (root "pkgs/slideshow-plugin")
+- (root "pkgs/srfi-lib-nonfree")
+ (root "pkgs/srfi-doc")
+- (root "pkgs/srfi-doc-nonfree")
+ (root "pkgs/syntax-color-doc")
+ (root "pkgs/web-server-lib")
+ (root "pkgs/unix-socket-lib")
+diff -urN racket-7.0.orig/share/pkgs/srfi/info.rkt racket-7.0/share/pkgs/srfi/info.rkt
+--- racket-7.0.orig/share/pkgs/srfi/info.rkt 2018-08-03 15:23:48.891563638 -0400
++++ racket-7.0/share/pkgs/srfi/info.rkt 2018-08-03 15:23:45.424540559 -0400
+@@ -1 +1 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "7.0"))) (define collection (quote multi)) (define deps (quote ("srfi-lib" "srfi-lib-nonfree" "srfi-doc" "srfi-doc-nonfree"))) (define implies (quote ("srfi-lib" "srfi-lib-nonfree" "srfi-doc" "srfi-doc-nonfree"))) (define pkg-desc "Legacy SRFI (Scheme) libraries") (define pkg-authors (quote (mflatt noel chongkai jay)))))
++(module info setup/infotab (#%module-begin (define package-content-state (quote (built "7.0"))) (define collection (quote multi)) (define deps (quote ("srfi-lib" "srfi-doc"))) (define implies (quote ("srfi-lib" "srfi-doc"))) (define pkg-desc "Legacy SRFI (Scheme) libraries") (define pkg-authors (quote (mflatt noel chongkai jay)))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/info.rkt racket-7.0/share/pkgs/srfi-doc-nonfree/info.rkt
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/info.rkt 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "7.0"))) (define collection (quote multi)) (define build-deps (quote ("mzscheme-doc" "scheme-lib" "base" "scribble-lib" "srfi-doc" "srfi-lib-nonfree" "racket-doc" "r5rs-doc" "r6rs-doc" "compatibility-lib"))) (define update-implies (quote ("srfi-lib-nonfree"))) (define pkg-desc "documentation part of \"srfi nonfree\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang info
+-
+-(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
+@@ -1,66 +0,0 @@
+-#lang scribble/doc
+-@(require srfi/scribblings/util
+- scribble/manual
+- scribble/eval
+- scriblib/render-cond
+- scribble/core
+- scribble/html-properties
+- (for-syntax scheme/base)
+- (for-label scheme/base
+- racket/stream))
+-
+-@; ----------------------------------------------------------------------
+-
+-@title{SRFI Nonfree Libraries and Documentation}
+-
+-The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
+-Implementation} (a.k.a. @deftech{SRFI}) process allows individual
+-members of the Scheme community to propose libraries and extensions to
+-be supported by multiple Scheme implementations.
+-
+-Racket is distributed with implementations of many SRFIs, most of
+-which can be implemented as libraries. To import the bindings of SRFI
+-@math{n}, use
+-
+-@racketblock[
+-(require @#,elem{@racketidfont{srfi/}@math{n}})
+-]
+-
+-This document lists the SRFIs that are supported by Racket and
+-provides a link to the original SRFI specification (which is also
+-distributed as part of Racket's documentation).
+-
+-The following SRFI specification documents are licensed restrictively.
+-
+-@table-of-contents[]
+-
+-
+-@; ----------------------------------------
+-
+-@srfi[5]{A compatible let form with signatures and rest arguments}
+-
+-@redirect[5 '(
+- (let #t "unnamed")
+-)]
+-
+-Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
+-
+-@; ----------------------------------------
+-
+-@srfi[29]{Localization}
+-
+-@redirect[29 '(
+- (current-language #f "current-language")
+- (current-country #f "current-country")
+- (current-locale-details #f "current-locale-details")
+- (declare-bundle! #f "declare-bundle!")
+- (store-bundle #f "store-bundle")
+- (load-bundle! #f "load-bundle!")
+- (localized-template #f "localized-template")
+-)]
+-
+-Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
+-
+-@; ----------------------------------------
+-
+-@index-section[]
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,507 +0,0 @@
+-<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
+-<html>
+- <head>
+- <meta name="generator" content="HTML Tidy, see www.w3.org">
+- <title>SRFI 29: Localization</title>
+- <meta name="author" content="Scott G. Miller">
+- <meta name="description" content="Localization">
+- </head>
+- <body>
+- <H1>Title</H1>
+-
+- SRFI 29: Localization
+-
+- <H1>Author</H1>
+-
+- Scott G. Miller
+-
+- <H1>Abstract</H1>
+-
+- This document specifies an interface to retrieving and
+- displaying locale sensitive messages. A Scheme program can
+- register one or more translations of templated messages, and
+- then write Scheme code that can transparently retrieve the
+- appropriate message for the locale under which the Scheme
+- system is running. <br>
+-
+-
+- <H1>Rationale</H1>
+-
+- <p>As any programmer that has ever had to deal with making his
+- or her code readable in more than one locale, the process of
+- sufficiently abstracting program messages from their
+- presentation to the user is non-trivial without help from the
+- programming language. Most modern programming language
+- libraries do provide some mechanism for performing this
+- separation.</p>
+-
+- <p>A portable API that allows a piece of code to run without
+- modification in different countries and under different
+- languages is a must for any non-trivial software project.
+- The interface should separate the logic of a program from
+- the myriad of translations that may be necessary.</p>
+-
+- <p>The interface described in this document provides such
+- functionality. The underlying implementation is also allowed to
+- use whatever datastructures it likes to provide access to the
+- translations in the most efficient manner possible. In
+- addition, the implementation is provided with standardized
+- functions that programs will use for accessing an external,
+- unspecified repository of translations.</p>
+-
+- <p>This interface <i>does not</i> cover all aspects of
+- localization, including support for non-latin characters,
+- number and date formatting, etc. Such functionality is the
+- scope of a future SRFI that may extend this one.</p>
+-
+- <H1>Dependencies</H1>
+-
+- An SRFI-29 conformant implementation must also implement
+- SRFI-28, Basic Format Strings. Message templates are strings
+- that must be processed by the <tt>format</tt> function
+- specified in that SRFI.
+-
+- <H1>Specification</H1>
+-
+- <h3>Message Bundles</h3>
+-
+- <p>A Message Bundle is a set of message templates and their
+- identifying keys. Each bundle contains one or more such
+- key/value pairs. The bundle itself is associated with a
+- <i>bundle specifier</i> which uniquely identifies the
+- bundle.</p>
+-
+- <h3>Bundle Specifiers</h3>
+-
+- <p>A Bundle Specifier is a Scheme list that describes, in order
+- of importance, the package and locale of a message bundle.
+- In most cases, a locale specifier will have between one
+- and three elements. The first element is a symbol denoting the
+- package for which this bundle applies. The second and third
+- elements denote a <i>locale</i>. The second element (first
+- element of the locale) if present, is the two letter, ISO 639-1
+- language code for the bundle. The third element, if present, is
+- a two letter ISO 3166-1 country code. In some cases, a
+- fourth element may be present, specifying the encoding used for
+- the bundle. All bundle specifier elements are Scheme
+- symbols.</p>
+-
+- <p>If only one translation is provided, it should be designated
+- only by a package name, for example <tt>(mathlib)</tt>. This
+- translation is called the <i>default</i> translation.</p>
+-
+- <h3>Bundle Searching</h3>
+-
+- <p>When a message template is retrieved from a bundle, the
+- Scheme implementation will provide the locale under which the
+- system is currently running. When the template is retrieved,
+- the package name will be specified. The Scheme system should
+- construct a Bundle Specifier from the provided package name and
+- the active locale. For example, when retrieving a message
+- template for French Canadian, in the <tt>mathlib</tt> package,
+- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
+- program may also retrieve the elements of the current locale
+- using the no-argument procedures:</p>
+-
+- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-language</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- When given no arguments, returns the current ISO 639-1
+- language code as a symbol. If provided with an
+- argument, the current language is set to that named by the
+- symbol for the currently executing Scheme thread (or for the
+- entire Scheme system if such a distinction is not possible).
+-
+- </blockquote>
+-
+- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-country</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- returns the current ISO 3166-1 country code as a symbol.
+- If provided with an argument, the current country is
+- set to that named by the symbol for the currently executing
+- Scheme thread (or for the entire Scheme system if such a
+- distinction is not possible).
+- </blockquote>
+-
+- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
+- symbol</i></tt>s<br>
+- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- Returns a list of additional locale details as a list of
+- symbols. This list may contain information about
+- encodings or other more specific information. If
+- provided with an argument, the current locale details are set
+- to those given in the currently executing Scheme thread (or
+- for the entire Scheme system if such a distinction is not
+- possible).
+- </blockquote>
+-
+- <p>The Scheme System should first check for a bundle with the
+- exact name provided. If no such bundle is found, the last
+- element from the list is removed and a search is tried for a
+- bundle with that name. If no bundle is then found, the list is
+- shortened by removing the last element again. If no message is
+- found and the bundle specifier is now the empty list, an error
+- should be raised.</p>
+-
+- <p>The reason for this search order is to provide the most
+- locale sensitive template possible, but to fall back on more
+- general templates if a translation has not yet been provided
+- for the given locale.</p>
+-
+- <h3>Message Templates</h3>
+-
+- <p>A message template is a localized message that may or may
+- not contain one of a number of formatting codes. A message
+- template is a Scheme string. The string is of a form that can
+- be processed by the <tt>format</tt> procedure found in many
+- Scheme systems and formally specified in SRFI-28 (Basic Format
+- Strings).</p>
+-
+- <p>This SRFI also extends SRFI-28 to provide an additional
+- <tt>format</tt> escape code:</p>
+-
+- <blockquote>
+- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
+- follows this code immediately to reference the [N]'th
+- optional value absolutely, rather than the next unconsumed
+- value. The referenced value is <i>not</i> consumed.
+- </blockquote>
+- This extension allows optional values to be positionally
+- referenced, so that message templates can be constructed that
+- can produce the proper word ordering for a language.
+-
+- <h3>Preparing Bundles</h3>
+- Before a bundle may be used by the Scheme system to retrieve
+- localized template messages, they must be made available to the
+- Scheme system. This SRFI specifies a way to portably
+- define the bundles, as well as store them in and retrieve them
+- from an unspecified system which may be provided by resources
+- outside the Scheme system.<br>
+-
+-
+- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
+- association-list</i> -> undefined<br>
+- </tt></p>
+-
+- <blockquote>
+- Declares a new bundle named by the given bundle-specifier.
+- The contents of the bundle are defined by the provided
+- association list. The list contains associations
+- between Scheme symbols and the message templates (Scheme
+- strings) they name. If a bundle already exists with the
+- given name, it is overwritten with the newly declared
+- bundle.<br>
+- </blockquote>
+- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to store a bundle named by the given bundle
+- specifier, and previously made available using
+- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
+- unspecified mechanism that may be persistent across Scheme
+- system restarts. If successful, a non-false value is
+- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
+- </blockquote>
+- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to retrieve a bundle from an unspecified mechanism
+- which stores bundles outside the Scheme system. If the
+- bundle was retrieved successfully, the function returns a
+- non-false value, and the bundle is immediately available to
+- the Scheme system. If the bundle could not be found or loaded
+- successfully, the function returns <tt>#f</tt>, and the
+- Scheme system's bundle registry remains unaffected.<br>
+- </blockquote>
+- A compliant Scheme system may choose not to provide any
+- external mechanism to store localized bundles. If it does
+- not, it must still provide implementations for
+- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
+- case, both functions must return <tt>#f</tt> regardless of the
+- arguments given. Users of this SRFI should recognize that the
+- inability to load or store a localized bundle in an external
+- repository is <i>not</i> a fatal error.<br>
+-
+-
+- <h3>Retrieving Localized Message Templates</h3>
+-
+- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
+- message-template-name</tt></i> <tt>-> <i>string or #f<br>
+- </i></tt></p>
+-
+- <blockquote>
+- Retrieves a localized message template for the given package
+- name and the given message template name (both symbols).
+- If no such message could be found, false (#f) is
+- returned.<br>
+- <br>
+- </blockquote>
+- After retrieving a template, the calling program can use
+- <tt>format</tt> to produce a string that can be displayed to
+- the user.<br>
+-
+-
+- <h2>Examples</h2>
+- The below example makes use of SRFI-29 to display simple,
+- localized messages. It also defines its bundles in such a
+- way that the Scheme system may store and retrieve the bundles
+- from a more efficient system catalog, if available.<br>
+-
+-<pre>
+-(let ((translations
+- '(((en) . ((time . "Its ~a, ~a.")
+- (goodbye . "Goodbye, ~a.")))
+- ((fr) . ((time . "~1@*~a, c'est ~a.")
+- (goodbye . "Au revoir, ~a."))))))
+- (for-each (lambda (translation)
+- (let ((bundle-name (cons 'hello-program (car translation))))
+- (if (not (load-bundle! bundle-name))
+- (begin
+- (declare-bundle! bundle-name (cdr translation))
+- (store-bundle! bundle-name)))))
+- translations))
+-
+-(define localized-message
+- (lambda (message-name . args)
+- (apply format (cons (localized-template 'hello-program
+- message-name)
+- args))))
+-
+-(let ((myname "Fred"))
+- (display (localized-message 'time "12:00" myname))
+- (display #\newline)
+-
+- (display (localized-message 'goodbye myname))
+- (display #\newline))
+-
+-;; Displays (English):
+-;; Its 12:00, Fred.
+-;; Goodbye, Fred.
+-;;
+-;; French:
+-;; Fred, c'est 12:00.
+-;; Au revoir, Fred.
+-</pre>
+-
+- <H1>Implementation</H1>
+-
+- <p>The implementation requires that the Scheme system provide a
+- definition for <tt>current-language</tt> and
+- <tt>current-country</tt> capable of distinguishing the correct
+- locale present during a Scheme session. The definitions of
+- those functions in the reference implementation are not capable
+- of that distinction. Their implementation is provided only so
+- that the following code can run in any R4RS scheme system.
+- <br>
+- </p>
+-
+- <p>In addition, the below implementation of a compliant
+- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
+- SRFI-23 (Error reporting)</p>
+-<pre>
+-;; The association list in which bundles will be stored
+-(define *localization-bundles* '())
+-
+-;; The current-language and current-country functions provided
+-;; here must be rewritten for each Scheme system to default to the
+-;; actual locale of the session
+-(define current-language
+- (let ((current-language-value 'en))
+- (lambda args
+- (if (null? args)
+- current-language-value
+- (set! current-language-value (car args))))))
+-
+-(define current-country
+- (let ((current-country-value 'us))
+- (lambda args
+- (if (null? args)
+- current-country-value
+- (set! current-country-value (car args))))))
+-
+-;; The load-bundle! and store-bundle! both return #f in this
+-;; reference implementation. A compliant implementation need
+-;; not rewrite these procedures.
+-(define load-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-(define store-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-;; Declare a bundle of templates with a given bundle specifier
+-(define declare-bundle!
+- (letrec ((remove-old-bundle
+- (lambda (specifier bundle)
+- (cond ((null? bundle) '())
+- ((equal? (caar bundle) specifier)
+- (cdr bundle))
+- (else (cons (car bundle)
+- (remove-old-bundle specifier
+- (cdr bundle))))))))
+- (lambda (bundle-specifier bundle-assoc-list)
+- (set! *localization-bundles*
+- (cons (cons bundle-specifier bundle-assoc-list)
+- (remove-old-bundle bundle-specifier
+- *localization-bundles*))))))
+-
+-;;Retrieve a localized template given its package name and a template name
+-(define localized-template
+- (letrec ((rdc
+- (lambda (ls)
+- (if (null? (cdr ls))
+- '()
+- (cons (car ls) (rdc (cdr ls))))))
+- (find-bundle
+- (lambda (specifier template-name)
+- (cond ((assoc specifier *localization-bundles*) =>
+- (lambda (bundle) bundle))
+- ((null? specifier) #f)
+- (else (find-bundle (rdc specifier)
+- template-name))))))
+- (lambda (package-name template-name)
+- (let loop ((specifier (cons package-name
+- (list (current-language)
+- (current-country)))))
+- (and (not (null? specifier))
+- (let ((bundle (find-bundle specifier template-name)))
+- (and bundle
+- (cond ((assq template-name bundle) => cdr)
+- ((null? (cdr specifier)) #f)
+- (else (loop (rdc specifier)))))))))))
+-
+-;;An SRFI-28 and SRFI-29 compliant version of format. It requires
+-;;SRFI-23 for error reporting.
+-(define format
+- (lambda (format-string . objects)
+- (let ((buffer (open-output-string)))
+- (let loop ((format-list (string->list format-string))
+- (objects objects)
+- (object-override #f))
+- (cond ((null? format-list) (get-output-string buffer))
+- ((char=? (car format-list) #\~)
+- (cond ((null? (cdr format-list))
+- (error 'format "Incomplete escape sequence"))
+- ((char-numeric? (cadr format-list))
+- (let posloop ((fl (cddr format-list))
+- (pos (string->number
+- (string (cadr format-list)))))
+- (cond ((null? fl)
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (null? (cdr fl)))
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (eq? (cadr fl) '#\*))
+- (loop (cddr fl) objects (list-ref objects pos)))
+- (else
+- (posloop (cdr fl)
+- (+ (* 10 pos)
+- (string->number
+- (string (car fl)))))))))
+- (else
+- (case (cadr format-list)
+- ((#\a)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (display (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\s)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (write (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\%)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\newline buffer)
+- (loop (cddr format-list) objects #f))
+- ((#\~)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\~ buffer)
+- (loop (cddr format-list) objects #f))
+- (else
+- (error 'format "Unrecognized escape sequence"))))))
+- (else (display (car format-list) buffer)
+- (loop (cdr format-list) objects #f)))))))
+-
+-</pre>
+-
+- <H1>Copyright</H1>
+-
+- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
+-
+- <p>This document and translations of it may be copied and
+- furnished to others, and derivative works that comment on or
+- otherwise explain it or assist in its implementation may be
+- prepared, copied, published and distributed, in whole or in
+- part, without restriction of any kind, provided that the above
+- copyright notice and this paragraph are included on all such
+- copies and derivative works. However, this document itself may
+- not be modified in any way, such as by removing the copyright
+- notice or references to the Scheme Request For Implementation
+- process or editors, except as needed for the purpose of
+- developing SRFIs in which case the procedures for copyrights
+- defined in the SRFI process must be followed, or as required to
+- translate it into languages other than English.</p>
+-
+- <p>The limited permissions granted above are perpetual and will
+- not be revoked by the authors or their successors or
+- assigns.</p>
+-
+- <p>This document and the information contained herein is
+- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
+- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
+- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
+- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
+- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
+- <hr>
+-
+- <address>
+- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
+- Rush</a>
+- </address>
+-
+- <address>
+- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
+- Miller</a>
+- </address>
+- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
+- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
+- Daylight Time 2002 <!-- hhmts end --> <br>
+- </body>
+-</html>
+-
+diff -urN racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
+--- racket-7.0.orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-08-03 15:23:48.895563664 -0400
++++ racket-7.0/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,345 +0,0 @@
+-<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
+-<html>
+-<head>
+- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
+-</head>
+-<body>
+-
+-<H1>Title</H1>
+-
+-SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
+-
+-<H1>Author</H1>
+-
+-Andy Gaynor
+-
+-<H1>Status</H1>
+-
+-This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
+-You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
+-<P><UL>
+-<LI>Received: 1999/2/2
+-<LI>Draft: 1999/2/10-1999/04/12
+-<LI>Final: 1999/4/26
+-<LI>Revised reference implementation: 2003/01/27
+-</UL>
+-
+-<H1>Abstract</H1>
+-
+-The <i>named-let</i> incarnation of the <code>let</code> form has two slight
+-inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
+-form makes no accommodation for rest arguments, an issue of functionality
+-and consistency. As defined, the <code>let</code> form does not accommodate
+-signature-style syntax, an issue of aesthetics and consistency. Both
+-issues are addressed here in a manner which is compatible with the traditional
+-<code>let</code> form but for minor extensions.
+-
+-<H1>Rationale</H1>
+-
+-<H2>Signature-style Syntax</H2>
+-
+-Consider the following two equivalent definitions:
+-
+-<p><pre>
+-(define fibonacci
+- (lambda (n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
+-
+-(define (fibonacci n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-Although there is a named-let analog for the former form, there is none
+-for the latter. To wit, suppose one wished to compute the 10th element
+-of the Fibonacci sequence using a named let:
+-
+-<p>
+-<pre>
+-(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-Values: 55
+-</pre>
+-
+-As it stands, one cannot equivalently write
+-
+-<p>
+-<pre>
+-(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-which is consistent with <code>define</code>'s signature-style form.
+-<p>Those that favor the signature style may prefer this extension.
+-In any case, it may be more appropriate to include all bound names within
+-the binding section. As presented, this straightforward extension
+-introduces no ambiguity or incompatibility with the existing definition
+-of let.
+-
+-<H2>Rest Arguments</H2>
+-
+-As it stands, one cannot write a named let with rest arguments, as in
+-
+-<p>
+-<pre>
+-(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))
+-</pre>
+-
+-otherwise equivalent to
+-
+-<p>
+-<pre>
+-(letrec ((blast (lambda (port . x)
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))))
+- (blast (current-output-port) (+ 1 2) 4 5))
+-</pre>
+-
+-While this example is rather contrived, the functionality is not.
+-There are several times when the author has used this construct in practice.
+-Regardless, there is little reason to deny the <code>let</code> form access to
+-all the features of lambda functionality.
+-
+-<H2>Symbols in Binding Sections</H2>
+-
+-Both the features above rely upon the placement of symbols in <code>let</code>
+-binding lists (this statement is intentially simplistic). The only
+-other apparent use of such symbol placement is to tersely bind variables
+-to unspecified values. For example, one might desire to use
+-<code>(let (foo bar baz) ...)</code>
+-to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
+-unspecified values.
+-
+-<p>This usage is considered less important in light of the rationales
+-presented above, and an alternate syntax is immediately apparent, as
+-in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
+-be preferable, consistently parenthesizing normal binding clauses.
+-
+-<H1>Specification</H1>
+-
+-<H2>Syntax</H2>
+-
+-<p>
+-A formal specification of the syntax follows. Below, body, expression,
+-and identifier are free. Each instantiation of binding-name must be
+-unique.
+-</p>
+-
+-<p>
+-<pre>
+- let = "(" "let" let-bindings body ")"
+- expressions = nothing | expression expressions
+- let-bindings = let-name bindings
+- |"(" let-name "." bindings ")"
+- let-name = identifier
+- bindings = "(" ")"
+- | rest-binding
+- | "(" normal-bindings ["." rest-binding] ")"
+-normal-bindings = nothing
+- | normal-binding normal-bindings
+-normal-binding = "(" binding-name expression ")"
+- binding-name = identifier
+- rest-binding = "(" binding-name expressions ")"
+-</pre>
+-
+-<p>
+-For clarity and convenience, an informal specification follows.
+-</p>
+-
+-<ol>
+-<li><a name="unnamed">Unnamed</a>
+-
+-<p><pre>
+-(let ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig">
+-Named, non-signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-sig">Named, signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-
+-<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-</li>
+-</ol>
+-
+-<H2>Semantics</H2>
+-
+-Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
+-and <code>letrec</code> forms, respectively.
+-
+-<ul>
+-<li>For informal syntax 1:
+-
+-<p><pre>
+-(($lambda (<parameter>...) <body>...) <argument>...)
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 2 and 3:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...) <body>...)))
+- (<name> <argument>...))
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 4 and 5:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...
+-
+-. <rest-parameter>) <body>...)))
+- (<name> <argument>... <rest-argument>...))
+-</pre>
+-</li>
+-</ul>
+-
+-<H1>Implementation</H1>
+-
+-Here is an implementation using <code>SYNTAX-RULES</code>.
+-
+-<p>
+-<pre>
+-;; Use your own standard let.
+-;; Or call a lambda.
+-;; (define-syntax standard-let
+-;;
+-;; (syntax-rules ()
+-;;
+-;; ((let ((var val) ...) body ...)
+-;; ((lambda (var ...) body ...) val ...))))
+-
+-(define-syntax let
+-
+- (syntax-rules ()
+-
+- ;; No bindings: use standard-let.
+- ((let () body ...)
+- (standard-let () body ...))
+- ;; Or call a lambda.
+- ;; ((lambda () body ...))
+-
+- ;; All standard bindings: use standard-let.
+- ((let ((var val) ...) body ...)
+- (standard-let ((var val) ...) body ...))
+- ;; Or call a lambda.
+- ;; ((lambda (var ...) body ...) val ...)
+-
+- ;; One standard binding: loop.
+- ;; The all-standard-bindings clause didn't match,
+- ;; so there must be a rest binding.
+- ((let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; Signature-style name: loop.
+- ((let (name binding ...) body ...)
+- (let-loop name (binding ...) () () (body ...)))
+-
+- ;; defun-style name: loop.
+- ((let name bindings body ...)
+- (let-loop name bindings () () (body ...)))))
+-
+-(define-syntax let-loop
+-
+- (syntax-rules ()
+-
+- ;; Standard binding: destructure and loop.
+- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
+- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
+-
+- ;; Rest binding, no name: use standard-let, listing the rest values.
+- ;; Because of let's first clause, there is no "no bindings, no name" clause.
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
+- ;; Or call a lambda with a rest parameter on all values.
+- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
+- ;; Or use one of several other reasonable alternatives.
+-
+- ;; No bindings, name: call a letrec'ed lambda.
+- ((let-loop name () (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ...) . body)))
+- name)
+- val ...))
+-
+- ;; Rest binding, name: call a letrec'ed lambda.
+- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ... . rest-var) . body)))
+- name)
+- val ... rest-val ...))))
+-</pre>
+-
+-
+-<H1>Copyright</H1>
+-
+-Copyright (C) Andy Gaynor (1999). All Rights Reserved.
+-<p>This document and translations of it may be copied and furnished to
+-others, and derivative works that comment on or otherwise explain it or
+-assist in its implementation may be prepared, copied, published and distributed,
+-in whole or in part, without restriction of any kind, provided that the
+-above copyright notice and this paragraph are included on all such copies
+-and derivative works. However, this document itself may not be modified
+-in any way, such as by removing the copyright notice or references to the
+-Scheme Request For Implementation process or editors, except as needed
+-for the purpose of developing SRFIs in which case the procedures for copyrights
+-defined in the SRFI process must be followed, or as required to translate
+-it into languages other than English.
+-<p>The limited permissions granted above are perpetual and will not be
+-revoked by the authors or their successors or assigns.
+-<p>This document and the information contained herein is provided on an
+-"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
+-EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
+-USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
+-WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
+-
+- <hr>
+- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
+-
+-</body>
+-</html>
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/info.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/info.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/info.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "7.0"))) (define collection (quote multi)) (define deps (quote ("scheme-lib" "base" "r6rs-lib" "srfi-lib" "compatibility-lib"))) (define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-#lang s-exp srfi/provider srfi/%3a5
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang s-exp srfi/provider srfi/5
+-
+-;; FIXME: "rest arguments" need to generate mutable lists
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,86 +0,0 @@
+-;;;
+-;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
+-;;; Time-stamp: <03/04/08 09:56:06 solsona>
+-;;;
+-;;; Usually, I would add a copyright notice, and the announce that
+-;;; this code is under the LGPL licence. Nevertheless, I only did the
+-;;; port to PLT Scheme v200, and here is the copyright notice,
+-;;; comments, and licence from the original source:
+-;;;
+-;;; Copyright (C) Andy Gaynor (1999-2003)
+-;;;
+-;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
+-;;; Renamed to s:let by Eli Barzilay
+-
+-#lang scheme/base
+-(provide s:let)
+-
+-(define-syntax s:let
+- (syntax-rules ()
+- ;; standard
+- ((s:let () body ...)
+- (let () body ...))
+- ((s:let ((var val) ...) body ...)
+- (let ((var val) ...) body ...))
+-
+- ;; rest style
+- ((s:let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; signature style
+- ((s:let (name bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ;; standard named style
+- ((s:let name (bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ))
+-
+-;; A loop to walk down the list of bindings.
+-
+-(define-syntax let-loop
+- (syntax-rules ()
+-
+- ;; No more bindings - make a LETREC.
+- ((let-loop name () (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ...) . body)))
+- name)
+- vals ...))
+-
+- ;; Rest binding, no name
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (let ((var val) ... (rest-var (list rest-val ...))) . body))
+-
+- ;; Process a (var val) pair.
+- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
+- (let-loop name (more ...) (vars ... var) (vals ... val) body))
+-
+- ;; End with a rest variable - make a LETREC.
+- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ... . rest-var) . body)))
+- name)
+- vals ... rest-vals ...))))
+-
+-;; Four loops - normal and `signature-style', each with and without a rest
+-;; binding.
+-;;
+-;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let fibonacci ((n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+diff -urN racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
+--- racket-7.0.orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-08-03 15:23:48.906563738 -0400
++++ racket-7.0/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,2 +0,0 @@
+-;; module loader for SRFI-5
+-#lang s-exp srfi/provider srfi/5/let #:unprefix s:
diff --git a/racket.spec b/racket.spec
index d3dc907..5fb06c9 100644
--- a/racket.spec
+++ b/racket.spec
@@ -1,38 +1,24 @@
Name: racket
-Version: 6.12
-Release: 8%{?dist}
+Version: 7.0
+Release: 1%{?dist}
Summary: General purpose programming language
License: GPLv3 and LGPLv3 and MIT
URL: https://racket-lang.org
Source0: https://mirror.racket-lang.org/installers/%{version}/%{name}-%{version}-src…
-# Update SRFI libraries to include upstream PR 5.
-# See: https://github.com/racket/srfi/pull/5 (merged)
-Patch0: racket-6.12-update-srfi.patch
-
# Remove SRFI library and docs with restrictive licensing.
# See: https://github.com/racket/srfi/issues/4 (open)
# Note: Upstream maintainers have confirmed this
# is safe, since the removed components are
# extra elements which nothing else in the
# package depends on.
-Patch1: racket-6.12-remove-nonfree.patch
-
-# Backport upstream fix of rpaths in racket/racket
-# which break check-buildroot
-# https://github.com/racket/racket/pull/1947 (merged)
-Patch2: racket-6.12-fix-rpaths.patch
-
-# Backport upstream fix of rpaths in racket/web-server
-# which break check-buildroot
-# https://github.com/racket/web-server/issues/36 (closed)
-Patch3: racket-6.12-fix-webserver-rpaths.patch
+Patch0: racket-7.0-remove-nonfree.patch
# Update racket-doc/search to use https://docs.racket-lang.org
# when no docs exist locally
# https://github.com/racket/scribble/pull/164 (open)
-Patch4: racket-6.12-fix-doc-open-url.patch
+Patch1: racket-7.0-fix-doc-open-url.patch
# Issue Building for s390x and armv7hl in koji
ExcludeArch: s390x armv7hl
@@ -131,6 +117,7 @@ for i in $DOCS_TO_FIX; do
${RPM_BUILD_ROOT}/%{_datadir}/doc/racket/$i
done
+# Remove the executable bit on legacy template file
chmod -x ${RPM_BUILD_ROOT}%{_libdir}/racket/starter-sh
%ldconfig_scriptlets
@@ -244,6 +231,10 @@ A local installation of the Racket documentation system.
%{_datadir}/doc/racket
%changelog
+* Fri Aug 3 2018 David Benoit <dbenoit(a)redhat.com> 7.0-1
+- Remove 6.12 patches and add update remove nonfree
+ srfi patch to 7.0
+
* Mon Jul 30 2018 David Benoit <dbenoit(a)redhat.com> 6.12-8
- Annotate dependencies with links to source code
- Move dependencies to racket-pkgs, since they are only used
diff --git a/sources b/sources
index 0fa455a..e43e6a3 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (racket-6.12-src.tgz) = e373cc1dd94f98d84dd2a105dad1cc2f2068118777b455af3f30ea3d190e6d601392bf17337652eeb84fa86d9c60aaa17563599b1772dc312211a8ac6e2c7060
+SHA512 (racket-7.0-src.tgz) = 98be3a7f0fbe38ae5febd2a9a3c81baede7dd7b3f3a9ed3ce68321389a68f74c6a8e24170013f7ca4fd725a79a3d5cfbda2e26f0f75163076e3b98f43094507f
commit 1b0a0799af2a5acee95845192bd3a772fd4a0f55
Author: dbenoit <dbenoit(a)redhat.com>
Date: Fri Aug 31 14:54:24 2018 -0400
fix sources file
diff --git a/.gitignore b/.gitignore
index e69de29..d013960 100644
--- a/.gitignore
+++ b/.gitignore
@@ -0,0 +1 @@
+/racket-6.12-src.tgz
diff --git a/sources b/sources
index 452da4f..0fa455a 100644
--- a/sources
+++ b/sources
@@ -1,6 +1 @@
-SHA512 (racket-6.12-fix-doc-open-url.patch) = 66a6a7f0f005a0da7acaf1ffbd53831eb796171994aa033bae5a2a973a79bb1c65212eaf4bf7ec02dfae5a186efcfa3e2a06b08bc31ff162cd4e08ab5b97bb94
-SHA512 (racket-6.12-fix-rpaths.patch) = d844d4d174b1cd92c61d16912aaf48afddbc9010acb2e5950d2d277d3110fc9256b68cbc38f06f69f6e2009f34085b32690a535049934d46d820e5b1491434e1
-SHA512 (racket-6.12-fix-webserver-rpaths.patch) = 807e356726dff3087bbcf505835082bcd2f0cdfb1d45e899d98bdf9b23715586b5c2a4a5703fff9ff2af9e46feb9936eaf685b6fb331388cd77dcc57389fbd76
-SHA512 (racket-6.12-remove-nonfree.patch) = 76e24e03728742073bdf4dc927da734195ddb6400cbcdaedecc09d94e7eac4168de4c9d30423bfd293b5fcd3580b0147b38c9061bded1205ec91e24f4868be2c
-SHA512 (racket-6.12-update-srfi.patch) = cfa97cc312ec3a40ad3dd4c4d662c949274325d6acc991c0d9b2327fc03b6c532cec0aa6e7f9a5563d0c23526742177367264f601f8e92055a1d6c3347e763a4
-SHA512 (racket.spec) = 5c53ad1906aa228848f7cabd72f377dd6da8f711669af515650d35fb40b5d3cb627c0b3daaf347cd2066ed250ce7fbcced7192e24f1aea7509eba1ceee487f49
+SHA512 (racket-6.12-src.tgz) = e373cc1dd94f98d84dd2a105dad1cc2f2068118777b455af3f30ea3d190e6d601392bf17337652eeb84fa86d9c60aaa17563599b1772dc312211a8ac6e2c7060
commit 335d72219631d93c3f29c29fd82070c0af593ef4
Author: dbenoit <dbenoit(a)redhat.com>
Date: Fri Aug 31 14:44:26 2018 -0400
fix gitignore
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..e69de29
diff --git a/sources b/sources
new file mode 100644
index 0000000..452da4f
--- /dev/null
+++ b/sources
@@ -0,0 +1,6 @@
+SHA512 (racket-6.12-fix-doc-open-url.patch) = 66a6a7f0f005a0da7acaf1ffbd53831eb796171994aa033bae5a2a973a79bb1c65212eaf4bf7ec02dfae5a186efcfa3e2a06b08bc31ff162cd4e08ab5b97bb94
+SHA512 (racket-6.12-fix-rpaths.patch) = d844d4d174b1cd92c61d16912aaf48afddbc9010acb2e5950d2d277d3110fc9256b68cbc38f06f69f6e2009f34085b32690a535049934d46d820e5b1491434e1
+SHA512 (racket-6.12-fix-webserver-rpaths.patch) = 807e356726dff3087bbcf505835082bcd2f0cdfb1d45e899d98bdf9b23715586b5c2a4a5703fff9ff2af9e46feb9936eaf685b6fb331388cd77dcc57389fbd76
+SHA512 (racket-6.12-remove-nonfree.patch) = 76e24e03728742073bdf4dc927da734195ddb6400cbcdaedecc09d94e7eac4168de4c9d30423bfd293b5fcd3580b0147b38c9061bded1205ec91e24f4868be2c
+SHA512 (racket-6.12-update-srfi.patch) = cfa97cc312ec3a40ad3dd4c4d662c949274325d6acc991c0d9b2327fc03b6c532cec0aa6e7f9a5563d0c23526742177367264f601f8e92055a1d6c3347e763a4
+SHA512 (racket.spec) = 5c53ad1906aa228848f7cabd72f377dd6da8f711669af515650d35fb40b5d3cb627c0b3daaf347cd2066ed250ce7fbcced7192e24f1aea7509eba1ceee487f49
commit f241dafce908418595fbb717060cbbefb6867e61
Author: dbenoit <dbenoit(a)redhat.com>
Date: Fri Aug 31 14:25:40 2018 -0400
add spec and patches
diff --git a/racket-6.12-fix-doc-open-url.patch b/racket-6.12-fix-doc-open-url.patch
new file mode 100644
index 0000000..7c497f5
--- /dev/null
+++ b/racket-6.12-fix-doc-open-url.patch
@@ -0,0 +1,69 @@
+diff -urN racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt racket-6.12/share/pkgs/scribble-lib/help/search.rkt
+--- racket-6.12-wbsvr/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:14:56.017273013 -0400
++++ racket-6.12/share/pkgs/scribble-lib/help/search.rkt 2018-03-19 17:15:37.839113929 -0400
+@@ -17,33 +17,38 @@
+ #:notify [notify void])
+ (define open-url (get-doc-open-url))
+ (cond
+- [open-url
+- (define dest-url (let ([u (string->url open-url)])
+- (combine-url/relative
+- u
+- (string-join
+- (for/list ([s (explode-path sub)])
+- (if (path? s)
+- (path-element->string s)
+- (format "~a" s)))
+- "/"))))
+- (notify (url->string dest-url))
+- (send-url (url->string
+- (struct-copy url dest-url
+- [fragment (or fragment
+- (url-fragment dest-url))]
+- [query (append
+- (url-query dest-url)
+- (if query
+- (url-query
+- (string->url
+- (format "q?~a" query)))
+- null))])))]
+- [else
+- (let* ([path (build-path (find-user-doc-dir) sub)]
+- [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
+- (notify path)
+- (send-url/file path #:fragment fragment #:query query))]))
++ [open-url
++ (define dest-url (let ([u (string->url open-url)])
++ (combine-url/relative
++ u
++ (string-join
++ (for/list ([s (explode-path sub)])
++ (if (path? s)
++ (path-element->string s)
++ (format "~a" s)))
++ "/"))))
++ (notify (url->string dest-url))
++ (send-url (url->string
++ (struct-copy url dest-url
++ [fragment (or fragment
++ (url-fragment dest-url))]
++ [query (append
++ (url-query dest-url)
++ (if query
++ (url-query
++ (string->url
++ (format "q?~a" query)))
++ null))])))]
++ [else
++ (let* ([path (build-path (find-user-doc-dir) sub)]
++ [path (if (file-exists? path) path (build-path (find-doc-dir) sub))])
++ (notify path)
++ (if (file-exists? path)
++ (send-url/file path #:fragment fragment #:query query)
++ (let ([part (lambda (pfx x) (if x (string-append pfx x) ""))])
++ (send-url (string-append
++ "https://docs.racket-lang.org/"
++ sub (part "#" fragment) (part "?" query))))))]))
+
+ ;; This is an example of changing this code to use the online manuals.
+ ;; Normally, it's better to set `doc-open-url` in "etc/config.rktd",
diff --git a/racket-6.12-fix-rpaths.patch b/racket-6.12-fix-rpaths.patch
new file mode 100644
index 0000000..f76730e
--- /dev/null
+++ b/racket-6.12-fix-rpaths.patch
@@ -0,0 +1,13 @@
+diff -urN racket-6.12-f/collects/racket/runtime-path.rkt racket-6.12/collects/racket/runtime-path.rkt
+--- racket-6.12-f/collects/racket/runtime-path.rkt 2018-02-06 16:04:29.969878308 -0500
++++ racket-6.12/collects/racket/runtime-path.rkt 2018-02-06 16:07:10.549176282 -0500
+@@ -186,7 +186,8 @@
+ (path-of
+ #,(datum->syntax
+ #'orig-stx
+- `(,#'this-expression-source-file))))
++ `(,#'this-expression-source-file)
++ #'orig-stx)))
+ #'void)])
+ (apply to-values (resolve-paths (#%variable-reference)
+ get-dir
diff --git a/racket-6.12-fix-webserver-rpaths.patch b/racket-6.12-fix-webserver-rpaths.patch
new file mode 100644
index 0000000..78614cf
--- /dev/null
+++ b/racket-6.12-fix-webserver-rpaths.patch
@@ -0,0 +1,35 @@
+diff -urN racket-6.12-rp/share/pkgs/web-server-lib/web-server/formlets/lib.rkt racket-6.12-wbsvr/share/pkgs/web-server-lib/web-server/formlets/lib.rkt
+--- racket-6.12-rp/share/pkgs/web-server-lib/web-server/formlets/lib.rkt 2018-02-22 18:42:06.131455756 -0500
++++ racket-6.12-wbsvr/share/pkgs/web-server-lib/web-server/formlets/lib.rkt 2018-03-20 15:07:09.666222028 -0400
+@@ -7,6 +7,7 @@
+ racket/function
+ racket/serialize
+ syntax/location
++ setup/collects
+ (for-syntax racket/base
+ syntax/parse))
+
+@@ -70,20 +71,20 @@
+ #:range-contracts (map (curry coerce-contract 'formlet/c)
+ contracts)))]))
+ (define quote-this-module-path
+- (quote-module-path))
++ (path->collects-relative (quote-module-path)))
+ (define-syntax formlet/c
+ (syntax-parser
+ [(_ range ...)
+ #:declare range (expr/c #'contract?
+ #:name "range contract argument")
+ #'(formlet/c** (-> listof-binding
+- (values (coerce-contract 'formlet/c range.c) ...)))]
++ (values (coerce-contract 'formlet/c range #;.c) ...)))]
+ [name:id
+ #`(contract
+ (-> contract? (... ...) contract?)
+ dynamic-formlet/c
+ quote-this-module-path
+- (quote-module-path)
++ (path->collects-relative (quote-module-path))
+ "formlet/c"
+ #'name)]))
+
diff --git a/racket-6.12-remove-nonfree.patch b/racket-6.12-remove-nonfree.patch
new file mode 100644
index 0000000..4e72a3e
--- /dev/null
+++ b/racket-6.12-remove-nonfree.patch
@@ -0,0 +1,1124 @@
+diff -urN racket-6.12-nf/share/links.rktd racket-6.12/share/links.rktd
+--- racket-6.12-nf/share/links.rktd 2018-01-31 13:04:06.757393819 -0500
++++ racket-6.12/share/links.rktd 2018-01-31 13:04:25.419309681 -0500
+@@ -133,7 +133,6 @@
+ (root "pkgs/r5rs-doc")
+ (root "pkgs/srfi-lite-lib")
+ (root "pkgs/srfi-lib")
+- (root "pkgs/srfi-lib-nonfree")
+ (root "pkgs/html-doc")
+ (root "pkgs/images-gui-lib")
+ (root "pkgs/images-doc")
+@@ -177,7 +176,6 @@
+ (root "pkgs/slideshow-exe")
+ (root "pkgs/slideshow-plugin")
+ (root "pkgs/srfi-doc")
+- (root "pkgs/srfi-doc-nonfree")
+ (root "pkgs/syntax-color-doc")
+ (root "pkgs/web-server-lib")
+ (root "pkgs/unix-socket-lib")
+diff -urN racket-6.12-nf/share/pkgs/srfi/info.rkt racket-6.12/share/pkgs/srfi/info.rkt
+--- racket-6.12-nf/share/pkgs/srfi/info.rkt 2018-01-31 13:04:07.216391749 -0500
++++ racket-6.12/share/pkgs/srfi/info.rkt 2018-01-31 13:05:32.601025227 -0500
+@@ -3,13 +3,9 @@
+ (define collection 'multi)
+
+ (define deps '("srfi-lib"
+- "srfi-lib-nonfree"
+- "srfi-doc"
+- "srfi-doc-nonfree"))
++ "srfi-doc"))
+ (define implies '("srfi-lib"
+- "srfi-lib-nonfree"
+- "srfi-doc"
+- "srfi-foc-nonfree"))
++ "srfi-doc"))
+
+ (define pkg-desc "Legacy SRFI (Scheme) libraries")
+
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/info.rkt 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,19 +0,0 @@
+-#lang info
+-
+-(define collection 'multi)
+-
+-(define build-deps '("mzscheme-doc"
+- "scheme-lib"
+- "base"
+- "scribble-lib"
+- "srfi-doc"
+- "srfi-lib-nonfree"
+- "racket-doc"
+- "r5rs-doc"
+- "r6rs-doc"
+- "compatibility-lib"))
+-(define update-implies '("srfi-lib-nonfree"))
+-
+-(define pkg-desc "documentation part of \"srfi nonfree\"")
+-
+-(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang info
+-
+-(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
+@@ -1,66 +0,0 @@
+-#lang scribble/doc
+-@(require srfi/scribblings/util
+- scribble/manual
+- scribble/eval
+- scriblib/render-cond
+- scribble/core
+- scribble/html-properties
+- (for-syntax scheme/base)
+- (for-label scheme/base
+- racket/stream))
+-
+-@; ----------------------------------------------------------------------
+-
+-@title{SRFI Nonfree Libraries and Documentation}
+-
+-The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
+-Implementation} (a.k.a. @deftech{SRFI}) process allows individual
+-members of the Scheme community to propose libraries and extensions to
+-be supported by multiple Scheme implementations.
+-
+-Racket is distributed with implementations of many SRFIs, most of
+-which can be implemented as libraries. To import the bindings of SRFI
+-@math{n}, use
+-
+-@racketblock[
+-(require @#,elem{@racketidfont{srfi/}@math{n}})
+-]
+-
+-This document lists the SRFIs that are supported by Racket and
+-provides a link to the original SRFI specification (which is also
+-distributed as part of Racket's documentation).
+-
+-The following SRFI specification documents are licensed restrictively.
+-
+-@table-of-contents[]
+-
+-
+-@; ----------------------------------------
+-
+-@srfi[5]{A compatible let form with signatures and rest arguments}
+-
+-@redirect[5 '(
+- (let #t "unnamed")
+-)]
+-
+-Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
+-
+-@; ----------------------------------------
+-
+-@srfi[29]{Localization}
+-
+-@redirect[29 '(
+- (current-language #f "current-language")
+- (current-country #f "current-country")
+- (current-locale-details #f "current-locale-details")
+- (declare-bundle! #f "declare-bundle!")
+- (store-bundle #f "store-bundle")
+- (load-bundle! #f "load-bundle!")
+- (localized-template #f "localized-template")
+-)]
+-
+-Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
+-
+-@; ----------------------------------------
+-
+-@index-section[]
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,507 +0,0 @@
+-<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
+-<html>
+- <head>
+- <meta name="generator" content="HTML Tidy, see www.w3.org">
+- <title>SRFI 29: Localization</title>
+- <meta name="author" content="Scott G. Miller">
+- <meta name="description" content="Localization">
+- </head>
+- <body>
+- <H1>Title</H1>
+-
+- SRFI 29: Localization
+-
+- <H1>Author</H1>
+-
+- Scott G. Miller
+-
+- <H1>Abstract</H1>
+-
+- This document specifies an interface to retrieving and
+- displaying locale sensitive messages. A Scheme program can
+- register one or more translations of templated messages, and
+- then write Scheme code that can transparently retrieve the
+- appropriate message for the locale under which the Scheme
+- system is running. <br>
+-
+-
+- <H1>Rationale</H1>
+-
+- <p>As any programmer that has ever had to deal with making his
+- or her code readable in more than one locale, the process of
+- sufficiently abstracting program messages from their
+- presentation to the user is non-trivial without help from the
+- programming language. Most modern programming language
+- libraries do provide some mechanism for performing this
+- separation.</p>
+-
+- <p>A portable API that allows a piece of code to run without
+- modification in different countries and under different
+- languages is a must for any non-trivial software project.
+- The interface should separate the logic of a program from
+- the myriad of translations that may be necessary.</p>
+-
+- <p>The interface described in this document provides such
+- functionality. The underlying implementation is also allowed to
+- use whatever datastructures it likes to provide access to the
+- translations in the most efficient manner possible. In
+- addition, the implementation is provided with standardized
+- functions that programs will use for accessing an external,
+- unspecified repository of translations.</p>
+-
+- <p>This interface <i>does not</i> cover all aspects of
+- localization, including support for non-latin characters,
+- number and date formatting, etc. Such functionality is the
+- scope of a future SRFI that may extend this one.</p>
+-
+- <H1>Dependencies</H1>
+-
+- An SRFI-29 conformant implementation must also implement
+- SRFI-28, Basic Format Strings. Message templates are strings
+- that must be processed by the <tt>format</tt> function
+- specified in that SRFI.
+-
+- <H1>Specification</H1>
+-
+- <h3>Message Bundles</h3>
+-
+- <p>A Message Bundle is a set of message templates and their
+- identifying keys. Each bundle contains one or more such
+- key/value pairs. The bundle itself is associated with a
+- <i>bundle specifier</i> which uniquely identifies the
+- bundle.</p>
+-
+- <h3>Bundle Specifiers</h3>
+-
+- <p>A Bundle Specifier is a Scheme list that describes, in order
+- of importance, the package and locale of a message bundle.
+- In most cases, a locale specifier will have between one
+- and three elements. The first element is a symbol denoting the
+- package for which this bundle applies. The second and third
+- elements denote a <i>locale</i>. The second element (first
+- element of the locale) if present, is the two letter, ISO 639-1
+- language code for the bundle. The third element, if present, is
+- a two letter ISO 3166-1 country code. In some cases, a
+- fourth element may be present, specifying the encoding used for
+- the bundle. All bundle specifier elements are Scheme
+- symbols.</p>
+-
+- <p>If only one translation is provided, it should be designated
+- only by a package name, for example <tt>(mathlib)</tt>. This
+- translation is called the <i>default</i> translation.</p>
+-
+- <h3>Bundle Searching</h3>
+-
+- <p>When a message template is retrieved from a bundle, the
+- Scheme implementation will provide the locale under which the
+- system is currently running. When the template is retrieved,
+- the package name will be specified. The Scheme system should
+- construct a Bundle Specifier from the provided package name and
+- the active locale. For example, when retrieving a message
+- template for French Canadian, in the <tt>mathlib</tt> package,
+- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
+- program may also retrieve the elements of the current locale
+- using the no-argument procedures:</p>
+-
+- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-language</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- When given no arguments, returns the current ISO 639-1
+- language code as a symbol. If provided with an
+- argument, the current language is set to that named by the
+- symbol for the currently executing Scheme thread (or for the
+- entire Scheme system if such a distinction is not possible).
+-
+- </blockquote>
+-
+- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-country</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- returns the current ISO 3166-1 country code as a symbol.
+- If provided with an argument, the current country is
+- set to that named by the symbol for the currently executing
+- Scheme thread (or for the entire Scheme system if such a
+- distinction is not possible).
+- </blockquote>
+-
+- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
+- symbol</i></tt>s<br>
+- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- Returns a list of additional locale details as a list of
+- symbols. This list may contain information about
+- encodings or other more specific information. If
+- provided with an argument, the current locale details are set
+- to those given in the currently executing Scheme thread (or
+- for the entire Scheme system if such a distinction is not
+- possible).
+- </blockquote>
+-
+- <p>The Scheme System should first check for a bundle with the
+- exact name provided. If no such bundle is found, the last
+- element from the list is removed and a search is tried for a
+- bundle with that name. If no bundle is then found, the list is
+- shortened by removing the last element again. If no message is
+- found and the bundle specifier is now the empty list, an error
+- should be raised.</p>
+-
+- <p>The reason for this search order is to provide the most
+- locale sensitive template possible, but to fall back on more
+- general templates if a translation has not yet been provided
+- for the given locale.</p>
+-
+- <h3>Message Templates</h3>
+-
+- <p>A message template is a localized message that may or may
+- not contain one of a number of formatting codes. A message
+- template is a Scheme string. The string is of a form that can
+- be processed by the <tt>format</tt> procedure found in many
+- Scheme systems and formally specified in SRFI-28 (Basic Format
+- Strings).</p>
+-
+- <p>This SRFI also extends SRFI-28 to provide an additional
+- <tt>format</tt> escape code:</p>
+-
+- <blockquote>
+- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
+- follows this code immediately to reference the [N]'th
+- optional value absolutely, rather than the next unconsumed
+- value. The referenced value is <i>not</i> consumed.
+- </blockquote>
+- This extension allows optional values to be positionally
+- referenced, so that message templates can be constructed that
+- can produce the proper word ordering for a language.
+-
+- <h3>Preparing Bundles</h3>
+- Before a bundle may be used by the Scheme system to retrieve
+- localized template messages, they must be made available to the
+- Scheme system. This SRFI specifies a way to portably
+- define the bundles, as well as store them in and retrieve them
+- from an unspecified system which may be provided by resources
+- outside the Scheme system.<br>
+-
+-
+- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
+- association-list</i> -> undefined<br>
+- </tt></p>
+-
+- <blockquote>
+- Declares a new bundle named by the given bundle-specifier.
+- The contents of the bundle are defined by the provided
+- association list. The list contains associations
+- between Scheme symbols and the message templates (Scheme
+- strings) they name. If a bundle already exists with the
+- given name, it is overwritten with the newly declared
+- bundle.<br>
+- </blockquote>
+- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to store a bundle named by the given bundle
+- specifier, and previously made available using
+- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
+- unspecified mechanism that may be persistent across Scheme
+- system restarts. If successful, a non-false value is
+- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
+- </blockquote>
+- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to retrieve a bundle from an unspecified mechanism
+- which stores bundles outside the Scheme system. If the
+- bundle was retrieved successfully, the function returns a
+- non-false value, and the bundle is immediately available to
+- the Scheme system. If the bundle could not be found or loaded
+- successfully, the function returns <tt>#f</tt>, and the
+- Scheme system's bundle registry remains unaffected.<br>
+- </blockquote>
+- A compliant Scheme system may choose not to provide any
+- external mechanism to store localized bundles. If it does
+- not, it must still provide implementations for
+- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
+- case, both functions must return <tt>#f</tt> regardless of the
+- arguments given. Users of this SRFI should recognize that the
+- inability to load or store a localized bundle in an external
+- repository is <i>not</i> a fatal error.<br>
+-
+-
+- <h3>Retrieving Localized Message Templates</h3>
+-
+- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
+- message-template-name</tt></i> <tt>-> <i>string or #f<br>
+- </i></tt></p>
+-
+- <blockquote>
+- Retrieves a localized message template for the given package
+- name and the given message template name (both symbols).
+- If no such message could be found, false (#f) is
+- returned.<br>
+- <br>
+- </blockquote>
+- After retrieving a template, the calling program can use
+- <tt>format</tt> to produce a string that can be displayed to
+- the user.<br>
+-
+-
+- <h2>Examples</h2>
+- The below example makes use of SRFI-29 to display simple,
+- localized messages. It also defines its bundles in such a
+- way that the Scheme system may store and retrieve the bundles
+- from a more efficient system catalog, if available.<br>
+-
+-<pre>
+-(let ((translations
+- '(((en) . ((time . "Its ~a, ~a.")
+- (goodbye . "Goodbye, ~a.")))
+- ((fr) . ((time . "~1@*~a, c'est ~a.")
+- (goodbye . "Au revoir, ~a."))))))
+- (for-each (lambda (translation)
+- (let ((bundle-name (cons 'hello-program (car translation))))
+- (if (not (load-bundle! bundle-name))
+- (begin
+- (declare-bundle! bundle-name (cdr translation))
+- (store-bundle! bundle-name)))))
+- translations))
+-
+-(define localized-message
+- (lambda (message-name . args)
+- (apply format (cons (localized-template 'hello-program
+- message-name)
+- args))))
+-
+-(let ((myname "Fred"))
+- (display (localized-message 'time "12:00" myname))
+- (display #\newline)
+-
+- (display (localized-message 'goodbye myname))
+- (display #\newline))
+-
+-;; Displays (English):
+-;; Its 12:00, Fred.
+-;; Goodbye, Fred.
+-;;
+-;; French:
+-;; Fred, c'est 12:00.
+-;; Au revoir, Fred.
+-</pre>
+-
+- <H1>Implementation</H1>
+-
+- <p>The implementation requires that the Scheme system provide a
+- definition for <tt>current-language</tt> and
+- <tt>current-country</tt> capable of distinguishing the correct
+- locale present during a Scheme session. The definitions of
+- those functions in the reference implementation are not capable
+- of that distinction. Their implementation is provided only so
+- that the following code can run in any R4RS scheme system.
+- <br>
+- </p>
+-
+- <p>In addition, the below implementation of a compliant
+- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
+- SRFI-23 (Error reporting)</p>
+-<pre>
+-;; The association list in which bundles will be stored
+-(define *localization-bundles* '())
+-
+-;; The current-language and current-country functions provided
+-;; here must be rewritten for each Scheme system to default to the
+-;; actual locale of the session
+-(define current-language
+- (let ((current-language-value 'en))
+- (lambda args
+- (if (null? args)
+- current-language-value
+- (set! current-language-value (car args))))))
+-
+-(define current-country
+- (let ((current-country-value 'us))
+- (lambda args
+- (if (null? args)
+- current-country-value
+- (set! current-country-value (car args))))))
+-
+-;; The load-bundle! and store-bundle! both return #f in this
+-;; reference implementation. A compliant implementation need
+-;; not rewrite these procedures.
+-(define load-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-(define store-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-;; Declare a bundle of templates with a given bundle specifier
+-(define declare-bundle!
+- (letrec ((remove-old-bundle
+- (lambda (specifier bundle)
+- (cond ((null? bundle) '())
+- ((equal? (caar bundle) specifier)
+- (cdr bundle))
+- (else (cons (car bundle)
+- (remove-old-bundle specifier
+- (cdr bundle))))))))
+- (lambda (bundle-specifier bundle-assoc-list)
+- (set! *localization-bundles*
+- (cons (cons bundle-specifier bundle-assoc-list)
+- (remove-old-bundle bundle-specifier
+- *localization-bundles*))))))
+-
+-;;Retrieve a localized template given its package name and a template name
+-(define localized-template
+- (letrec ((rdc
+- (lambda (ls)
+- (if (null? (cdr ls))
+- '()
+- (cons (car ls) (rdc (cdr ls))))))
+- (find-bundle
+- (lambda (specifier template-name)
+- (cond ((assoc specifier *localization-bundles*) =>
+- (lambda (bundle) bundle))
+- ((null? specifier) #f)
+- (else (find-bundle (rdc specifier)
+- template-name))))))
+- (lambda (package-name template-name)
+- (let loop ((specifier (cons package-name
+- (list (current-language)
+- (current-country)))))
+- (and (not (null? specifier))
+- (let ((bundle (find-bundle specifier template-name)))
+- (and bundle
+- (cond ((assq template-name bundle) => cdr)
+- ((null? (cdr specifier)) #f)
+- (else (loop (rdc specifier)))))))))))
+-
+-;;An SRFI-28 and SRFI-29 compliant version of format. It requires
+-;;SRFI-23 for error reporting.
+-(define format
+- (lambda (format-string . objects)
+- (let ((buffer (open-output-string)))
+- (let loop ((format-list (string->list format-string))
+- (objects objects)
+- (object-override #f))
+- (cond ((null? format-list) (get-output-string buffer))
+- ((char=? (car format-list) #\~)
+- (cond ((null? (cdr format-list))
+- (error 'format "Incomplete escape sequence"))
+- ((char-numeric? (cadr format-list))
+- (let posloop ((fl (cddr format-list))
+- (pos (string->number
+- (string (cadr format-list)))))
+- (cond ((null? fl)
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (null? (cdr fl)))
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (eq? (cadr fl) '#\*))
+- (loop (cddr fl) objects (list-ref objects pos)))
+- (else
+- (posloop (cdr fl)
+- (+ (* 10 pos)
+- (string->number
+- (string (car fl)))))))))
+- (else
+- (case (cadr format-list)
+- ((#\a)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (display (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\s)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (write (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\%)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\newline buffer)
+- (loop (cddr format-list) objects #f))
+- ((#\~)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\~ buffer)
+- (loop (cddr format-list) objects #f))
+- (else
+- (error 'format "Unrecognized escape sequence"))))))
+- (else (display (car format-list) buffer)
+- (loop (cdr format-list) objects #f)))))))
+-
+-</pre>
+-
+- <H1>Copyright</H1>
+-
+- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
+-
+- <p>This document and translations of it may be copied and
+- furnished to others, and derivative works that comment on or
+- otherwise explain it or assist in its implementation may be
+- prepared, copied, published and distributed, in whole or in
+- part, without restriction of any kind, provided that the above
+- copyright notice and this paragraph are included on all such
+- copies and derivative works. However, this document itself may
+- not be modified in any way, such as by removing the copyright
+- notice or references to the Scheme Request For Implementation
+- process or editors, except as needed for the purpose of
+- developing SRFIs in which case the procedures for copyrights
+- defined in the SRFI process must be followed, or as required to
+- translate it into languages other than English.</p>
+-
+- <p>The limited permissions granted above are perpetual and will
+- not be revoked by the authors or their successors or
+- assigns.</p>
+-
+- <p>This document and the information contained herein is
+- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
+- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
+- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
+- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
+- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
+- <hr>
+-
+- <address>
+- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
+- Rush</a>
+- </address>
+-
+- <address>
+- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
+- Miller</a>
+- </address>
+- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
+- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
+- Daylight Time 2002 <!-- hhmts end --> <br>
+- </body>
+-</html>
+-
+diff -urN racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
+--- racket-6.12-nf/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-01-31 13:04:07.221391727 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,345 +0,0 @@
+-<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
+-<html>
+-<head>
+- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
+-</head>
+-<body>
+-
+-<H1>Title</H1>
+-
+-SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
+-
+-<H1>Author</H1>
+-
+-Andy Gaynor
+-
+-<H1>Status</H1>
+-
+-This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
+-You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
+-<P><UL>
+-<LI>Received: 1999/2/2
+-<LI>Draft: 1999/2/10-1999/04/12
+-<LI>Final: 1999/4/26
+-<LI>Revised reference implementation: 2003/01/27
+-</UL>
+-
+-<H1>Abstract</H1>
+-
+-The <i>named-let</i> incarnation of the <code>let</code> form has two slight
+-inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
+-form makes no accommodation for rest arguments, an issue of functionality
+-and consistency. As defined, the <code>let</code> form does not accommodate
+-signature-style syntax, an issue of aesthetics and consistency. Both
+-issues are addressed here in a manner which is compatible with the traditional
+-<code>let</code> form but for minor extensions.
+-
+-<H1>Rationale</H1>
+-
+-<H2>Signature-style Syntax</H2>
+-
+-Consider the following two equivalent definitions:
+-
+-<p><pre>
+-(define fibonacci
+- (lambda (n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
+-
+-(define (fibonacci n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-Although there is a named-let analog for the former form, there is none
+-for the latter. To wit, suppose one wished to compute the 10th element
+-of the Fibonacci sequence using a named let:
+-
+-<p>
+-<pre>
+-(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-Values: 55
+-</pre>
+-
+-As it stands, one cannot equivalently write
+-
+-<p>
+-<pre>
+-(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-which is consistent with <code>define</code>'s signature-style form.
+-<p>Those that favor the signature style may prefer this extension.
+-In any case, it may be more appropriate to include all bound names within
+-the binding section. As presented, this straightforward extension
+-introduces no ambiguity or incompatibility with the existing definition
+-of let.
+-
+-<H2>Rest Arguments</H2>
+-
+-As it stands, one cannot write a named let with rest arguments, as in
+-
+-<p>
+-<pre>
+-(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))
+-</pre>
+-
+-otherwise equivalent to
+-
+-<p>
+-<pre>
+-(letrec ((blast (lambda (port . x)
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))))
+- (blast (current-output-port) (+ 1 2) 4 5))
+-</pre>
+-
+-While this example is rather contrived, the functionality is not.
+-There are several times when the author has used this construct in practice.
+-Regardless, there is little reason to deny the <code>let</code> form access to
+-all the features of lambda functionality.
+-
+-<H2>Symbols in Binding Sections</H2>
+-
+-Both the features above rely upon the placement of symbols in <code>let</code>
+-binding lists (this statement is intentially simplistic). The only
+-other apparent use of such symbol placement is to tersely bind variables
+-to unspecified values. For example, one might desire to use
+-<code>(let (foo bar baz) ...)</code>
+-to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
+-unspecified values.
+-
+-<p>This usage is considered less important in light of the rationales
+-presented above, and an alternate syntax is immediately apparent, as
+-in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
+-be preferable, consistently parenthesizing normal binding clauses.
+-
+-<H1>Specification</H1>
+-
+-<H2>Syntax</H2>
+-
+-<p>
+-A formal specification of the syntax follows. Below, body, expression,
+-and identifier are free. Each instantiation of binding-name must be
+-unique.
+-</p>
+-
+-<p>
+-<pre>
+- let = "(" "let" let-bindings body ")"
+- expressions = nothing | expression expressions
+- let-bindings = let-name bindings
+- |"(" let-name "." bindings ")"
+- let-name = identifier
+- bindings = "(" ")"
+- | rest-binding
+- | "(" normal-bindings ["." rest-binding] ")"
+-normal-bindings = nothing
+- | normal-binding normal-bindings
+-normal-binding = "(" binding-name expression ")"
+- binding-name = identifier
+- rest-binding = "(" binding-name expressions ")"
+-</pre>
+-
+-<p>
+-For clarity and convenience, an informal specification follows.
+-</p>
+-
+-<ol>
+-<li><a name="unnamed">Unnamed</a>
+-
+-<p><pre>
+-(let ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig">
+-Named, non-signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-sig">Named, signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-
+-<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-</li>
+-</ol>
+-
+-<H2>Semantics</H2>
+-
+-Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
+-and <code>letrec</code> forms, respectively.
+-
+-<ul>
+-<li>For informal syntax 1:
+-
+-<p><pre>
+-(($lambda (<parameter>...) <body>...) <argument>...)
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 2 and 3:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...) <body>...)))
+- (<name> <argument>...))
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 4 and 5:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...
+-
+-. <rest-parameter>) <body>...)))
+- (<name> <argument>... <rest-argument>...))
+-</pre>
+-</li>
+-</ul>
+-
+-<H1>Implementation</H1>
+-
+-Here is an implementation using <code>SYNTAX-RULES</code>.
+-
+-<p>
+-<pre>
+-;; Use your own standard let.
+-;; Or call a lambda.
+-;; (define-syntax standard-let
+-;;
+-;; (syntax-rules ()
+-;;
+-;; ((let ((var val) ...) body ...)
+-;; ((lambda (var ...) body ...) val ...))))
+-
+-(define-syntax let
+-
+- (syntax-rules ()
+-
+- ;; No bindings: use standard-let.
+- ((let () body ...)
+- (standard-let () body ...))
+- ;; Or call a lambda.
+- ;; ((lambda () body ...))
+-
+- ;; All standard bindings: use standard-let.
+- ((let ((var val) ...) body ...)
+- (standard-let ((var val) ...) body ...))
+- ;; Or call a lambda.
+- ;; ((lambda (var ...) body ...) val ...)
+-
+- ;; One standard binding: loop.
+- ;; The all-standard-bindings clause didn't match,
+- ;; so there must be a rest binding.
+- ((let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; Signature-style name: loop.
+- ((let (name binding ...) body ...)
+- (let-loop name (binding ...) () () (body ...)))
+-
+- ;; defun-style name: loop.
+- ((let name bindings body ...)
+- (let-loop name bindings () () (body ...)))))
+-
+-(define-syntax let-loop
+-
+- (syntax-rules ()
+-
+- ;; Standard binding: destructure and loop.
+- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
+- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
+-
+- ;; Rest binding, no name: use standard-let, listing the rest values.
+- ;; Because of let's first clause, there is no "no bindings, no name" clause.
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
+- ;; Or call a lambda with a rest parameter on all values.
+- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
+- ;; Or use one of several other reasonable alternatives.
+-
+- ;; No bindings, name: call a letrec'ed lambda.
+- ((let-loop name () (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ...) . body)))
+- name)
+- val ...))
+-
+- ;; Rest binding, name: call a letrec'ed lambda.
+- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ... . rest-var) . body)))
+- name)
+- val ... rest-val ...))))
+-</pre>
+-
+-
+-<H1>Copyright</H1>
+-
+-Copyright (C) Andy Gaynor (1999). All Rights Reserved.
+-<p>This document and translations of it may be copied and furnished to
+-others, and derivative works that comment on or otherwise explain it or
+-assist in its implementation may be prepared, copied, published and distributed,
+-in whole or in part, without restriction of any kind, provided that the
+-above copyright notice and this paragraph are included on all such copies
+-and derivative works. However, this document itself may not be modified
+-in any way, such as by removing the copyright notice or references to the
+-Scheme Request For Implementation process or editors, except as needed
+-for the purpose of developing SRFIs in which case the procedures for copyrights
+-defined in the SRFI process must be followed, or as required to translate
+-it into languages other than English.
+-<p>The limited permissions granted above are perpetual and will not be
+-revoked by the authors or their successors or assigns.
+-<p>This document and the information contained herein is provided on an
+-"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
+-EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
+-USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
+-WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
+-
+- <hr>
+- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
+-
+-</body>
+-</html>
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/info.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/info.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,14 +0,0 @@
+-#lang info
+-
+-(define collection 'multi)
+-
+-(define deps '("scheme-lib"
+- "base"
+- "r6rs-lib"
+- "srfi-lib"
+- "compatibility-lib"))
+-
+-
+-(define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"")
+-
+-(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-#lang s-exp srfi/provider srfi/%3a5
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang s-exp srfi/provider srfi/5
+-
+-;; FIXME: "rest arguments" need to generate mutable lists
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,86 +0,0 @@
+-;;;
+-;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
+-;;; Time-stamp: <03/04/08 09:56:06 solsona>
+-;;;
+-;;; Usually, I would add a copyright notice, and the announce that
+-;;; this code is under the LGPL licence. Nevertheless, I only did the
+-;;; port to PLT Scheme v200, and here is the copyright notice,
+-;;; comments, and licence from the original source:
+-;;;
+-;;; Copyright (C) Andy Gaynor (1999-2003)
+-;;;
+-;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
+-;;; Renamed to s:let by Eli Barzilay
+-
+-#lang scheme/base
+-(provide s:let)
+-
+-(define-syntax s:let
+- (syntax-rules ()
+- ;; standard
+- ((s:let () body ...)
+- (let () body ...))
+- ((s:let ((var val) ...) body ...)
+- (let ((var val) ...) body ...))
+-
+- ;; rest style
+- ((s:let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; signature style
+- ((s:let (name bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ;; standard named style
+- ((s:let name (bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ))
+-
+-;; A loop to walk down the list of bindings.
+-
+-(define-syntax let-loop
+- (syntax-rules ()
+-
+- ;; No more bindings - make a LETREC.
+- ((let-loop name () (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ...) . body)))
+- name)
+- vals ...))
+-
+- ;; Rest binding, no name
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (let ((var val) ... (rest-var (list rest-val ...))) . body))
+-
+- ;; Process a (var val) pair.
+- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
+- (let-loop name (more ...) (vars ... var) (vals ... val) body))
+-
+- ;; End with a rest variable - make a LETREC.
+- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ... . rest-var) . body)))
+- name)
+- vals ... rest-vals ...))))
+-
+-;; Four loops - normal and `signature-style', each with and without a rest
+-;; binding.
+-;;
+-;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let fibonacci ((n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+diff -urN racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
+--- racket-6.12-nf/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-01-31 13:04:07.236391659 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,2 +0,0 @@
+-;; module loader for SRFI-5
+-#lang s-exp srfi/provider srfi/5/let #:unprefix s:
diff --git a/racket-6.12-update-srfi.patch b/racket-6.12-update-srfi.patch
new file mode 100644
index 0000000..1894e32
--- /dev/null
+++ b/racket-6.12-update-srfi.patch
@@ -0,0 +1,4369 @@
+diff -urN racket-6.12-orig/share/links.rktd racket-6.12/share/links.rktd
+--- racket-6.12-orig/share/links.rktd 2018-01-26 16:12:13.000000000 -0500
++++ racket-6.12/share/links.rktd 2018-01-31 12:22:57.600504036 -0500
+@@ -133,6 +133,7 @@
+ (root "pkgs/r5rs-doc")
+ (root "pkgs/srfi-lite-lib")
+ (root "pkgs/srfi-lib")
++ (root "pkgs/srfi-lib-nonfree")
+ (root "pkgs/html-doc")
+ (root "pkgs/images-gui-lib")
+ (root "pkgs/images-doc")
+@@ -176,6 +177,7 @@
+ (root "pkgs/slideshow-exe")
+ (root "pkgs/slideshow-plugin")
+ (root "pkgs/srfi-doc")
++ (root "pkgs/srfi-doc-nonfree")
+ (root "pkgs/syntax-color-doc")
+ (root "pkgs/web-server-lib")
+ (root "pkgs/unix-socket-lib")
+diff -urN racket-6.12-orig/share/pkgs/srfi/info.rkt racket-6.12/share/pkgs/srfi/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi/info.rkt 2018-01-26 16:10:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi/info.rkt 2018-01-31 12:22:58.139502019 -0500
+@@ -1 +1,16 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("srfi-lib" "srfi-doc"))) (define implies (quote ("srfi-lib" "srfi-doc"))) (define pkg-desc "Legacy SRFI (Scheme) libraries") (define pkg-authors (quote (mflatt noel chongkai jay)))))
++#lang info
++
++(define collection 'multi)
++
++(define deps '("srfi-lib"
++ "srfi-lib-nonfree"
++ "srfi-doc"
++ "srfi-doc-nonfree"))
++(define implies '("srfi-lib"
++ "srfi-lib-nonfree"
++ "srfi-doc"
++ "srfi-foc-nonfree"))
++
++(define pkg-desc "Legacy SRFI (Scheme) libraries")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/info.rkt racket-6.12/share/pkgs/srfi-doc/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-doc/info.rkt 2018-01-26 16:10:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/info.rkt 2018-01-31 12:22:58.139502019 -0500
+@@ -1 +1,24 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define build-deps (quote ("mzscheme-doc" "scheme-lib" "base" "scribble-lib" "srfi-lib" "racket-doc" "r5rs-doc" "r6rs-doc" "compatibility-lib"))) (define update-implies (quote ("srfi-lib"))) (define pkg-desc "documentation part of \"srfi\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
++#lang info
++
++(define collection 'multi)
++
++(define build-deps '("mzscheme-doc"
++ "scheme-lib"
++ "base"
++ "scribble-lib"
++ "srfi-lib"
++ "racket-doc"
++ "r5rs-doc"
++ "r6rs-doc"
++ "compatibility-lib"))
++
++(define deps '("scheme-lib"
++ "base"
++ "scribble-lib"
++ "compatibility-lib"))
++
++(define update-implies '("srfi-lib"))
++
++(define pkg-desc "documentation part of \"srfi\"")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl
+--- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi.scrbl 2018-01-31 12:22:58.144502000 -0500
+@@ -1,5 +1,6 @@
+ #lang scribble/doc
+-@(require scribble/manual
++@(require srfi/scribblings/util
++ scribble/manual
+ scribble/eval
+ scriblib/render-cond
+ scribble/core
+@@ -8,54 +9,6 @@
+ (for-label scheme/base
+ racket/stream))
+
+-@(define-syntax (srfi stx)
+- (syntax-case stx ()
+- [(_ num #:subdir subdir? . title)
+- (with-syntax ([srfi/n (string->symbol (format "srfi/~a" (syntax-e #'num)))])
+- #'(begin
+- (section #:tag (format "srfi-~a" num)
+- #:style 'unnumbered
+- (format "SRFI ~a: " num)
+- . title)
+- (defmodule srfi/n)
+- "Original specification: "
+- (let* ([label (format "SRFI ~a" num)]
+- [sub (if subdir? (format "srfi-~a/" num) "")]
+- [url ( (b) (format "~a/srfi-std/~asrfi-~a.html" b sub num))])
+- (cond-element
+- [(or latex text) @link[(url "http://docs.racket-lang.org") label]]
+- [else @link[(url ".") label]]))))]
+- [(_ num . title) #'(srfi num #:subdir #f . title)]))
+-
+-@;{ The `lst' argument is a list of
+- (list sym syntactic-form? html-anchor) }
+-@(define (redirect n lst #:subdir [subdir? #f])
+- (let ([file (if subdir?
+- (format "srfi-~a/srfi-~a.html" n n)
+- (format "srfi-~a.html" n))]
+- [mod-path (string->symbol (format "srfi/~a" n))])
+- (make-binding-redirect-elements mod-path
+- (map (lambda (b)
+- (list (car b) (cadr b)
+- (build-path "srfi-std" file)
+- (caddr b)))
+- lst))))
+-
+-@(define in-core
+- (case-lambda
+- [() (in-core ".")]
+- [(k) @elem{This SRFI's bindings are also available in
+- @racketmodname[racket/base]@|k|}]))
+-
+-@(begin
+- (define-syntax-rule (def-mz mz-if)
+- (begin
+- (require (for-label mzscheme))
+- (define mz-if (racket if))))
+- (def-mz mz-if))
+-
+-@(define srfi-std (style #f (list (install-resource "srfi-std"))))
+-
+ @; ----------------------------------------------------------------------
+
+ @title{SRFIs: Libraries}
+@@ -245,14 +198,6 @@
+
+ @; ----------------------------------------
+
+-@srfi[5]{A compatible let form with signatures and rest arguments}
+-
+-@redirect[5 '(
+- (let #t "unnamed")
+-)]
+-
+-@; ----------------------------------------
+-
+ @srfi[6]{Basic String Ports}
+
+ @redirect[6 '(
+@@ -671,20 +616,6 @@
+
+ @; ----------------------------------------
+
+-@srfi[29]{Localization}
+-
+-@redirect[29 '(
+- (current-language #f "current-language")
+- (current-country #f "current-country")
+- (current-locale-details #f "current-locale-details")
+- (declare-bundle! #f "declare-bundle!")
+- (store-bundle #f "store-bundle")
+- (load-bundle! #f "load-bundle!")
+- (localized-template #f "localized-template")
+-)]
+-
+-@; ----------------------------------------
+-
+ @srfi[30]{Nested Multi-line Comments}
+
+ This SRFI's syntax is part of Racket's default reader.
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html
+--- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,507 +0,0 @@
+-<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
+-<html>
+- <head>
+- <meta name="generator" content="HTML Tidy, see www.w3.org">
+- <title>SRFI 29: Localization</title>
+- <meta name="author" content="Scott G. Miller">
+- <meta name="description" content="Localization">
+- </head>
+- <body>
+- <H1>Title</H1>
+-
+- SRFI 29: Localization
+-
+- <H1>Author</H1>
+-
+- Scott G. Miller
+-
+- <H1>Abstract</H1>
+-
+- This document specifies an interface to retrieving and
+- displaying locale sensitive messages. A Scheme program can
+- register one or more translations of templated messages, and
+- then write Scheme code that can transparently retrieve the
+- appropriate message for the locale under which the Scheme
+- system is running. <br>
+-
+-
+- <H1>Rationale</H1>
+-
+- <p>As any programmer that has ever had to deal with making his
+- or her code readable in more than one locale, the process of
+- sufficiently abstracting program messages from their
+- presentation to the user is non-trivial without help from the
+- programming language. Most modern programming language
+- libraries do provide some mechanism for performing this
+- separation.</p>
+-
+- <p>A portable API that allows a piece of code to run without
+- modification in different countries and under different
+- languages is a must for any non-trivial software project.
+- The interface should separate the logic of a program from
+- the myriad of translations that may be necessary.</p>
+-
+- <p>The interface described in this document provides such
+- functionality. The underlying implementation is also allowed to
+- use whatever datastructures it likes to provide access to the
+- translations in the most efficient manner possible. In
+- addition, the implementation is provided with standardized
+- functions that programs will use for accessing an external,
+- unspecified repository of translations.</p>
+-
+- <p>This interface <i>does not</i> cover all aspects of
+- localization, including support for non-latin characters,
+- number and date formatting, etc. Such functionality is the
+- scope of a future SRFI that may extend this one.</p>
+-
+- <H1>Dependencies</H1>
+-
+- An SRFI-29 conformant implementation must also implement
+- SRFI-28, Basic Format Strings. Message templates are strings
+- that must be processed by the <tt>format</tt> function
+- specified in that SRFI.
+-
+- <H1>Specification</H1>
+-
+- <h3>Message Bundles</h3>
+-
+- <p>A Message Bundle is a set of message templates and their
+- identifying keys. Each bundle contains one or more such
+- key/value pairs. The bundle itself is associated with a
+- <i>bundle specifier</i> which uniquely identifies the
+- bundle.</p>
+-
+- <h3>Bundle Specifiers</h3>
+-
+- <p>A Bundle Specifier is a Scheme list that describes, in order
+- of importance, the package and locale of a message bundle.
+- In most cases, a locale specifier will have between one
+- and three elements. The first element is a symbol denoting the
+- package for which this bundle applies. The second and third
+- elements denote a <i>locale</i>. The second element (first
+- element of the locale) if present, is the two letter, ISO 639-1
+- language code for the bundle. The third element, if present, is
+- a two letter ISO 3166-1 country code. In some cases, a
+- fourth element may be present, specifying the encoding used for
+- the bundle. All bundle specifier elements are Scheme
+- symbols.</p>
+-
+- <p>If only one translation is provided, it should be designated
+- only by a package name, for example <tt>(mathlib)</tt>. This
+- translation is called the <i>default</i> translation.</p>
+-
+- <h3>Bundle Searching</h3>
+-
+- <p>When a message template is retrieved from a bundle, the
+- Scheme implementation will provide the locale under which the
+- system is currently running. When the template is retrieved,
+- the package name will be specified. The Scheme system should
+- construct a Bundle Specifier from the provided package name and
+- the active locale. For example, when retrieving a message
+- template for French Canadian, in the <tt>mathlib</tt> package,
+- the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
+- program may also retrieve the elements of the current locale
+- using the no-argument procedures:</p>
+-
+- <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-language</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- When given no arguments, returns the current ISO 639-1
+- language code as a symbol. If provided with an
+- argument, the current language is set to that named by the
+- symbol for the currently executing Scheme thread (or for the
+- entire Scheme system if such a distinction is not possible).
+-
+- </blockquote>
+-
+- <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
+- <i>symbol</i></tt><br>
+- <tt><b>current-country</b> <i>symbol</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- returns the current ISO 3166-1 country code as a symbol.
+- If provided with an argument, the current country is
+- set to that named by the symbol for the currently executing
+- Scheme thread (or for the entire Scheme system if such a
+- distinction is not possible).
+- </blockquote>
+-
+- <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
+- symbol</i></tt>s<br>
+- <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
+- undefined</tt><br>
+- </p>
+-
+- <blockquote>
+- Returns a list of additional locale details as a list of
+- symbols. This list may contain information about
+- encodings or other more specific information. If
+- provided with an argument, the current locale details are set
+- to those given in the currently executing Scheme thread (or
+- for the entire Scheme system if such a distinction is not
+- possible).
+- </blockquote>
+-
+- <p>The Scheme System should first check for a bundle with the
+- exact name provided. If no such bundle is found, the last
+- element from the list is removed and a search is tried for a
+- bundle with that name. If no bundle is then found, the list is
+- shortened by removing the last element again. If no message is
+- found and the bundle specifier is now the empty list, an error
+- should be raised.</p>
+-
+- <p>The reason for this search order is to provide the most
+- locale sensitive template possible, but to fall back on more
+- general templates if a translation has not yet been provided
+- for the given locale.</p>
+-
+- <h3>Message Templates</h3>
+-
+- <p>A message template is a localized message that may or may
+- not contain one of a number of formatting codes. A message
+- template is a Scheme string. The string is of a form that can
+- be processed by the <tt>format</tt> procedure found in many
+- Scheme systems and formally specified in SRFI-28 (Basic Format
+- Strings).</p>
+-
+- <p>This SRFI also extends SRFI-28 to provide an additional
+- <tt>format</tt> escape code:</p>
+-
+- <blockquote>
+- <tt>~[n]@*</tt> - Causes a value-requiring escape code that
+- follows this code immediately to reference the [N]'th
+- optional value absolutely, rather than the next unconsumed
+- value. The referenced value is <i>not</i> consumed.
+- </blockquote>
+- This extension allows optional values to be positionally
+- referenced, so that message templates can be constructed that
+- can produce the proper word ordering for a language.
+-
+- <h3>Preparing Bundles</h3>
+- Before a bundle may be used by the Scheme system to retrieve
+- localized template messages, they must be made available to the
+- Scheme system. This SRFI specifies a way to portably
+- define the bundles, as well as store them in and retrieve them
+- from an unspecified system which may be provided by resources
+- outside the Scheme system.<br>
+-
+-
+- <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
+- association-list</i> -> undefined<br>
+- </tt></p>
+-
+- <blockquote>
+- Declares a new bundle named by the given bundle-specifier.
+- The contents of the bundle are defined by the provided
+- association list. The list contains associations
+- between Scheme symbols and the message templates (Scheme
+- strings) they name. If a bundle already exists with the
+- given name, it is overwritten with the newly declared
+- bundle.<br>
+- </blockquote>
+- <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to store a bundle named by the given bundle
+- specifier, and previously made available using
+- <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
+- unspecified mechanism that may be persistent across Scheme
+- system restarts. If successful, a non-false value is
+- returned. If unsuccessful, <tt>#f</tt> is returned.<br>
+- </blockquote>
+- <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
+- boolean</tt><br>
+-
+-
+- <blockquote>
+- Attempts to retrieve a bundle from an unspecified mechanism
+- which stores bundles outside the Scheme system. If the
+- bundle was retrieved successfully, the function returns a
+- non-false value, and the bundle is immediately available to
+- the Scheme system. If the bundle could not be found or loaded
+- successfully, the function returns <tt>#f</tt>, and the
+- Scheme system's bundle registry remains unaffected.<br>
+- </blockquote>
+- A compliant Scheme system may choose not to provide any
+- external mechanism to store localized bundles. If it does
+- not, it must still provide implementations for
+- <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
+- case, both functions must return <tt>#f</tt> regardless of the
+- arguments given. Users of this SRFI should recognize that the
+- inability to load or store a localized bundle in an external
+- repository is <i>not</i> a fatal error.<br>
+-
+-
+- <h3>Retrieving Localized Message Templates</h3>
+-
+- <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
+- message-template-name</tt></i> <tt>-> <i>string or #f<br>
+- </i></tt></p>
+-
+- <blockquote>
+- Retrieves a localized message template for the given package
+- name and the given message template name (both symbols).
+- If no such message could be found, false (#f) is
+- returned.<br>
+- <br>
+- </blockquote>
+- After retrieving a template, the calling program can use
+- <tt>format</tt> to produce a string that can be displayed to
+- the user.<br>
+-
+-
+- <h2>Examples</h2>
+- The below example makes use of SRFI-29 to display simple,
+- localized messages. It also defines its bundles in such a
+- way that the Scheme system may store and retrieve the bundles
+- from a more efficient system catalog, if available.<br>
+-
+-<pre>
+-(let ((translations
+- '(((en) . ((time . "Its ~a, ~a.")
+- (goodbye . "Goodbye, ~a.")))
+- ((fr) . ((time . "~1@*~a, c'est ~a.")
+- (goodbye . "Au revoir, ~a."))))))
+- (for-each (lambda (translation)
+- (let ((bundle-name (cons 'hello-program (car translation))))
+- (if (not (load-bundle! bundle-name))
+- (begin
+- (declare-bundle! bundle-name (cdr translation))
+- (store-bundle! bundle-name)))))
+- translations))
+-
+-(define localized-message
+- (lambda (message-name . args)
+- (apply format (cons (localized-template 'hello-program
+- message-name)
+- args))))
+-
+-(let ((myname "Fred"))
+- (display (localized-message 'time "12:00" myname))
+- (display #\newline)
+-
+- (display (localized-message 'goodbye myname))
+- (display #\newline))
+-
+-;; Displays (English):
+-;; Its 12:00, Fred.
+-;; Goodbye, Fred.
+-;;
+-;; French:
+-;; Fred, c'est 12:00.
+-;; Au revoir, Fred.
+-</pre>
+-
+- <H1>Implementation</H1>
+-
+- <p>The implementation requires that the Scheme system provide a
+- definition for <tt>current-language</tt> and
+- <tt>current-country</tt> capable of distinguishing the correct
+- locale present during a Scheme session. The definitions of
+- those functions in the reference implementation are not capable
+- of that distinction. Their implementation is provided only so
+- that the following code can run in any R4RS scheme system.
+- <br>
+- </p>
+-
+- <p>In addition, the below implementation of a compliant
+- <tt>format</tt> requires SRFI-6 (Basic String Ports) and
+- SRFI-23 (Error reporting)</p>
+-<pre>
+-;; The association list in which bundles will be stored
+-(define *localization-bundles* '())
+-
+-;; The current-language and current-country functions provided
+-;; here must be rewritten for each Scheme system to default to the
+-;; actual locale of the session
+-(define current-language
+- (let ((current-language-value 'en))
+- (lambda args
+- (if (null? args)
+- current-language-value
+- (set! current-language-value (car args))))))
+-
+-(define current-country
+- (let ((current-country-value 'us))
+- (lambda args
+- (if (null? args)
+- current-country-value
+- (set! current-country-value (car args))))))
+-
+-;; The load-bundle! and store-bundle! both return #f in this
+-;; reference implementation. A compliant implementation need
+-;; not rewrite these procedures.
+-(define load-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-(define store-bundle!
+- (lambda (bundle-specifier)
+- #f))
+-
+-;; Declare a bundle of templates with a given bundle specifier
+-(define declare-bundle!
+- (letrec ((remove-old-bundle
+- (lambda (specifier bundle)
+- (cond ((null? bundle) '())
+- ((equal? (caar bundle) specifier)
+- (cdr bundle))
+- (else (cons (car bundle)
+- (remove-old-bundle specifier
+- (cdr bundle))))))))
+- (lambda (bundle-specifier bundle-assoc-list)
+- (set! *localization-bundles*
+- (cons (cons bundle-specifier bundle-assoc-list)
+- (remove-old-bundle bundle-specifier
+- *localization-bundles*))))))
+-
+-;;Retrieve a localized template given its package name and a template name
+-(define localized-template
+- (letrec ((rdc
+- (lambda (ls)
+- (if (null? (cdr ls))
+- '()
+- (cons (car ls) (rdc (cdr ls))))))
+- (find-bundle
+- (lambda (specifier template-name)
+- (cond ((assoc specifier *localization-bundles*) =>
+- (lambda (bundle) bundle))
+- ((null? specifier) #f)
+- (else (find-bundle (rdc specifier)
+- template-name))))))
+- (lambda (package-name template-name)
+- (let loop ((specifier (cons package-name
+- (list (current-language)
+- (current-country)))))
+- (and (not (null? specifier))
+- (let ((bundle (find-bundle specifier template-name)))
+- (and bundle
+- (cond ((assq template-name bundle) => cdr)
+- ((null? (cdr specifier)) #f)
+- (else (loop (rdc specifier)))))))))))
+-
+-;;An SRFI-28 and SRFI-29 compliant version of format. It requires
+-;;SRFI-23 for error reporting.
+-(define format
+- (lambda (format-string . objects)
+- (let ((buffer (open-output-string)))
+- (let loop ((format-list (string->list format-string))
+- (objects objects)
+- (object-override #f))
+- (cond ((null? format-list) (get-output-string buffer))
+- ((char=? (car format-list) #\~)
+- (cond ((null? (cdr format-list))
+- (error 'format "Incomplete escape sequence"))
+- ((char-numeric? (cadr format-list))
+- (let posloop ((fl (cddr format-list))
+- (pos (string->number
+- (string (cadr format-list)))))
+- (cond ((null? fl)
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (null? (cdr fl)))
+- (error 'format "Incomplete escape sequence"))
+- ((and (eq? (car fl) '#\@)
+- (eq? (cadr fl) '#\*))
+- (loop (cddr fl) objects (list-ref objects pos)))
+- (else
+- (posloop (cdr fl)
+- (+ (* 10 pos)
+- (string->number
+- (string (car fl)))))))))
+- (else
+- (case (cadr format-list)
+- ((#\a)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (display (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\s)
+- (cond (object-override
+- (begin
+- (display object-override buffer)
+- (loop (cddr format-list) objects #f)))
+- ((null? objects)
+- (error 'format "No value for escape sequence"))
+- (else
+- (begin
+- (write (car objects) buffer)
+- (loop (cddr format-list)
+- (cdr objects) #f)))))
+- ((#\%)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\newline buffer)
+- (loop (cddr format-list) objects #f))
+- ((#\~)
+- (if object-override
+- (error 'format "Escape sequence following positional override does not require a value"))
+- (display #\~ buffer)
+- (loop (cddr format-list) objects #f))
+- (else
+- (error 'format "Unrecognized escape sequence"))))))
+- (else (display (car format-list) buffer)
+- (loop (cdr format-list) objects #f)))))))
+-
+-</pre>
+-
+- <H1>Copyright</H1>
+-
+- Copyright (C) Scott G. Miller (2002). All Rights Reserved.
+-
+- <p>This document and translations of it may be copied and
+- furnished to others, and derivative works that comment on or
+- otherwise explain it or assist in its implementation may be
+- prepared, copied, published and distributed, in whole or in
+- part, without restriction of any kind, provided that the above
+- copyright notice and this paragraph are included on all such
+- copies and derivative works. However, this document itself may
+- not be modified in any way, such as by removing the copyright
+- notice or references to the Scheme Request For Implementation
+- process or editors, except as needed for the purpose of
+- developing SRFIs in which case the procedures for copyrights
+- defined in the SRFI process must be followed, or as required to
+- translate it into languages other than English.</p>
+-
+- <p>The limited permissions granted above are perpetual and will
+- not be revoked by the authors or their successors or
+- assigns.</p>
+-
+- <p>This document and the information contained herein is
+- provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
+- EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
+- BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
+- HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
+- OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
+- <hr>
+-
+- <address>
+- Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
+- Rush</a>
+- </address>
+-
+- <address>
+- Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
+- Miller</a>
+- </address>
+- <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
+- <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
+- Daylight Time 2002 <!-- hhmts end --> <br>
+- </body>
+-</html>
+-
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html
+--- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
+@@ -1,345 +0,0 @@
+-<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
+-<html>
+-<head>
+- <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
+-</head>
+-<body>
+-
+-<H1>Title</H1>
+-
+-SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
+-
+-<H1>Author</H1>
+-
+-Andy Gaynor
+-
+-<H1>Status</H1>
+-
+-This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
+-You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
+-<P><UL>
+-<LI>Received: 1999/2/2
+-<LI>Draft: 1999/2/10-1999/04/12
+-<LI>Final: 1999/4/26
+-<LI>Revised reference implementation: 2003/01/27
+-</UL>
+-
+-<H1>Abstract</H1>
+-
+-The <i>named-let</i> incarnation of the <code>let</code> form has two slight
+-inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
+-form makes no accommodation for rest arguments, an issue of functionality
+-and consistency. As defined, the <code>let</code> form does not accommodate
+-signature-style syntax, an issue of aesthetics and consistency. Both
+-issues are addressed here in a manner which is compatible with the traditional
+-<code>let</code> form but for minor extensions.
+-
+-<H1>Rationale</H1>
+-
+-<H2>Signature-style Syntax</H2>
+-
+-Consider the following two equivalent definitions:
+-
+-<p><pre>
+-(define fibonacci
+- (lambda (n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1)))))
+-
+-(define (fibonacci n i f0 f1)
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-Although there is a named-let analog for the former form, there is none
+-for the latter. To wit, suppose one wished to compute the 10th element
+-of the Fibonacci sequence using a named let:
+-
+-<p>
+-<pre>
+-(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-Values: 55
+-</pre>
+-
+-As it stands, one cannot equivalently write
+-
+-<p>
+-<pre>
+-(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+- (if (= i n)
+- f0
+- (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-</pre>
+-
+-which is consistent with <code>define</code>'s signature-style form.
+-<p>Those that favor the signature style may prefer this extension.
+-In any case, it may be more appropriate to include all bound names within
+-the binding section. As presented, this straightforward extension
+-introduces no ambiguity or incompatibility with the existing definition
+-of let.
+-
+-<H2>Rest Arguments</H2>
+-
+-As it stands, one cannot write a named let with rest arguments, as in
+-
+-<p>
+-<pre>
+-(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))
+-</pre>
+-
+-otherwise equivalent to
+-
+-<p>
+-<pre>
+-(letrec ((blast (lambda (port . x)
+- (if (null? x)
+- 'just-a-silly-contrived-example
+- (begin
+- (write (car x) port)
+- (apply blast port (cdr x)))))))
+- (blast (current-output-port) (+ 1 2) 4 5))
+-</pre>
+-
+-While this example is rather contrived, the functionality is not.
+-There are several times when the author has used this construct in practice.
+-Regardless, there is little reason to deny the <code>let</code> form access to
+-all the features of lambda functionality.
+-
+-<H2>Symbols in Binding Sections</H2>
+-
+-Both the features above rely upon the placement of symbols in <code>let</code>
+-binding lists (this statement is intentially simplistic). The only
+-other apparent use of such symbol placement is to tersely bind variables
+-to unspecified values. For example, one might desire to use
+-<code>(let (foo bar baz) ...)</code>
+-to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
+-unspecified values.
+-
+-<p>This usage is considered less important in light of the rationales
+-presented above, and an alternate syntax is immediately apparent, as
+-in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
+-be preferable, consistently parenthesizing normal binding clauses.
+-
+-<H1>Specification</H1>
+-
+-<H2>Syntax</H2>
+-
+-<p>
+-A formal specification of the syntax follows. Below, body, expression,
+-and identifier are free. Each instantiation of binding-name must be
+-unique.
+-</p>
+-
+-<p>
+-<pre>
+- let = "(" "let" let-bindings body ")"
+- expressions = nothing | expression expressions
+- let-bindings = let-name bindings
+- |"(" let-name "." bindings ")"
+- let-name = identifier
+- bindings = "(" ")"
+- | rest-binding
+- | "(" normal-bindings ["." rest-binding] ")"
+-normal-bindings = nothing
+- | normal-binding normal-bindings
+-normal-binding = "(" binding-name expression ")"
+- binding-name = identifier
+- rest-binding = "(" binding-name expressions ")"
+-</pre>
+-
+-<p>
+-For clarity and convenience, an informal specification follows.
+-</p>
+-
+-<ol>
+-<li><a name="unnamed">Unnamed</a>
+-
+-<p><pre>
+-(let ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig">
+-Named, non-signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-sig">Named, signature-style, no rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...)
+- <body>...)
+-</pre>
+-</li>
+-
+-<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
+-
+-<p><pre>
+-(let <name> ((<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-
+-<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
+-
+-<p><pre>
+-(let (<name> (<parameter> <argument>)...
+-
+-. (<rest-parameter> <rest-argument>...))
+- <body>...)
+-</pre>
+-</li>
+-</ol>
+-
+-<H2>Semantics</H2>
+-
+-Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
+-and <code>letrec</code> forms, respectively.
+-
+-<ul>
+-<li>For informal syntax 1:
+-
+-<p><pre>
+-(($lambda (<parameter>...) <body>...) <argument>...)
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 2 and 3:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...) <body>...)))
+- (<name> <argument>...))
+-</pre>
+-</li>
+-
+-<li>For informal syntaxes 4 and 5:
+-
+-<p>
+-<pre>
+-($letrec ((<name> ($lambda (<parameter>...
+-
+-. <rest-parameter>) <body>...)))
+- (<name> <argument>... <rest-argument>...))
+-</pre>
+-</li>
+-</ul>
+-
+-<H1>Implementation</H1>
+-
+-Here is an implementation using <code>SYNTAX-RULES</code>.
+-
+-<p>
+-<pre>
+-;; Use your own standard let.
+-;; Or call a lambda.
+-;; (define-syntax standard-let
+-;;
+-;; (syntax-rules ()
+-;;
+-;; ((let ((var val) ...) body ...)
+-;; ((lambda (var ...) body ...) val ...))))
+-
+-(define-syntax let
+-
+- (syntax-rules ()
+-
+- ;; No bindings: use standard-let.
+- ((let () body ...)
+- (standard-let () body ...))
+- ;; Or call a lambda.
+- ;; ((lambda () body ...))
+-
+- ;; All standard bindings: use standard-let.
+- ((let ((var val) ...) body ...)
+- (standard-let ((var val) ...) body ...))
+- ;; Or call a lambda.
+- ;; ((lambda (var ...) body ...) val ...)
+-
+- ;; One standard binding: loop.
+- ;; The all-standard-bindings clause didn't match,
+- ;; so there must be a rest binding.
+- ((let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; Signature-style name: loop.
+- ((let (name binding ...) body ...)
+- (let-loop name (binding ...) () () (body ...)))
+-
+- ;; defun-style name: loop.
+- ((let name bindings body ...)
+- (let-loop name bindings () () (body ...)))))
+-
+-(define-syntax let-loop
+-
+- (syntax-rules ()
+-
+- ;; Standard binding: destructure and loop.
+- ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
+- (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
+-
+- ;; Rest binding, no name: use standard-let, listing the rest values.
+- ;; Because of let's first clause, there is no "no bindings, no name" clause.
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
+- ;; Or call a lambda with a rest parameter on all values.
+- ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
+- ;; Or use one of several other reasonable alternatives.
+-
+- ;; No bindings, name: call a letrec'ed lambda.
+- ((let-loop name () (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ...) . body)))
+- name)
+- val ...))
+-
+- ;; Rest binding, name: call a letrec'ed lambda.
+- ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
+- ((letrec ((name (lambda (var ... . rest-var) . body)))
+- name)
+- val ... rest-val ...))))
+-</pre>
+-
+-
+-<H1>Copyright</H1>
+-
+-Copyright (C) Andy Gaynor (1999). All Rights Reserved.
+-<p>This document and translations of it may be copied and furnished to
+-others, and derivative works that comment on or otherwise explain it or
+-assist in its implementation may be prepared, copied, published and distributed,
+-in whole or in part, without restriction of any kind, provided that the
+-above copyright notice and this paragraph are included on all such copies
+-and derivative works. However, this document itself may not be modified
+-in any way, such as by removing the copyright notice or references to the
+-Scheme Request For Implementation process or editors, except as needed
+-for the purpose of developing SRFIs in which case the procedures for copyrights
+-defined in the SRFI process must be followed, or as required to translate
+-it into languages other than English.
+-<p>The limited permissions granted above are perpetual and will not be
+-revoked by the authors or their successors or assigns.
+-<p>This document and the information contained herein is provided on an
+-"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
+-EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
+-USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
+-WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
+-
+- <hr>
+- <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
+-
+-</body>
+-</html>
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/util.rkt racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/util.rkt
+--- racket-6.12-orig/share/pkgs/srfi-doc/srfi/scribblings/util.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc/srfi/scribblings/util.rkt 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,61 @@
++#lang scribble/doc
++
++@(require scribble/manual
++ scribble/eval
++ scriblib/render-cond
++ scribble/core
++ scribble/html-properties
++ (for-syntax scheme/base)
++ (for-label scheme/base
++ racket/stream))
++
++@(provide (all-defined-out))
++
++@(define-syntax (srfi stx)
++ (syntax-case stx ()
++ [(_ num #:subdir subdir? . title)
++ (with-syntax ([srfi/n (string->symbol (format "srfi/~a" (syntax-e #'num)))])
++ #'(begin
++ (section #:tag (format "srfi-~a" num)
++ #:style 'unnumbered
++ (format "SRFI ~a: " num)
++ . title)
++ (defmodule srfi/n)
++ "Original specification: "
++ (let* ([label (format "SRFI ~a" num)]
++ [sub (if subdir? (format "srfi-~a/" num) "")]
++ [url ( (b) (format "~a/srfi-std/~asrfi-~a.html" b sub num))])
++ (cond-element
++ [(or latex text) @link[(url "http://docs.racket-lang.org") label]]
++ [else @link[(url ".") label]]))))]
++ [(_ num . title) #'(srfi num #:subdir #f . title)]))
++
++@;{ The `lst' argument is a list of
++ (list sym syntactic-form? html-anchor) }
++@(define (redirect n lst #:subdir [subdir? #f])
++ (let ([file (if subdir?
++ (format "srfi-~a/srfi-~a.html" n n)
++ (format "srfi-~a.html" n))]
++ [mod-path (string->symbol (format "srfi/~a" n))])
++ (make-binding-redirect-elements mod-path
++ (map (lambda (b)
++ (list (car b) (cadr b)
++ (build-path "srfi-std" file)
++ (caddr b)))
++ lst))))
++
++@(define in-core
++ (case-lambda
++ [() (in-core ".")]
++ [(k) @elem{This SRFI's bindings are also available in
++ @racketmodname[racket/base]@|k|}]))
++
++@(begin
++ (define-syntax-rule (def-mz mz-if)
++ (begin
++ (require (for-label mzscheme))
++ (define mz-if (racket if))))
++ (def-mz mz-if))
++
++@(define srfi-std (style #f (list (install-resource "srfi-std"))))
++
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/info.rkt 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,19 @@
++#lang info
++
++(define collection 'multi)
++
++(define build-deps '("mzscheme-doc"
++ "scheme-lib"
++ "base"
++ "scribble-lib"
++ "srfi-doc"
++ "srfi-lib-nonfree"
++ "racket-doc"
++ "r5rs-doc"
++ "r6rs-doc"
++ "compatibility-lib"))
++(define update-implies '("srfi-lib-nonfree"))
++
++(define pkg-desc "documentation part of \"srfi nonfree\"")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/info.rkt 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,3 @@
++#lang info
++
++(define scribblings '(("srfi-nf.scrbl" (multi-page) (library 100))))
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-nf.scrbl 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,66 @@
++#lang scribble/doc
++@(require srfi/scribblings/util
++ scribble/manual
++ scribble/eval
++ scriblib/render-cond
++ scribble/core
++ scribble/html-properties
++ (for-syntax scheme/base)
++ (for-label scheme/base
++ racket/stream))
++
++@; ----------------------------------------------------------------------
++
++@title{SRFI Nonfree Libraries and Documentation}
++
++The @link[#:style srfi-std "http://srfi.schemers.org/"]{Scheme Requests for
++Implementation} (a.k.a. @deftech{SRFI}) process allows individual
++members of the Scheme community to propose libraries and extensions to
++be supported by multiple Scheme implementations.
++
++Racket is distributed with implementations of many SRFIs, most of
++which can be implemented as libraries. To import the bindings of SRFI
++@math{n}, use
++
++@racketblock[
++(require @#,elem{@racketidfont{srfi/}@math{n}})
++]
++
++This document lists the SRFIs that are supported by Racket and
++provides a link to the original SRFI specification (which is also
++distributed as part of Racket's documentation).
++
++The following SRFI specification documents are licensed restrictively.
++
++@table-of-contents[]
++
++
++@; ----------------------------------------
++
++@srfi[5]{A compatible let form with signatures and rest arguments}
++
++@redirect[5 '(
++ (let #t "unnamed")
++)]
++
++Racket provides this SRFI in the @racket[srfi-lib-nonfree] package.
++
++@; ----------------------------------------
++
++@srfi[29]{Localization}
++
++@redirect[29 '(
++ (current-language #f "current-language")
++ (current-country #f "current-country")
++ (current-locale-details #f "current-locale-details")
++ (declare-bundle! #f "declare-bundle!")
++ (store-bundle #f "store-bundle")
++ (load-bundle! #f "load-bundle!")
++ (localized-template #f "localized-template")
++)]
++
++Racket provides a free implementation of this SRFI in the @racket[srfi-lib] package. Only the SRFI specification document is nonfree.
++
++@; ----------------------------------------
++
++@index-section[]
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-29.html 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,507 @@
++<!DOCTYPE html PUBLIC "-//IETF//DTD HTML//EN">
++<html>
++ <head>
++ <meta name="generator" content="HTML Tidy, see www.w3.org">
++ <title>SRFI 29: Localization</title>
++ <meta name="author" content="Scott G. Miller">
++ <meta name="description" content="Localization">
++ </head>
++ <body>
++ <H1>Title</H1>
++
++ SRFI 29: Localization
++
++ <H1>Author</H1>
++
++ Scott G. Miller
++
++ <H1>Abstract</H1>
++
++ This document specifies an interface to retrieving and
++ displaying locale sensitive messages. A Scheme program can
++ register one or more translations of templated messages, and
++ then write Scheme code that can transparently retrieve the
++ appropriate message for the locale under which the Scheme
++ system is running. <br>
++
++
++ <H1>Rationale</H1>
++
++ <p>As any programmer that has ever had to deal with making his
++ or her code readable in more than one locale, the process of
++ sufficiently abstracting program messages from their
++ presentation to the user is non-trivial without help from the
++ programming language. Most modern programming language
++ libraries do provide some mechanism for performing this
++ separation.</p>
++
++ <p>A portable API that allows a piece of code to run without
++ modification in different countries and under different
++ languages is a must for any non-trivial software project.
++ The interface should separate the logic of a program from
++ the myriad of translations that may be necessary.</p>
++
++ <p>The interface described in this document provides such
++ functionality. The underlying implementation is also allowed to
++ use whatever datastructures it likes to provide access to the
++ translations in the most efficient manner possible. In
++ addition, the implementation is provided with standardized
++ functions that programs will use for accessing an external,
++ unspecified repository of translations.</p>
++
++ <p>This interface <i>does not</i> cover all aspects of
++ localization, including support for non-latin characters,
++ number and date formatting, etc. Such functionality is the
++ scope of a future SRFI that may extend this one.</p>
++
++ <H1>Dependencies</H1>
++
++ An SRFI-29 conformant implementation must also implement
++ SRFI-28, Basic Format Strings. Message templates are strings
++ that must be processed by the <tt>format</tt> function
++ specified in that SRFI.
++
++ <H1>Specification</H1>
++
++ <h3>Message Bundles</h3>
++
++ <p>A Message Bundle is a set of message templates and their
++ identifying keys. Each bundle contains one or more such
++ key/value pairs. The bundle itself is associated with a
++ <i>bundle specifier</i> which uniquely identifies the
++ bundle.</p>
++
++ <h3>Bundle Specifiers</h3>
++
++ <p>A Bundle Specifier is a Scheme list that describes, in order
++ of importance, the package and locale of a message bundle.
++ In most cases, a locale specifier will have between one
++ and three elements. The first element is a symbol denoting the
++ package for which this bundle applies. The second and third
++ elements denote a <i>locale</i>. The second element (first
++ element of the locale) if present, is the two letter, ISO 639-1
++ language code for the bundle. The third element, if present, is
++ a two letter ISO 3166-1 country code. In some cases, a
++ fourth element may be present, specifying the encoding used for
++ the bundle. All bundle specifier elements are Scheme
++ symbols.</p>
++
++ <p>If only one translation is provided, it should be designated
++ only by a package name, for example <tt>(mathlib)</tt>. This
++ translation is called the <i>default</i> translation.</p>
++
++ <h3>Bundle Searching</h3>
++
++ <p>When a message template is retrieved from a bundle, the
++ Scheme implementation will provide the locale under which the
++ system is currently running. When the template is retrieved,
++ the package name will be specified. The Scheme system should
++ construct a Bundle Specifier from the provided package name and
++ the active locale. For example, when retrieving a message
++ template for French Canadian, in the <tt>mathlib</tt> package,
++ the bundle specifier '<tt>(mathlib fr ca)</tt>' is used. A
++ program may also retrieve the elements of the current locale
++ using the no-argument procedures:</p>
++
++ <p><b><a name="current-language"></a><tt>current-language</tt></b> <tt>->
++ <i>symbol</i></tt><br>
++ <tt><b>current-language</b> <i>symbol</i> ->
++ undefined</tt><br>
++ </p>
++
++ <blockquote>
++ When given no arguments, returns the current ISO 639-1
++ language code as a symbol. If provided with an
++ argument, the current language is set to that named by the
++ symbol for the currently executing Scheme thread (or for the
++ entire Scheme system if such a distinction is not possible).
++
++ </blockquote>
++
++ <p><b><a name="current-country"></a><tt>current-country</tt></b> <tt>->
++ <i>symbol</i></tt><br>
++ <tt><b>current-country</b> <i>symbol</i> ->
++ undefined</tt><br>
++ </p>
++
++ <blockquote>
++ returns the current ISO 3166-1 country code as a symbol.
++ If provided with an argument, the current country is
++ set to that named by the symbol for the currently executing
++ Scheme thread (or for the entire Scheme system if such a
++ distinction is not possible).
++ </blockquote>
++
++ <p><b><a name="current-locale-details"></a><tt>current-locale-details</tt></b> <tt>-> <i>list of
++ symbol</i></tt>s<br>
++ <tt><b>current-locale-details</b> <i>list-of-symbols</i> ->
++ undefined</tt><br>
++ </p>
++
++ <blockquote>
++ Returns a list of additional locale details as a list of
++ symbols. This list may contain information about
++ encodings or other more specific information. If
++ provided with an argument, the current locale details are set
++ to those given in the currently executing Scheme thread (or
++ for the entire Scheme system if such a distinction is not
++ possible).
++ </blockquote>
++
++ <p>The Scheme System should first check for a bundle with the
++ exact name provided. If no such bundle is found, the last
++ element from the list is removed and a search is tried for a
++ bundle with that name. If no bundle is then found, the list is
++ shortened by removing the last element again. If no message is
++ found and the bundle specifier is now the empty list, an error
++ should be raised.</p>
++
++ <p>The reason for this search order is to provide the most
++ locale sensitive template possible, but to fall back on more
++ general templates if a translation has not yet been provided
++ for the given locale.</p>
++
++ <h3>Message Templates</h3>
++
++ <p>A message template is a localized message that may or may
++ not contain one of a number of formatting codes. A message
++ template is a Scheme string. The string is of a form that can
++ be processed by the <tt>format</tt> procedure found in many
++ Scheme systems and formally specified in SRFI-28 (Basic Format
++ Strings).</p>
++
++ <p>This SRFI also extends SRFI-28 to provide an additional
++ <tt>format</tt> escape code:</p>
++
++ <blockquote>
++ <tt>~[n]@*</tt> - Causes a value-requiring escape code that
++ follows this code immediately to reference the [N]'th
++ optional value absolutely, rather than the next unconsumed
++ value. The referenced value is <i>not</i> consumed.
++ </blockquote>
++ This extension allows optional values to be positionally
++ referenced, so that message templates can be constructed that
++ can produce the proper word ordering for a language.
++
++ <h3>Preparing Bundles</h3>
++ Before a bundle may be used by the Scheme system to retrieve
++ localized template messages, they must be made available to the
++ Scheme system. This SRFI specifies a way to portably
++ define the bundles, as well as store them in and retrieve them
++ from an unspecified system which may be provided by resources
++ outside the Scheme system.<br>
++
++
++ <p><b><a name="declare-bundle!"></a><tt>declare-bundle!</tt></b> <tt><i>bundle-specifier
++ association-list</i> -> undefined<br>
++ </tt></p>
++
++ <blockquote>
++ Declares a new bundle named by the given bundle-specifier.
++ The contents of the bundle are defined by the provided
++ association list. The list contains associations
++ between Scheme symbols and the message templates (Scheme
++ strings) they name. If a bundle already exists with the
++ given name, it is overwritten with the newly declared
++ bundle.<br>
++ </blockquote>
++ <tt><a name="store-bundle"></a><b>store-bundle</b> <i>bundle-specifier</i> ->
++ boolean</tt><br>
++
++
++ <blockquote>
++ Attempts to store a bundle named by the given bundle
++ specifier, and previously made available using
++ <tt>declare-bundle!</tt> or <tt>load-bundle!</tt>, in an
++ unspecified mechanism that may be persistent across Scheme
++ system restarts. If successful, a non-false value is
++ returned. If unsuccessful, <tt>#f</tt> is returned.<br>
++ </blockquote>
++ <tt><a name="load-bundle!"></a><b>load-bundle!</b> <i>bundle-specifier</i> ->
++ boolean</tt><br>
++
++
++ <blockquote>
++ Attempts to retrieve a bundle from an unspecified mechanism
++ which stores bundles outside the Scheme system. If the
++ bundle was retrieved successfully, the function returns a
++ non-false value, and the bundle is immediately available to
++ the Scheme system. If the bundle could not be found or loaded
++ successfully, the function returns <tt>#f</tt>, and the
++ Scheme system's bundle registry remains unaffected.<br>
++ </blockquote>
++ A compliant Scheme system may choose not to provide any
++ external mechanism to store localized bundles. If it does
++ not, it must still provide implementations for
++ <tt>store-bundle</tt> and <tt>load-bundle!</tt>. In such a
++ case, both functions must return <tt>#f</tt> regardless of the
++ arguments given. Users of this SRFI should recognize that the
++ inability to load or store a localized bundle in an external
++ repository is <i>not</i> a fatal error.<br>
++
++
++ <h3>Retrieving Localized Message Templates</h3>
++
++ <p><a name="localized-template"></a><b><tt>localized-template</tt></b> <i><tt>package-name
++ message-template-name</tt></i> <tt>-> <i>string or #f<br>
++ </i></tt></p>
++
++ <blockquote>
++ Retrieves a localized message template for the given package
++ name and the given message template name (both symbols).
++ If no such message could be found, false (#f) is
++ returned.<br>
++ <br>
++ </blockquote>
++ After retrieving a template, the calling program can use
++ <tt>format</tt> to produce a string that can be displayed to
++ the user.<br>
++
++
++ <h2>Examples</h2>
++ The below example makes use of SRFI-29 to display simple,
++ localized messages. It also defines its bundles in such a
++ way that the Scheme system may store and retrieve the bundles
++ from a more efficient system catalog, if available.<br>
++
++<pre>
++(let ((translations
++ '(((en) . ((time . "Its ~a, ~a.")
++ (goodbye . "Goodbye, ~a.")))
++ ((fr) . ((time . "~1@*~a, c'est ~a.")
++ (goodbye . "Au revoir, ~a."))))))
++ (for-each (lambda (translation)
++ (let ((bundle-name (cons 'hello-program (car translation))))
++ (if (not (load-bundle! bundle-name))
++ (begin
++ (declare-bundle! bundle-name (cdr translation))
++ (store-bundle! bundle-name)))))
++ translations))
++
++(define localized-message
++ (lambda (message-name . args)
++ (apply format (cons (localized-template 'hello-program
++ message-name)
++ args))))
++
++(let ((myname "Fred"))
++ (display (localized-message 'time "12:00" myname))
++ (display #\newline)
++
++ (display (localized-message 'goodbye myname))
++ (display #\newline))
++
++;; Displays (English):
++;; Its 12:00, Fred.
++;; Goodbye, Fred.
++;;
++;; French:
++;; Fred, c'est 12:00.
++;; Au revoir, Fred.
++</pre>
++
++ <H1>Implementation</H1>
++
++ <p>The implementation requires that the Scheme system provide a
++ definition for <tt>current-language</tt> and
++ <tt>current-country</tt> capable of distinguishing the correct
++ locale present during a Scheme session. The definitions of
++ those functions in the reference implementation are not capable
++ of that distinction. Their implementation is provided only so
++ that the following code can run in any R4RS scheme system.
++ <br>
++ </p>
++
++ <p>In addition, the below implementation of a compliant
++ <tt>format</tt> requires SRFI-6 (Basic String Ports) and
++ SRFI-23 (Error reporting)</p>
++<pre>
++;; The association list in which bundles will be stored
++(define *localization-bundles* '())
++
++;; The current-language and current-country functions provided
++;; here must be rewritten for each Scheme system to default to the
++;; actual locale of the session
++(define current-language
++ (let ((current-language-value 'en))
++ (lambda args
++ (if (null? args)
++ current-language-value
++ (set! current-language-value (car args))))))
++
++(define current-country
++ (let ((current-country-value 'us))
++ (lambda args
++ (if (null? args)
++ current-country-value
++ (set! current-country-value (car args))))))
++
++;; The load-bundle! and store-bundle! both return #f in this
++;; reference implementation. A compliant implementation need
++;; not rewrite these procedures.
++(define load-bundle!
++ (lambda (bundle-specifier)
++ #f))
++
++(define store-bundle!
++ (lambda (bundle-specifier)
++ #f))
++
++;; Declare a bundle of templates with a given bundle specifier
++(define declare-bundle!
++ (letrec ((remove-old-bundle
++ (lambda (specifier bundle)
++ (cond ((null? bundle) '())
++ ((equal? (caar bundle) specifier)
++ (cdr bundle))
++ (else (cons (car bundle)
++ (remove-old-bundle specifier
++ (cdr bundle))))))))
++ (lambda (bundle-specifier bundle-assoc-list)
++ (set! *localization-bundles*
++ (cons (cons bundle-specifier bundle-assoc-list)
++ (remove-old-bundle bundle-specifier
++ *localization-bundles*))))))
++
++;;Retrieve a localized template given its package name and a template name
++(define localized-template
++ (letrec ((rdc
++ (lambda (ls)
++ (if (null? (cdr ls))
++ '()
++ (cons (car ls) (rdc (cdr ls))))))
++ (find-bundle
++ (lambda (specifier template-name)
++ (cond ((assoc specifier *localization-bundles*) =>
++ (lambda (bundle) bundle))
++ ((null? specifier) #f)
++ (else (find-bundle (rdc specifier)
++ template-name))))))
++ (lambda (package-name template-name)
++ (let loop ((specifier (cons package-name
++ (list (current-language)
++ (current-country)))))
++ (and (not (null? specifier))
++ (let ((bundle (find-bundle specifier template-name)))
++ (and bundle
++ (cond ((assq template-name bundle) => cdr)
++ ((null? (cdr specifier)) #f)
++ (else (loop (rdc specifier)))))))))))
++
++;;An SRFI-28 and SRFI-29 compliant version of format. It requires
++;;SRFI-23 for error reporting.
++(define format
++ (lambda (format-string . objects)
++ (let ((buffer (open-output-string)))
++ (let loop ((format-list (string->list format-string))
++ (objects objects)
++ (object-override #f))
++ (cond ((null? format-list) (get-output-string buffer))
++ ((char=? (car format-list) #\~)
++ (cond ((null? (cdr format-list))
++ (error 'format "Incomplete escape sequence"))
++ ((char-numeric? (cadr format-list))
++ (let posloop ((fl (cddr format-list))
++ (pos (string->number
++ (string (cadr format-list)))))
++ (cond ((null? fl)
++ (error 'format "Incomplete escape sequence"))
++ ((and (eq? (car fl) '#\@)
++ (null? (cdr fl)))
++ (error 'format "Incomplete escape sequence"))
++ ((and (eq? (car fl) '#\@)
++ (eq? (cadr fl) '#\*))
++ (loop (cddr fl) objects (list-ref objects pos)))
++ (else
++ (posloop (cdr fl)
++ (+ (* 10 pos)
++ (string->number
++ (string (car fl)))))))))
++ (else
++ (case (cadr format-list)
++ ((#\a)
++ (cond (object-override
++ (begin
++ (display object-override buffer)
++ (loop (cddr format-list) objects #f)))
++ ((null? objects)
++ (error 'format "No value for escape sequence"))
++ (else
++ (begin
++ (display (car objects) buffer)
++ (loop (cddr format-list)
++ (cdr objects) #f)))))
++ ((#\s)
++ (cond (object-override
++ (begin
++ (display object-override buffer)
++ (loop (cddr format-list) objects #f)))
++ ((null? objects)
++ (error 'format "No value for escape sequence"))
++ (else
++ (begin
++ (write (car objects) buffer)
++ (loop (cddr format-list)
++ (cdr objects) #f)))))
++ ((#\%)
++ (if object-override
++ (error 'format "Escape sequence following positional override does not require a value"))
++ (display #\newline buffer)
++ (loop (cddr format-list) objects #f))
++ ((#\~)
++ (if object-override
++ (error 'format "Escape sequence following positional override does not require a value"))
++ (display #\~ buffer)
++ (loop (cddr format-list) objects #f))
++ (else
++ (error 'format "Unrecognized escape sequence"))))))
++ (else (display (car format-list) buffer)
++ (loop (cdr format-list) objects #f)))))))
++
++</pre>
++
++ <H1>Copyright</H1>
++
++ Copyright (C) Scott G. Miller (2002). All Rights Reserved.
++
++ <p>This document and translations of it may be copied and
++ furnished to others, and derivative works that comment on or
++ otherwise explain it or assist in its implementation may be
++ prepared, copied, published and distributed, in whole or in
++ part, without restriction of any kind, provided that the above
++ copyright notice and this paragraph are included on all such
++ copies and derivative works. However, this document itself may
++ not be modified in any way, such as by removing the copyright
++ notice or references to the Scheme Request For Implementation
++ process or editors, except as needed for the purpose of
++ developing SRFIs in which case the procedures for copyrights
++ defined in the SRFI process must be followed, or as required to
++ translate it into languages other than English.</p>
++
++ <p>The limited permissions granted above are perpetual and will
++ not be revoked by the authors or their successors or
++ assigns.</p>
++
++ <p>This document and the information contained herein is
++ provided on an "AS IS" basis and THE AUTHOR AND THE SRFI
++ EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING
++ BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION
++ HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES
++ OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.</p>
++ <hr>
++
++ <address>
++ Editor: <a href="mailto:srfi-editors@srfi.schemers.org">David
++ Rush</a>
++ </address>
++
++ <address>
++ Author: <a href="mailto:scgmille@freenetproject.org">Scott G.
++ Miller</a>
++ </address>
++ <!-- Created: Tue Sep 29 19:20:08 EDT 1998 -->
++ <!-- hhmts start -->Last modified: Mon Jun 17 12:00:08 Pacific
++ Daylight Time 2002 <!-- hhmts end --> <br>
++ </body>
++</html>
++
+diff -urN racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html
+--- racket-6.12-orig/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-doc-nonfree/srfi/scribblings/srfi-std/srfi-5.html 2018-01-31 12:22:58.144502000 -0500
+@@ -0,0 +1,345 @@
++<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 3.2//EN">
++<html>
++<head>
++ <title>SRFI 5: A compatible let form with signatures and rest arguments</title>
++</head>
++<body>
++
++<H1>Title</H1>
++
++SRFI-5: A compatible <code>let</code> form with signatures and rest arguments
++
++<H1>Author</H1>
++
++Andy Gaynor
++
++<H1>Status</H1>
++
++This SRFI is currently in ``final'' status. To see an explanation of each status that a SRFI can hold, see <A HREF="http://srfi.schemers.org/srfi-process.html">here</A>.
++You can access the discussion on this SRFI via <A HREF="http://srfi.schemers.org/srfi-5/mail-archive/maillist.html">the archive of the mailing list</A>.
++<P><UL>
++<LI>Received: 1999/2/2
++<LI>Draft: 1999/2/10-1999/04/12
++<LI>Final: 1999/4/26
++<LI>Revised reference implementation: 2003/01/27
++</UL>
++
++<H1>Abstract</H1>
++
++The <i>named-let</i> incarnation of the <code>let</code> form has two slight
++inconsistencies with the <code>define</code> form. As defined, the <code>let</code>
++form makes no accommodation for rest arguments, an issue of functionality
++and consistency. As defined, the <code>let</code> form does not accommodate
++signature-style syntax, an issue of aesthetics and consistency. Both
++issues are addressed here in a manner which is compatible with the traditional
++<code>let</code> form but for minor extensions.
++
++<H1>Rationale</H1>
++
++<H2>Signature-style Syntax</H2>
++
++Consider the following two equivalent definitions:
++
++<p><pre>
++(define fibonacci
++ (lambda (n i f0 f1)
++ (if (= i n)
++ f0
++ (fibonacci n (+ i 1) f1 (+ f0 f1)))))
++
++(define (fibonacci n i f0 f1)
++ (if (= i n)
++ f0
++ (fibonacci n (+ i 1) f1 (+ f0 f1))))
++</pre>
++
++Although there is a named-let analog for the former form, there is none
++for the latter. To wit, suppose one wished to compute the 10th element
++of the Fibonacci sequence using a named let:
++
++<p>
++<pre>
++(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
++ (if (= i n)
++ f0
++ (fibonacci n (+ i 1) f1 (+ f0 f1))))
++Values: 55
++</pre>
++
++As it stands, one cannot equivalently write
++
++<p>
++<pre>
++(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
++ (if (= i n)
++ f0
++ (fibonacci n (+ i 1) f1 (+ f0 f1))))
++</pre>
++
++which is consistent with <code>define</code>'s signature-style form.
++<p>Those that favor the signature style may prefer this extension.
++In any case, it may be more appropriate to include all bound names within
++the binding section. As presented, this straightforward extension
++introduces no ambiguity or incompatibility with the existing definition
++of let.
++
++<H2>Rest Arguments</H2>
++
++As it stands, one cannot write a named let with rest arguments, as in
++
++<p>
++<pre>
++(let (blast (port (current-output-port)) . (x (+ 1 2) 4 5))
++ (if (null? x)
++ 'just-a-silly-contrived-example
++ (begin
++ (write (car x) port)
++ (apply blast port (cdr x)))))
++</pre>
++
++otherwise equivalent to
++
++<p>
++<pre>
++(letrec ((blast (lambda (port . x)
++ (if (null? x)
++ 'just-a-silly-contrived-example
++ (begin
++ (write (car x) port)
++ (apply blast port (cdr x)))))))
++ (blast (current-output-port) (+ 1 2) 4 5))
++</pre>
++
++While this example is rather contrived, the functionality is not.
++There are several times when the author has used this construct in practice.
++Regardless, there is little reason to deny the <code>let</code> form access to
++all the features of lambda functionality.
++
++<H2>Symbols in Binding Sections</H2>
++
++Both the features above rely upon the placement of symbols in <code>let</code>
++binding lists (this statement is intentially simplistic). The only
++other apparent use of such symbol placement is to tersely bind variables
++to unspecified values. For example, one might desire to use
++<code>(let (foo bar baz) ...)</code>
++to bind <code>foo</code>, <code>bar</code>, and <code>baz</code> to
++unspecified values.
++
++<p>This usage is considered less important in light of the rationales
++presented above, and an alternate syntax is immediately apparent, as
++in <code>(let ((foo) (bar) (baz)) ...)</code> This may even
++be preferable, consistently parenthesizing normal binding clauses.
++
++<H1>Specification</H1>
++
++<H2>Syntax</H2>
++
++<p>
++A formal specification of the syntax follows. Below, body, expression,
++and identifier are free. Each instantiation of binding-name must be
++unique.
++</p>
++
++<p>
++<pre>
++ let = "(" "let" let-bindings body ")"
++ expressions = nothing | expression expressions
++ let-bindings = let-name bindings
++ |"(" let-name "." bindings ")"
++ let-name = identifier
++ bindings = "(" ")"
++ | rest-binding
++ | "(" normal-bindings ["." rest-binding] ")"
++normal-bindings = nothing
++ | normal-binding normal-bindings
++normal-binding = "(" binding-name expression ")"
++ binding-name = identifier
++ rest-binding = "(" binding-name expressions ")"
++</pre>
++
++<p>
++For clarity and convenience, an informal specification follows.
++</p>
++
++<ol>
++<li><a name="unnamed">Unnamed</a>
++
++<p><pre>
++(let ((<parameter> <argument>)...)
++ <body>...)
++</pre>
++</li>
++
++<li><a name="named-non-sig">
++Named, non-signature-style, no rest argument</a>
++
++<p><pre>
++(let <name> ((<parameter> <argument>)...)
++ <body>...)
++</pre>
++</li>
++
++<li><a name="named-sig">Named, signature-style, no rest argument</a>
++
++<p><pre>
++(let (<name> (<parameter> <argument>)...)
++ <body>...)
++</pre>
++</li>
++
++<li><a name="named-non-sig-rest">Named, non-signature-style, rest argument</a>
++
++<p><pre>
++(let <name> ((<parameter> <argument>)...
++
++. (<rest-parameter> <rest-argument>...))
++ <body>...)
++</pre>
++
++<li><a name="named-sig-rest">Named, signature-style, rest argument</a>
++
++<p><pre>
++(let (<name> (<parameter> <argument>)...
++
++. (<rest-parameter> <rest-argument>...))
++ <body>...)
++</pre>
++</li>
++</ol>
++
++<H2>Semantics</H2>
++
++Let <code>$lambda</code> and <code>$letrec</code> be hygienic bindings for the <code>lambda</code>
++and <code>letrec</code> forms, respectively.
++
++<ul>
++<li>For informal syntax 1:
++
++<p><pre>
++(($lambda (<parameter>...) <body>...) <argument>...)
++</pre>
++</li>
++
++<li>For informal syntaxes 2 and 3:
++
++<p>
++<pre>
++($letrec ((<name> ($lambda (<parameter>...) <body>...)))
++ (<name> <argument>...))
++</pre>
++</li>
++
++<li>For informal syntaxes 4 and 5:
++
++<p>
++<pre>
++($letrec ((<name> ($lambda (<parameter>...
++
++. <rest-parameter>) <body>...)))
++ (<name> <argument>... <rest-argument>...))
++</pre>
++</li>
++</ul>
++
++<H1>Implementation</H1>
++
++Here is an implementation using <code>SYNTAX-RULES</code>.
++
++<p>
++<pre>
++;; Use your own standard let.
++;; Or call a lambda.
++;; (define-syntax standard-let
++;;
++;; (syntax-rules ()
++;;
++;; ((let ((var val) ...) body ...)
++;; ((lambda (var ...) body ...) val ...))))
++
++(define-syntax let
++
++ (syntax-rules ()
++
++ ;; No bindings: use standard-let.
++ ((let () body ...)
++ (standard-let () body ...))
++ ;; Or call a lambda.
++ ;; ((lambda () body ...))
++
++ ;; All standard bindings: use standard-let.
++ ((let ((var val) ...) body ...)
++ (standard-let ((var val) ...) body ...))
++ ;; Or call a lambda.
++ ;; ((lambda (var ...) body ...) val ...)
++
++ ;; One standard binding: loop.
++ ;; The all-standard-bindings clause didn't match,
++ ;; so there must be a rest binding.
++ ((let ((var val) . bindings) body ...)
++ (let-loop #f bindings (var) (val) (body ...)))
++
++ ;; Signature-style name: loop.
++ ((let (name binding ...) body ...)
++ (let-loop name (binding ...) () () (body ...)))
++
++ ;; defun-style name: loop.
++ ((let name bindings body ...)
++ (let-loop name bindings () () (body ...)))))
++
++(define-syntax let-loop
++
++ (syntax-rules ()
++
++ ;; Standard binding: destructure and loop.
++ ((let-loop name ((var0 val0) binding ...) (var ... ) (val ... ) body)
++ (let-loop name ( binding ...) (var ... var0) (val ... val0) body))
++
++ ;; Rest binding, no name: use standard-let, listing the rest values.
++ ;; Because of let's first clause, there is no "no bindings, no name" clause.
++ ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
++ (standard-let ((var val) ... (rest-var (list rest-val ...))) . body))
++ ;; Or call a lambda with a rest parameter on all values.
++ ;; ((lambda (var ... . rest-var) . body) val ... rest-val ...))
++ ;; Or use one of several other reasonable alternatives.
++
++ ;; No bindings, name: call a letrec'ed lambda.
++ ((let-loop name () (var ...) (val ...) body)
++ ((letrec ((name (lambda (var ...) . body)))
++ name)
++ val ...))
++
++ ;; Rest binding, name: call a letrec'ed lambda.
++ ((let-loop name (rest-var rest-val ...) (var ...) (val ...) body)
++ ((letrec ((name (lambda (var ... . rest-var) . body)))
++ name)
++ val ... rest-val ...))))
++</pre>
++
++
++<H1>Copyright</H1>
++
++Copyright (C) Andy Gaynor (1999). All Rights Reserved.
++<p>This document and translations of it may be copied and furnished to
++others, and derivative works that comment on or otherwise explain it or
++assist in its implementation may be prepared, copied, published and distributed,
++in whole or in part, without restriction of any kind, provided that the
++above copyright notice and this paragraph are included on all such copies
++and derivative works. However, this document itself may not be modified
++in any way, such as by removing the copyright notice or references to the
++Scheme Request For Implementation process or editors, except as needed
++for the purpose of developing SRFIs in which case the procedures for copyrights
++defined in the SRFI process must be followed, or as required to translate
++it into languages other than English.
++<p>The limited permissions granted above are perpetual and will not be
++revoked by the authors or their successors or assigns.
++<p>This document and the information contained herein is provided on an
++"AS IS" basis and THE AUTHOR AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES,
++EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE
++USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
++WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
++
++ <hr>
++ <address>Editor: <a href="mailto:srfi-editors@srfi.schemers.org">Mike Sperber</a></address>
++
++</body>
++</html>
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/info.rkt racket-6.12/share/pkgs/srfi-lib/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/info.rkt 2018-01-26 16:10:02.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/info.rkt 2018-01-31 12:22:58.144502000 -0500
+@@ -1 +1,15 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("scheme-lib" "base" "srfi-lite-lib" "r6rs-lib" "compatibility-lib"))) (define implies (quote ("srfi-lite-lib"))) (define pkg-desc "implementation (no documentation) part of \"srfi\"") (define pkg-authors (quote (mflatt noel chongkai jay)))))
++#lang info
++
++(define collection 'multi)
++
++(define deps '("scheme-lib"
++ "base"
++ "srfi-lite-lib"
++ "r6rs-lib"
++ "compatibility-lib"))
++
++(define implies '("srfi-lite-lib"))
++
++(define pkg-desc "implementation (no documentation) part of \"srfi\"")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/sort.txt racket-6.12/share/pkgs/srfi-lib/srfi/32/sort.txt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/sort.txt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/32/sort.txt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,1069 +0,0 @@
+-The SRFI-32 sort libraries -*- outline -*-
+-Olin Shivers
+-First draft: 1998/10/19
+-Last update: 2002/7/21
+-
+-[Todo: del-list-neighbor-dups!
+- vector-copy -> subvector
+- use srfi-23 for reporting errors
+- use srfi-16 for n-aries?
+-
+-Emacs should display this document in outline mode. Say c-h m for
+-instructions on how to move through it by sections (e.g., c-c c-n, c-c c-p).
+-
+-* Table of contents
+--------------------
+-Abstract
+-Procedure index
+-Introduction
+-What's wrong with the current state of affairs?
+-Design rules
+- What vs. how
+- Consistency across function signatures
+- Data parameter first, less-than parameter after
+- Ordering, comparison functions & stability
+- All vector operations accept optional subrange parameters
+- Required vs. allowed side-effects
+-Procedure specification
+- Procedure naming and functionality
+- Types of parameters and return values
+- sort-lib - general sorting package
+- Algorithm-specific sorting packages
+-Algorithmic properties
+-Topics to be resolved during discussion phase
+-Porting and optimisation
+-References & Links
+-Acknowledgements
+-Copyright
+-
+-
+-* Abstract
+-----------
+-Current Scheme sorting packages are, every one of them, surprisingly bad. I've
+-designed the API for a full-featured sort toolkit, which I propose as a SRFI.
+-
+-The spec comes with 1200 lines of high-quality reference code: tightly
+-written, highly commented, portable code, available for free. Implementors
+-want this code. It's better than what you have.
+-
+--------------------------------------------------------------------------------
+-* Procedure index
+------------------
+-list-sorted? vector-sorted?
+-
+-list-merge vector-merge
+-list-sort vector-sort
+-list-stable-sort vector-stable-sort
+-list-delete-neighbor-dups vector-delete-neighbor-dups
+-
+-list-merge! vector-merge!
+-list-sort! vector-sort!
+-list-stable-sort! vector-stable-sort!
+-list-delete-neighbor-dups! vector-delete-neighbor-dups!
+-
+-quick-sort heap-sort insert-sort list-merge-sort vector-merge-sort
+-quick-sort! heap-sort! insert-sort! list-merge-sort! vector-merge-sort!
+-quick-sort3!
+-
+-vector-binary-search
+-vector-binary-search3
+-
+--------------------------------------------------------------------------------
+-* Introduction
+---------------
+-As I'll detail below, I wasn't very happy with the state of the Scheme
+-world for sorting and merging lists and vectors. So I have designed and
+-written a fairly comprehensive sorting & merging toolkit. It is
+-
+- - very portable,
+-
+- - much better code than what is currently in Elk, Gambit, Bigloo,
+- Scheme->C, MzScheme, RScheme, Scheme48, MIT Scheme, or slib, and
+-
+- - priced to move: free code.
+-
+-The package includes
+- - Vector insert sort (stable)
+- - Vector heap sort
+- - Vector quick sort (with median-of-3 pivot picking)
+- - Vector merge sort (stable)
+- - Pure and destructive list merge sort (stable)
+- - Stable vector and list merge
+- - Miscellaneous sort-related procedures: Vector and list merging,
+- sorted? predicates, vector binary search, vector and list
+- delete-equal-neighbor procedures.
+- - A general, non-algorithmic set of procedure names for general sorting
+- and merging.
+-
+-Scheme programmers may want to adopt this package. I'd like Scheme
+-implementors to adopt this code and its API -- in fact, the code is a bribe to
+-make it easy for implementors to converge on the suggested API. I mean, you'd
+-really have to be a boor to take this free code I wrote and mutate its
+-interface over to your incompatible, unportable API, wouldn't you? But you
+-could, of course -- it's freely available. More in the spirit of the offering,
+-you could make this API available, and then also write a little module
+-providing your old interface that is defined in terms of this API. "Scheme
+-implementors," in this context, includes slib, which is not a standalone
+-implementation of Scheme, but rather an influential collection of API's and
+-code.
+-
+-The code is tightly bummed. It is clearly written, and commented in my usual
+-voluminous style. This includes notes on porting and implementation-specific
+-optimisations.
+-
+-
+--------------------------------------------------------------------------------
+-* What's wrong with the current state of affairs?
+--------------------------------------------------
+-
+-It's just amazing to me that in 2002, sorting and merging hasn't been
+-completely put to bed. These are well-understood algorithms, each of them well
+-under a page of code. The straightforward algorithms are basic, core stuff --
+-sophomore-level. But if you tour the major Scheme implementations out there on
+-the Net, you find badly written code that provides extremely spotty coverage
+-of the algorithm space. One implementation even has a buggy implementation
+-that has been in use for about 20 years. Another has an O(n^2) algorithm...
+-implemented in C for speed.
+-
+-Open source-code is a wonderful thing. In a couple of hours, I was able to
+-download and check the sources of 9 Scheme systems. Here are my notes from the
+-systems I checked. You can skip to the next section if you aren't morbidly
+-curious.
+-
+-slib
+- sorted? vector-or-list <
+- merge list1 list2 <
+- merge! list1 list2 <
+- sort vector-or-list <
+- sort! vector-or-list <
+-
+- Richard O'Keefe's stable list merge sort is right idea, but implemented
+- using gratuitous variable side effects. It also does redundant SET-CDR!s.
+- The vector sort converts to list, merge sorts, then reconverts
+- to vector. This is a bad idea -- non-local pointer chasing bad; vector
+- shuffling good. If you must allocate temp storage, might as well allocate
+- a temp vector and use vector merge sort.
+-
+-MIT Scheme
+- sort! vector <
+- merge-sort! vector <
+- quick-sort! vector <
+-
+- sort vector-or-list <
+- merge-sort vector-or-list <
+- quick-sort vector-or-list <
+-
+- Naive vector quicksort: loser, for worst-case performance reasons.
+- List sort by "list->vector; quicksort; vector->list," hence also loser.
+- A clever stable vector merge sort, albeit not very bummed.
+-
+-Scheme 48 & T
+- sort-list list <
+- sort-list! list <
+- list-merge! list1 list2 <
+-
+- Bob Nix's implementation of online merge-sort, written in the early 80's.
+- Conses unnecessary bookkeeping structure, which isn't necessary with a
+- proper recursive formulation. Also, does redundant SET-CDR!s. No vector
+- sort. Also, has a bug -- is claimed to be a stable sort, but isn't! To see
+- this, get the S48 code, and try
+- (define (my< x y) (< (abs x) (abs y)))
+- (list-merge! (list 0 2) (list -2) my<) ; -> (0 2 -2)
+- (list-merge! (list 2) (list 0 -2) my<) ; -> (0 -2 2)
+- This could be fixed very easily, but it isn't worth it given the
+- other problems with the algorithm.
+-
+-RScheme
+- vector-sort! vector <
+- sort collection <
+-
+- Good basic implementation of vector heapsort, which has O(n lg n)
+- worst-case time. Code ugly, needs tuning. List sort by "list->vector;
+- sort; vector->list." Nothing for stable sorting.
+-
+-MzScheme
+- quicksort lis <
+- mergesort alox <
+-
+- Sorts lists with (list->vector; quicksort; vector->list) -- but the core
+- quicksort is not available for vector sorting. Nothing for stable sorting.
+- Quicksort picks pivot naively, inducing O(n^2) worse-case behaviour on a
+- fairly common case: an already-sorted list.
+-
+-Bigloo, STK
+- sort vector-or-list <
+- Uses an O(n^2) algorithm... implemented in C for speed. Hmm.
+- (See runtime/Ieee/vector.scm and runtime/Clib/cvector.c)
+-
+-Gambit
+- sort-list list <
+- Nothing for vectors. Simple, slow, unstable merge sort for lists.
+-
+-Elk
+- Another naive quicksort. Lists handled by converting to vector.
+- sort vector-or-list <
+- sort! vector-or-list <
+-
+-Chez Scheme
+- merge < list1 list2
+- merge! < list1 list2
+- sort < list
+- sort! < list
+-
+- These are stable. I have not seen the source code.
+-
+-Common Lisp
+- sort sequence < [key]
+- stable-sort sequence < [key]
+- merge result-type sequence1 sequence2 < [key]
+-
+- The sort procedures are allowed, but not required, to be destructive.
+-
+-SML/NJ
+- sort: ('a*'a -> bool) -> 'a list -> 'a list
+- "Smooth applicative merge sort," which is stable.
+- There is also a highly bummed quicksort for vectors.
+-
+-The right solution: Implement a full toolbox of carefully written standard sort
+-routines.
+-
+-Having the source of all these above-cited Schemes available for study made
+-life a lot easier writing this code. I appreciate the authors making their
+-source available under such open terms.
+-
+-
+--------------------------------------------------------------------------------
+-* Design rules
+---------------
+-
+-** What vs. how
+-===============
+-There are two different interfaces: "what" (simple) & "how" (detailed).
+-
+- - Simple: you specify semantics: datatype (list or vector),
+- mutability, and stability.
+-
+- - Detailed: you specify the actual algorithm (quick, heap,
+- insert, merge). Different algorithms have different properties,
+- both semantic & pragmatic, so these exports are necessary.
+-
+- It is necessarily the case that the specifications of these procedures
+- make statements about execution "pragmatics." For example, the sole
+- distinction between heap sort and quick sort -- both of which are
+- provided by this library -- is one of execution time, which is not a
+- "semantic" distinction. Similar resource-use statements are made about
+- "iterative" procedures, meaning that they can execute on input of
+- arbitrary size in a constant number of stack frames.
+-
+-** Consistency across function signatures
+-=========================================
+-The two interfaces share common function signatures wherever
+-possible, to facilitate switching a given call from one procedure
+-to another.
+-
+-** Less-than parameter first, data parameter after
+-==================================================
+-These procedures uniformly observe the following parameter order:
+-the data to be sorted comes after the comparison function.
+-That is, we write
+- (sort < lis)
+-not
+- (sort lis <).
+-
+-With the sole exception of Chez Scheme, this is the exact opposite of
+-every sort function out there in current use in the Scheme world. (See
+-the summary of related APIs above.) However, it is consistent with common
+-practice across Scheme libraries in general to put the ordering function
+-first -- the "operation currying" convention. (E.g., consider FOR-EACH or
+-MAP or FIND.)
+-
+-The original draft of this SRFI used the data-first/comparison-last convention
+-for backwards compatibility -- a decision I made with internal misgivings.
+-Happily, however, the overwhelming response from the discussion phase
+-supported "cleaning up" this issue and re-converging the parameter order with
+-the general Scheme "op currying" convention. So the original decision was
+-inverted in favor of the comparison-first/data-last convention.
+-
+-** Ordering, comparison functions & stability
+-=============================================
+-These routines take a < comparison function, not a <= comparison
+-function, and they sort into increasing order. The difference between
+-a < spec and a <= spec comes up in three places:
+- - the definition of an ordered or sorted data set,
+- - the definition of a stable sorting algorithm, and
+- - correctness of quicksort.
+-
+-+ We say that a data set (a list or vector) is *sorted* or *ordered*
+- if it contains no adjacent pair of values ... X Y ... such that Y < X.
+-
+- In other words, scanning across the data never takes a "downwards" step.
+-
+- If you use a <= procedure where these algorithms expect a <
+- procedure, you may not get the answers you expect. For example,
+- the LIST-SORTED? function will return false if you pass it a <= comparison
+- function and an ordered list containing adjacent equal elements.
+-
+-+ A "stable" sort is one that preserves the pre-existing order of equal
+- elements. Suppose, for example, that we sort a list of numbers by
+- comparing their absolute values, i.e., using comparison function
+- (lambda (x y) (< (abs x) (abs y)))
+- If we sort a list that contains both 3 and -3:
+- ... 3 ... -3 ...
+- then a stable sort is an algorithm that will not swap the order
+- of these two elements, that is, the answer is guaranteed to to look like
+- ... 3 -3 ...
+- not
+- ... -3 3 ...
+-
+- Choosing < for the comparison function instead of <= affects how stability
+- is coded. Given an adjacent pair X Y, (< y x) means "Y should be moved in
+- front of X" -- otherwise, leave things as they are. So using a <= function
+- where a < function is expected will *invert* stability.
+-
+- This is due to the definition of equality, given a < comparator:
+- (and (not (< x y))
+- (not (< y x)))
+- The definition is rather different, given a <= comparator:
+- (and (<= x y)
+- (<= y x))
+-
+-+ A "stable" merge is one that reliably favors one of its data sets
+- when equal items appear in both data sets. *All merge operations in
+- this library are stable*, breaking ties between data sets in favor
+- of the first data set -- elements of the first list come before equal
+- elements in the second list.
+-
+- So, if we are merging two lists of numbers ordered by absolute value,
+- the stable merge operation LIST-MERGE
+- (list-merge (lambda (x y) (< (abs x) (abs y)))
+- '(0 -2 4 8 -10) '(-1 3 -4 7))
+- reliably places the 4 of the first list before the equal-comparing -4
+- of the second list:
+- (0 -1 -2 4 -4 7 8 -10)
+-
+-+ Some sort algorithms will *not work correctly* if given a <= when they
+- expect a < comparison (or vice-versa). For example, violating quicksort's
+- spec may cause it to produce wrong answers, diverge, raise an error, or do
+- some fourth thing. To see why, consider the left-scan part of the standard
+- quicksort partition step:
+- (let ((i (let scan ((i i)) (if (elt< (vector-ref v i) pivot)
+- (scan (+ i 1))
+- i))))
+- ...)
+- Consider applying this loop to a vector of all zeroes (hence, PIVOT, as
+- well, is zero), but erroneously using <= for the ELT< function. The loop
+- will scan right off the end of the vector, producing a vector-index error.
+- The guarantee that the scan loop will terminate before running off the end
+- of the vector depends critically upon ELT< performing as a true, irreflexive
+- < relation. Running off the end of the vector is only one of a variety of
+- possibly ways to lose -- other, variant implementations of quicksort can,
+- instead, loop forever on some data sets if ELT< is a <= predicate.
+-
+-In short, if your comparison function F answers true to (F x x), then
+- - using a stable sorting or merging algorithm will not give you a
+- stable sort or merge,
+- - LIST-SORTED? may surprise you, and
+- - quicksort may fail in a variety of possible ways.
+-Note that you can synthesize a < function from a <= function with
+- (lambda (x y) (not (<= y x)))
+-if need be.
+-
+-Precise definitions give sharp edges to tools, but require care in use.
+-"Measure twice, cut once."
+-
+-I have adopted the choice of < from Common Lisp. One would assume the definers
+-of Common Lisp had a good reason for adopting < instead of <=, but canvassing
+-several of the principal actors in the definition process has turned up no
+-better reason than "an arbitrary but consistent choice." At minimum, then,
+-this SRFI extends the coverage of that consistent choice.
+-
+-** All vector operations accept optional subrange parameters
+-============================================================
+-The vector operations specified below all take optional START/END arguments
+-indicating a selected subrange of a vector's elements. If a START parameter or
+-START/END parameter pair is given to such a procedure, they must be exact,
+-non-negative integers, such that
+- 0 <= START <= END <= (VECTOR-LENGTH V)
+-where V is the related vector parameter. If not specified, they default to 0
+-and the length of the vector, respectively. They are interpreted to select the
+-range [START,END), that is, all elements from index START (inclusive) up to,
+-but not including, index END.
+-
+-** Required vs. allowed side-effects
+-====================================
+-LIST-SORT! and LIST-STABLE-SORT! are allowed, but not required,
+-to alter their arguments' cons cells to construct the result list. This is
+-consistent with the what-not-how character of the group of procedures
+-to which they belong (the "sort-lib" package).
+-
+-The LIST-DELETE-NEIGHBOR-DUPS!, LIST-MERGE! and LIST-MERGE-SORT! procedures,
+-on the other hand, provide specific algorithms, and, as such, explicitly
+-commit to the use of side-effects on their input lists in order to guarantee
+-their key algorithmic properties (e.g., linear-time operation, constant-space
+-stack use).
+-
+--------------------------------------------------------------------------------
+-* Procedure specification
+--------------------------
+-The procedures are split into several packages. In a Scheme system that has a
+-module or package system, these procedures should be contained in modules
+-named as follows:
+- Package name Functionality
+- ------------ -------------
+- sort-lib General sorting for lists & vectors
+- sorted?-lib Sorted predicates for lists & vectors
+- list-merge-sort-lib List merge sort
+- vector-merge-sort-lib Vector merge sort
+- vector-heap-sort-lib Vector heap sort
+- vector-quick-sort-lib Vector quick sort
+- vector-insert-sort-lib Vector insertion sort
+- delndup-lib List and vector delete neighbor duplicates
+- binsearch-lib Vector binary search
+-
+-A Scheme system without a module system should provide all of the bindings
+-defined in all of these modules as components of the "SRFI-32" package.
+-
+-Note that there is no "list insert sort" package, as you might as well always
+-use list merge sort. The reference implementation's destructive list merge
+-sort will do fewer SET-CDR!s than a destructive insert sort.
+-
+-** Procedure naming and functionality
+-=====================================
+-Almost all of the procedures described below are variants of two basic
+-operations: sorting and merging. These procedures are consistently named
+-by composing a set of basic lexemes to indicate what they do.
+-
+-Lexeme Meaning
+------- -------
+-"sort" The procedure sorts its input data set by some < comparison function.
+-
+-"merge" The procedure merges two ordered data sets into a single ordered
+- result.
+-
+-"stable" This lexeme indicates that the sort is a stable one.
+-
+-"vector" The procedure operates upon vectors.
+-
+-"list" The procedure operates upon lists.
+-
+-"!" Procedures that end in "!" are allowed, and sometimes required,
+- to reuse their input storage to construct their answer.
+-
+-** Types of parameters and return values
+-========================================
+-In the procedures specified below,
+- - A LIS parameter is a list;
+-
+- - A V parameter is a vector;
+-
+- - A < or = parameter is a procedure accepting two arguments taken from the
+- specified procedure's data set(s), and returning a boolean;
+-
+- - START and END parameters are exact, non-negative integers that
+- serve as vector indices selecting a subrange of some associated vector.
+- When specified, they must satisfy the relation
+- 0 <= start <= end <= (vector-length v)
+- where V is the associated vector.
+-
+-Passing values to procedures with these parameters that do not satisfy these
+-types is an error.
+-
+-If a procedure is said to return "unspecified," this means that nothing at all
+-is said about what the procedure returns, not even the number of return
+-values. Such a procedure is not even required to be consistent from call to
+-call in the nature or number of its return values. It is simply required to
+-return a value (or values) that may be passed to a command continuation, e.g.
+-as the value of an expression appearing as a non-terminal subform of a BEGIN
+-expression. Note that in R5RS, this restricts such a procedure to returning a
+-single value; non-R5RS systems may not even provide this restriction.
+-
+-** sort-lib - general sorting package
+-=====================================
+-This library provides basic sorting and merging functionality suitable for
+-general programming. The procedures are named by their semantic properties,
+-i.e., what they do to the data (sort, stable sort, merge, and so forth).
+-
+- Procedure Suggested algorithm
+- -------------------------------------------------------------------------
+- list-sorted? < lis -> boolean
+- list-merge < lis1 lis2 -> list
+- list-merge! < lis1 lis2 -> list
+- list-sort < lis -> list (vector heap or quick)
+- list-sort! < lis -> list (list merge sort)
+- list-stable-sort < lis -> list (vector merge sort)
+- list-stable-sort! < lis -> list (list merge sort)
+- list-delete-neighbor-dups = lis -> list
+- list-delete-neighbor-dups! = lis -> list
+-
+- vector-sorted? < v [start end] -> boolean
+- vector-merge < v1 v2 [start1 end1 start2 end2] -> vector
+- vector-merge! < v v1 v2 [start start1 end1 start2 end2] -> unspecified
+- vector-sort < v [start end] -> vector (heap or quick sort)
+- vector-sort! < v [start end] -> unspecified (heap or quick sort)
+- vector-stable-sort < v [start end] -> vector (vector merge sort)
+- vector-stable-sort! < v [start end] -> unspecified (vector merge sort)
+- vector-delete-neighbor-dups = v [start end] -> vector
+- vector-delete-neighbor-dups! = target source [t-start s-start s-end] -> t-end
+-
+- LIST-SORTED? and VECTOR-SORTED? return true if their input list or vector
+- is in sorted order, as determined by their < comparison parameter.
+-
+- All four merge operations are stable: an element of the initial list LIS1
+- or vector V1 will come before an equal-comparing element in the second
+- list LIS2 or vector V2 in the result.
+-
+- The procedures
+- LIST-MERGE
+- LIST-SORT
+- LIST-STABLE-SORT
+- LIST-DELETE-NEIGHBOR-DUPS
+- do not alter their inputs and are allowed to return a value that shares
+- a common tail with a list argument.
+-
+- The procedures
+- LIST-SORT!
+- LIST-STABLE-SORT!
+- are "linear update" operators -- they are allowed, but not required, to
+- alter the cons cells of their arguments to produce their results.
+-
+- On the other hand, the procedures
+- LIST-DELETE-NEIGHBOR-DUPS!
+- LIST-MERGE!
+- make only a single, iterative, linear-time pass over their argument lists,
+- using SET-CDR!s to rearrange the cells of the lists into the final result
+- -- they work "in place." Hence, any cons cell appearing in the result must
+- have originally appeared in an input. The intent of this
+- iterative-algorithm commitment is to allow the programmer to be sure that
+- if, for example, LIST-MERGE! is asked to merge two ten-million-element
+- lists, the operation will complete without performing some extremely
+- (possibly twenty-million) deep recursion.
+-
+- The vector procedures
+- VECTOR-SORT
+- VECTOR-STABLE-SORT
+- VECTOR-DELETE-NEIGHBOR-DUPS
+- do not alter their inputs, but allocate a fresh vector for their result,
+- of length END - START.
+-
+- The vector procedures
+- VECTOR-SORT!
+- VECTOR-STABLE-SORT!
+- sort their data in-place. (But note that VECTOR-STABLE-SORT! may
+- allocate temporary storage proportional to the size of the input --
+- I am not aware of O(n lg n) stable vector-sorting algorithms that
+- run in constant space.)
+-
+- VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
+-
+- VECTOR-MERGE! writes its result into vector V, beginning at index START,
+- for indices less than END = START + (END1-START1) + (END2-START2). The
+- target subvector
+- V[start,end)
+- may not overlap either source subvector
+- V1[start1,end1)
+- V2[start2,end2).
+-
+- The ...-DELETE-NEIGHBOR-DUPS-... procedures:
+- These procedures delete adjacent duplicate elements from a list or a
+- vector, using a given element-equality procedure. The first/leftmost
+- element of a run of equal elements is the one that survives. The list or
+- vector is not otherwise disordered.
+-
+- These procedures are linear time -- much faster than the O(n^2) general
+- duplicate-element deletors that do not assume any "bunching" of elements
+- (such as the ones provided by SRFI-1). If you want to delete duplicate
+- elements from a large list or vector, you can sort the elements to bring
+- equal items together, then use one of these procedures, for a total time
+- of O(n lg n).
+-
+- The comparison function = passed to these procedures is always applied
+- (= x y)
+- where X comes before Y in the containing list or vector.
+-
+- - LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
+- may share storage with the input list.
+-
+- - VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
+- rather allocates a fresh vector to hold the result.
+-
+- - LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
+- mutate its input list in order to construct its answer.
+-
+- - VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
+- answer, packing its answer into the index range [start,end'), where
+- END' is the non-negative exact integer returned as its value. It
+- returns END' as its result. The vector is not altered outside the range
+- [start,end').
+-
+- - VECTOR-DELETE-NEIGHBOR-DUPS! scans vector SOURCE in range
+- [S-START,S-END), writing its result to vector TARGET beginning at index
+- T-START. It returns exact, non-negative integer T-END, which indicates
+- that the results of the operation are found in index range
+- [T-START,T-END) of TARGET; elements of TARGET outside this range
+- are unaltered.
+-
+- It is an error for memory cell TARGET[T-START] to be a memory cell in
+- the region SOURCE[1 + S-START, S-END). In a Scheme implementation
+- that does not allow distinct vectors to share storage, this means
+- that one of the following must be true:
+- 1. (not (eq? source target))
+- 2. t-start not-in [s-start + 1, s-end)
+-
+- - Examples:
+- (list-delete-neighbor-dups = '(1 1 2 7 7 7 0 -2 -2))
+- => (1 2 7 0 -2)
+-
+- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2))
+- => #(1 2 7 0 -2)
+-
+- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2) 3 7)
+- => #(7 0 -2)
+-
+- ;; Result left in v[3,9):
+- (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
+- (cons (vector-delete-neighbor-dups! = v 3)
+- v))
+- => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
+-
+-
+-** Algorithm-specific sorting packages
+-======================================
+-These packages provide more specific sorting functionality, that is,
+-specific commitment to particular algorithms that have particular
+-pragmatic consequences (such as memory locality, asymptotic running time)
+-beyond their semantic behaviour (sorting, stable sorting, merging, etc.).
+-Programmers that need a particular algorithm can use one of these packages.
+-
+-sorted?-lib - sorted predicates
+- list-sorted? < lis -> boolean
+- vector-sorted? < v [start end] -> boolean
+-
+- Return #f iff there is an adjacent pair ... X Y ... in the input
+- list or vector such that Y < X. The optional START/END range
+- arguments restrict VECTOR-SORTED? to the indicated subvector.
+-
+-list-merge-sort-lib - list merge sort
+- list-merge-sort < lis -> list
+- list-merge-sort! < lis -> list
+- list-merge lis1 < lis2 -> list
+- list-merge! lis1 < lis2 -> list
+-
+- The sort procedures sort their data using a list merge sort, which is
+- stable. (The reference implementation is, additionally, a "natural" sort.
+- See below for the properties of this algorithm.)
+-
+- The ! procedures are destructive -- they use SET-CDR!s to rearrange the
+- cells of the lists into the proper order. As such, they do not allocate
+- any extra cons cells -- they are "in place" sorts. Additionally,
+- LIST-MERGE! is iterative -- it can operate on arguments of arbitrary size
+- with a constant number of stack frames.
+-
+- The merge operations are stable: an element of LIS1 will come before an
+- equal-comparing element in LIS2 in the result list.
+-
+-vector-merge-sort-lib - vector merge sort
+- vector-merge-sort < v [start end temp] -> vector
+- vector-merge-sort! < v [start end temp] -> unspecified
+- vector-merge < v1 v2 [start1 end1 start2 end2] -> vector
+- vector-merge! < v v1 v2 [start start1 end1 start2 end2] -> unspecified
+-
+- The sort procedures sort their data using vector merge sort, which is
+- stable. (The reference implementation is, additionally, a "natural" sort.
+- See below for the properties of this algorithm.)
+-
+- The optional START/END arguments provide for sorting of subranges, and
+- default to 0 and the length of the corresponding vector.
+-
+- Merge-sorting a vector requires the allocation of a temporary "scratch"
+- work vector for the duration of the sort. This scratch vector can be
+- passed in by the client as the optional TEMP argument; if so, the supplied
+- vector must be of size >= END, and will not be altered outside the range
+- [start,end). If not supplied, the sort routines allocate one themselves.
+-
+- The merge operations are stable: an element of V1 will come before an
+- equal-comparing element in V2 in the result vector.
+-
+- VECTOR-MERGE-SORT! leaves its result in V[start,end).
+-
+- VECTOR-MERGE-SORT returns a vector of length END-START.
+-
+- VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
+-
+- VECTOR-MERGE! writes its result into vector V, beginning at index START,
+- for indices less than END = START + (END1-START1) + (END2-START2). The
+- target subvector
+- V[start,end)
+- may not overlap either source subvector
+- V1[start1,end1)
+- V2[start2,end2).
+-
+-vector-heap-sort-lib - vector heap sort
+- heap-sort < v [start end] -> vector
+- heap-sort! < v [start end] -> unspecified
+-
+- These procedures sort their data using heap sort,
+- which is not a stable sorting algorithm.
+-
+- HEAP-SORT returns a vector of length END-START.
+- HEAP-SORT! is in-place, leaving its result in V[start,end).
+-
+-vector-quick-sort-lib - vector quick sort
+- quick-sort < v [start end] -> vector
+- quick-sort! < v [start end] -> unspecified
+- quick-sort3! c v [start end] -> unspecified
+-
+- These procedures sort their data using quick sort,
+- which is not a stable sorting algorithm.
+-
+- QUICK-SORT returns a vector of length END-START.
+- QUICK-SORT! is in-place, leaving its result in V[start,end).
+-
+- QUICK-SORT3! is a variant of quick-sort that takes a three-way
+- comparison function C. C compares a pair of elements and returns
+- an exact integer whose sign indicates their relationship:
+- (c x y) < 0 => x<y
+- (c x y) = 0 => x=y
+- (c x y) > 0 => x>y
+- To help remember the relationship between the sign of the result and
+- the relation, use the function - as the model for C: (- x y) < 0
+- means that x < y; (- x y) > 0 means that x > y.
+-
+- The extra discrimination provided by the three-way comparison can
+- provide significant speedups when sorting data sets with many duplicates,
+- especially when the comparison function is relatively expensive (e.g.,
+- comparing long strings).
+-
+- WARNING: Some sort algorithms, such as insertion sort or heap sort,
+- can tolerate being passed a <= comparison function when they expect a <
+- function -- insertion and merge sort may simply invert stability; and
+- heap sort will run a bit slower, but otherwise produce a correct answer.
+-
+- Quicksort, however, is much more critically sensitive to the distinction
+- between a < and a <= comparison. If QUICK-SORT or QUICK-SORT! expect a <
+- comparison function, and are erroneously given a <= function, they may,
+- depending on implementation, produce an unsorted result, go into an
+- infinite loop, cause a run-time error, occasionally produce a correct
+- result, or do some fifth thing.
+-
+- Implementors may wish to write QUICKSORT3! so that it (a) tests the
+- comparison function (by checking that (c v[start] v[start]) produces
+- false), or (b) is tolerant of an erroneous <= function, or (c) both.
+- Clients of this function, however, should not count on this.
+-
+-vector-insert-sort-lib - vector insertion sort
+- insert-sort < v [start end] -> vector
+- insert-sort! < v [start end] -> unspecified
+-
+- These procedures stably sort their data using insertion sort.
+-
+- INSERT-SORT returns a vector of length END-START.
+- INSERT-SORT! is in-place, leaving its result in V[start,end).
+-
+-delndup-lib - list and vector delete neighbor duplicates
+- list-delete-neighbor-dups = lis -> list
+- list-delete-neighbor-dups! = lis -> list
+-
+- vector-delete-neighbor-dups = v [start end] -> vector
+- vector-delete-neighbor-dups! = v [start end] -> end'
+-
+- These procedures delete adjacent duplicate elements from a list or
+- a vector, using a given element-equality procedure =. The first/leftmost
+- element of a run of equal elements is the one that survives. The list
+- or vector is not otherwise disordered.
+-
+- These procedures are linear time -- much faster than the O(n^2) general
+- duplicate-element deletors that do not assume any "bunching" of elements
+- (such as the ones provided by SRFI-1). If you want to delete duplicate
+- elements from a large list or vector, you can sort the elements to bring
+- equal items together, then use one of these procedures, for a total time
+- of O(n lg n).
+-
+- The comparison function = passed to these procedures is always applied
+- (= x y)
+- where X comes before Y in the containing list or vector.
+-
+- LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
+- may share storage with the input list.
+-
+- VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
+- rather allocates a fresh vector to hold the result.
+-
+- LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
+- mutate its input list in order to construct its answer.
+-
+- VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
+- answer, packing its answer into the index range [start,end'), where
+- END' is the non-negative exact integer returned as its value. It
+- returns END' as its result. The vector is not altered outside the range
+- [start,end').
+-
+- Examples:
+- (list-delete-neighbor-dups = '(1 1 2 7 7 7 0 -2 -2))
+- => (1 2 7 0 -2)
+-
+- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2))
+- => #(1 2 7 0 -2)
+-
+- (vector-delete-neighbor-dups = '#(1 1 2 7 7 7 0 -2 -2) 3 7)
+- => #(7 0 -2)
+-
+- ;; Result left in v[3,9):
+- (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
+- (cons (vector-delete-neighbor-dups! = v 3)
+- v))
+- => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
+-
+-binsearch-lib - vector binary search lib
+- vector-binary-search elt< elt->key key v [start end] -> integer-or-false
+- vector-binary-search3 c v [start end] -> integer-or-false
+-
+- VECTOR-BINARY-SEARCH searches vector V in range [START,END) (which
+- default to 0 and the length of V, respectively) for an element whose
+- associated key is equal to KEY. The procedure ELT->KEY is used to map
+- an element to its associated key. The elements of the vector are assumed
+- to be ordered by the ELT< relation on these keys. That is,
+- (vector-sorted? (lambda (x y) (elt< (elt->key x) (elt->key y)))
+- v start end) => true
+- An element E of V is a match for KEY if it's neither less nor greater
+- than the key:
+- (and (not (elt< (elt->key e) key))
+- (not (elt< key (elt->key e))))
+- If there is such an element, the procedure returns its index in the
+- vector as an exact integer. If there is no such element in the searched
+- range, the procedure returns false.
+-
+- (vector-binary-search < car 4 '#((1 . one) (3 . three)
+- (4 . four) (25 . twenty-five)))
+- => 2
+-
+- (vector-binary-search < car 7 '#((1 . one) (3 . three)
+- (4 . four) (25 . twenty-five)))
+- => #f
+-
+- VECTOR-BINARY-SEARCH3 is a variant that uses a three-way comparison
+- function C. C compares its parameter to the search key, and returns an
+- exact integer whose sign indicates its relationship to the search key.
+- (c x) < 0 => x < search-key
+- (c x) = 0 => x = search-key
+- (c x) > 0 => x > search-key
+-
+- (vector-binary-search3 (lambda (elt) (- (car elt) 4))
+- '#((1 . one) (3 . three)
+- (4 . four) (25 . twenty-five)))
+- => 2
+-
+- Rationale:
+- - Why isn't VECTOR-BINARY-SEARCH's ELT->KEY computation simply absorbed
+- into the < function? It is separated out because the < function is
+- applied twice inside the binary-search inner loop, once with the search
+- key for the first argument and the element key for the second argument,
+- and once, with the reverse argument order. This is not necessary for
+- VECTOR-BINARY-SEARCH3.
+-
+- - When a comparison operation is able to produce a three-way
+- discrimination, the inner loop of the binary search can trim the number
+- of per-iteration comparisons from an average of 1.5 to a guaranteed
+- single comparison per iteration. This can be a significant savings when
+- searching with an expensive comparison operation (e.g., one that
+- uses string compare, sends email, references a database, or queries
+- a network service such as a web server).
+-
+- - Failure is signaled by false (rather than, say, -1) so that searches
+- can be used in conditional forms such as
+- (or (vector-binary-search ...) ...)
+- or
+- (cond ((vector-binary-search ...) => index-consumer)
+- ...)
+-
+--------------------------------------------------------------------------------
+-* Algorithmic properties
+-------------------------
+-Different sort and merge algorithms have different properties.
+-Choose the algorithm that matches your needs:
+-
+-Vector insert sort
+- Stable, but only suitable for small vectors -- O(n^2).
+-
+-Vector quick sort
+- Not stable. Is fast on average -- O(n lg n) -- but has bad worst-case
+- behaviour. Has good memory locality for big vectors (unlike heap sort).
+- A clever pivot-picking trick (median of three samples) helps avoid
+- worst-case behaviour, but pathological cases can still blow up.
+-
+-Vector heap sort
+- Not stable. Guaranteed fast -- O(n lg n) *worst* case. Poor locality
+- on large vectors. A very reliable workhorse.
+-
+-Vector merge sort
+- Stable. Not in-place -- requires a temporary buffer of equal size.
+- Fast -- O(n lg n) -- and has good memory locality for large vectors.
+-
+- The implementation of vector merge sort provided by this SRFI's reference
+- implementation is, additionally, a "natural" sort, meaning that it
+- exploits existing order in the input data, providing O(n) best case.
+-
+-Destructive list merge sort
+- Stable, fast and in-place (i.e., allocates no new cons cells). "Fast"
+- means O(n lg n) worse-case, and substantially better if the data
+- is already mostly ordered, all the way down to linear time for
+- a completely-ordered input list (i.e., it is a "natural" sort).
+-
+- Note that sorting lists involves chasing pointers through memory, which
+- can be a loser on modern machine architectures because of poor cache &
+- page locality. Pointer *writing*, which is what the SET-CDR!s of a
+- destructive list-sort algorithm do, is even worse, especially if your
+- Scheme has a generational GC -- the writes will thrash the write-barrier.
+- Sorting vectors has inherently better locality.
+-
+- This SRFI's destructive list merge and merge sort implementations are
+- opportunistic -- they avoid redundant SET-CDR!s, and try to take long
+- already-ordered runs of list structure as-is when doing the merges.
+-
+-Pure list merge sort
+- Stable and fast -- O(n lg n) worst-case, and possibly O(n), depending
+- upon the input list (see discussion above).
+-
+-
+-Algorithm Stable? Worst case Average case In-place
+-------------------------------------------------------
+-Vector insert Yes O(n^2) O(n^2) Yes
+-Vector quick No O(n^2) O(n lg n) Yes
+-Vector heap No O(n lg n) O(n lg n) Yes
+-Vector merge Yes O(n lg n) O(n lg n) No
+-List merge Yes O(n lg n) O(n lg n) Either
+-
+-
+--------------------------------------------------------------------------------
+-* Porting and optimisation
+---------------------------
+-This package should be trivial to port. There are only four non-R4RS bits
+-in the code:
+-- Use of multiple-value return, with the R5RS VALUES procedure, and the
+- simple (RECEIVE (var ...) mv-exp body ...) multiple-value binding macro
+- of SRFI-8.
+-
+-- A VECTOR-COPY procedure. This is a tiny little procedure:
+- (vector-copy v [start end])
+-
+-- Use of the LET-OPTIONALS macro from scsh to parse and default optional
+- arguments to three routines. Again, easy to port the macro or rewrite
+- the code to parse, default, and error check the args by hand.
+-
+-- Calls to an ERROR function for complaining about bad arguments.
+-
+-This code is tightly bummed, as far as I can go in portable Scheme.
+-
+-You could speed up the vector code a lot by error-checking the procedure
+-parameters and then shifting over to fixnum-specific arithmetic and dangerous
+-vector-indexing and vector-setting primitives. The comments in the code
+-indicate where the initial error checks would have to be added. There are
+-several (QUOTIENT N 2)'s that could be changed to a fixnum right-shift, as
+-well, in both the list and vector code (SRFI 33 provides such an operator).
+-The code is designed to enable this -- each file usually exports one or two
+-"safe" procedures that end up calling an internal "dangerous" primitive. The
+-little exported cover procedures are where you move the error checks.
+-
+-This should provide *big* speedups. In fact, all the code bumming I've done
+-pretty much disappears in the noise unless you have a good compiler and also
+-can dump the vector-index checks and generic arithmetic -- so I've really just
+-set things up for you to exploit.
+-
+-The optional-arg parsing, defaulting, and error checking is done with a
+-portable R4RS macro. But if your Scheme has a faster mechanism (e.g., Chez),
+-you should definitely port over to it. Note that argument defaulting and
+-error-checking are interleaved -- you don't have to error-check defaulted
+-START/END args to see if they are fixnums that are legal vector indices for
+-the corresponding vector, etc.
+-
+-
+--------------------------------------------------------------------------------
+-* References & Links
+---------------------
+-
+-This document, in HTML:
+- http://srfi.schemers.org/srfi-32/srfi-32.html
+- [This link may not be valid while the SRFI is in draft form.]
+-
+-This document, in simple text format:
+- http://srfi.schemers.org/srfi-32/srfi-32.txt
+-
+-Archive of SRFI-32 discussion-list email:
+- http://srfi.schemers.org/srfi-32/mail-archive/maillist.html
+-
+-SRFI web site:
+- http://srfi.schemers.org/
+-
+-[CommonLisp]
+- Common Lisp: the Language
+- Guy L. Steele Jr. (editor).
+- Digital Press, Maynard, Mass., second edition 1990.
+- Available at http://www.elwood.com/alu/table/references.htm#cltl2
+-
+- The Common Lisp "HyperSpec," produced by Kent Pitman, is essentially
+- the ANSI spec for Common Lisp:
+- http://www.xanalys.com/software_tools/reference/HyperSpec/
+-
+-[R5RS]
+- Revised^5 Report on the Algorithmic Language Scheme,
+- R. Kelsey, W. Clinger, J. Rees (editors).
+- Higher-Order and Symbolic Computation, Vol. 11, No. 1, September, 1998.
+- and ACM SIGPLAN Notices, Vol. 33, No. 9, October, 1998.
+-
+- Available at http://www.schemers.org/Documents/Standards/
+-
+-
+--------------------------------------------------------------------------------
+-* Acknowledgements
+-------------------
+-
+-I thank the authors of the open source I consulted when designing this
+-library, particularly Richard O'Keefe, Donovan Kolby and the MIT Scheme Team.
+-
+-
+--------------------------------------------------------------------------------
+-* Copyright
+------------
+-
+-** SRFI text
+-============
+-This document is copyright (C) Olin Shivers (1998, 1999).
+-All Rights Reserved.
+-
+-This document and translations of it may be copied and furnished to others,
+-and derivative works that comment on or otherwise explain it or assist in its
+-implementation may be prepared, copied, published and distributed, in whole or
+-in part, without restriction of any kind, provided that the above copyright
+-notice and this paragraph are included on all such copies and derivative
+-works. However, this document itself may not be modified in any way, such as
+-by removing the copyright notice or references to the Scheme Request For
+-Implementation process or editors, except as needed for the purpose of
+-developing SRFIs in which case the procedures for copyrights defined in the
+-SRFI process must be followed, or as required to translate it into languages
+-other than English.
+-
+-The limited permissions granted above are perpetual and will not be revoked by
+-the authors or their successors or assigns.
+-
+-This document and the information contained herein is provided on an "AS IS"
+-basis and THE AUTHORS AND THE SRFI EDITORS DISCLAIM ALL WARRANTIES, EXPRESS OR
+-IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE
+-INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF
+-MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
+-
+-** Reference implementation
+-===========================
+-Short summary: no restrictions.
+-
+-While I wrote all of this code myself, I read a lot of code before I began
+-writing. However, all such code is, itself, either open source or public
+-domain, rendering irrelevant any issue of "copyright taint."
+-
+-The natural merge sorts (pure list, destructive list, and vector) are not only
+-my own code, but are implementations of an algorithm of my own devising. They
+-run in O(n lg n) worst case, O(n) best case, and require only a logarithmic
+-number of stack frames. And they are stable. And the destructive-list variant
+-allocates zero cons cells; it simply rearranges the cells of the input list.
+-
+-Hence the reference implementation is
+- Copyright (c) 1998 by Olin Shivers.
+-and made available under the same copyright as the SRFI text (see above).
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/srfi-32.txt racket-6.12/share/pkgs/srfi-lib/srfi/32/srfi-32.txt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/32/srfi-32.txt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/32/srfi-32.txt 2018-01-31 12:22:58.152501970 -0500
+@@ -0,0 +1,938 @@
++The SRFI-32 sort libraries -*- outline -*-
++Olin Shivers
++First draft: 1998/10/19
++Last update: 2002/7/21
++
++Emacs should display this document in outline mode. Say c-h m for
++instructions on how to move through it by sections (e.g., c-c c-n, c-c c-p).
++
++* Table of contents
++-------------------
++Abstract
++Procedure index
++Introduction
++What's wrong with the current state of affairs?
++Design rules
++ What vs. how
++ Consistency across function signatures
++ Data parameter first, less-than parameter after
++ Ordering, comparison functions & stability
++ All vector operations accept optional subrange parameters
++ Required vs. allowed side-effects
++Procedure specification
++ Procedure naming and functionality
++ Types of parameters and return values
++ sort-lib - general sorting package
++ Algorithm-specific sorting packages
++Algorithmic properties
++Topics to be resolved during discussion phase
++Porting and optimisation
++References & Links
++Acknowledgements
++Copyright
++
++
++* Abstract
++----------
++Current Scheme sorting packages are, every one of them, surprisingly bad. I've
++designed the API for a full-featured sort toolkit, which I propose as an SRFI.
++The spec comes with 1200 lines of high-quality reference code: tightly
++written, highly commented, portable code, available for free. Implementors
++want this code. It's better than what you have.
++
++-------------------------------------------------------------------------------
++* Procedure index
++-----------------
++list-sorted? vector-sorted?
++
++list-merge vector-merge
++list-sort vector-sort
++list-stable-sort vector-stable-sort
++list-delete-neighbor-dups vector-delete-neighbor-dups
++
++list-merge! vector-merge!
++list-sort! vector-sort!
++list-stable-sort! vector-stable-sort!
++list-delete-neighbor-dups! vector-delete-neighbor-dups!
++
++heap-sort quick-sort insert-sort list-merge-sort vector-merge-sort
++heap-sort! quick-sort! insert-sort! list-merge-sort! vector-merge-sort!
++
++-------------------------------------------------------------------------------
++* Introduction
++--------------
++As I'll detail bewlow, I wasn't very happy with the state of the Scheme
++world for sorting and merging lists and vectors. So I have designed and
++written a fairly comprehensive sorting & merging toolkit. It is
++
++ - very portable,
++
++ - much better code than what is currently in Elk, Gambit, Bigloo,
++ Scheme->C, MzScheme, RScheme, Scheme48, MIT Scheme, or slib, and
++
++ - priced to move: free code.
++
++The package includes
++ - Vector insert sort (stable)
++ - Vector heap sort
++ - Vector quick sort (with median-of-3 pivot picking)
++ - Vector merge sort (stable)
++ - Pure and destructive list merge sort (stable)
++ - Stable vector and list merge
++ - Miscellaneous sort-related procedures: Vector and list merging,
++ sorted? predicates, vector binary search, vector and list
++ delete-equal-neighbor procedures.
++ - A general, non-algorithmic set of procedure names for general sorting
++ and merging.
++
++Scheme programmers may want to adopt this package. I'd like Scheme
++implementors to adopt this code and its API -- in fact, the code is a bribe to
++make it easy for implementors to converge on the suggested API. I mean, you'd
++really have to be a boor to take this free code I wrote and mutate its
++interface over to your incompatible, unportable API, wouldn't you? But you
++could, of course -- it's freely available. More in the spirit of the offering,
++you could make this API available, and then also write a little module
++providing your old interface that is defined in terms of this API. "Scheme
++implementors," in this context, includes slib, which isn't really a standalone
++implementation of Scheme, but is an influential collection of API's and code.
++
++The code is tightly bummed. It is clearly written, and commented in my usual
++voluminous style. This includes notes on porting and implementation-specific
++optimisations.
++
++
++-------------------------------------------------------------------------------
++* What's wrong with the current state of affairs?
++-------------------------------------------------
++
++It's just amazing to me that in 2002, sorting and merging hasn't been
++completely put to bed. These are well-understood algorithms, each of them well
++under a page of code. The straightforward algorithms are basic, core stuff --
++sophomore-level. But if you tour the major Scheme implementations out there on
++the Net, you find badly written code that provides extremely spotty coverage
++of the algorithm space. One implementation even has a buggy implementation
++that has been in use for about 20 years!
++
++Open source-code is a wonderful thing. In a couple of hours, I was able to
++download and check the sources of 9 Scheme systems. Here are my notes from the
++systems I checked. You can skip to the next section if you aren't morbidly
++curious.
++
++slib
++ sorted? vector-or-list <
++ merge list1 list2 <
++ merge! list1 list2 <
++ sort vector-or-list <
++ sort! vector-or-list <
++
++ Richard O'Keefe's stable list merge sort is right idea, but implemented
++ using gratuitous variable side effects. It also does redundant SET-CDR!s.
++ The vector sort converts to list, merge sorts, then reconverts
++ to vector. This is a bad idea -- non-local pointer chasing bad; vector
++ shuffling good.
++
++MIT Scheme
++ sort! vector <
++ merge-sort! vector <
++ quick-sort! vector <
++
++ sort vector-or-list <
++ merge-sort vector-or-list <
++ quick-sort vector-or-list <
++
++ Naive vector quicksort: loser, for worst-case performance reasons.
++ List sort by "list->vector; quicksort; vector->list," hence also loser.
++ A clever stable vector merge sort, albeit not very bummed.
++
++Scheme 48 & T
++ sort-list list <
++ sort-list! list <
++ list-merge! list1 list2 <
++
++ Bob Nix's implementation of online merge-sort, written in the early 80's.
++ Conses unnecessary bookkeeping structure, which isn't necessary with a
++ proper recursive formulation. Also, does redundant SET-CDR!s. No vector
++ sort. Also, has a bug -- is claimed to be a stable sort, but isn't! To see
++ this, get the S48 code, and try
++ (define (my< x y) (< (quotient x 2) (quotient y 2)))
++ (list-merge! (list 0 2) (list 3) my<) ; -> (0 2 3)
++ (list-merge! (list 2) (list 0 3) my<) ; -> (0 3 2)
++ This could be fixed very easily, but it isn't worth it given the
++ other problems with the algorithm.
++
++RScheme
++ vector-sort! vector <
++ sort collection <
++
++ Good basic implementation of vector heapsort, which has O(n lg n)
++ worst-case time. Code ugly, needs tuning. List sort by "list->vector;
++ sort; vector->list", which allocates unneeded temp storage. Nothing
++ for stable sorting.
++
++MzScheme
++ Naive quicksort -- but not available for vector sorting, even
++ though it internally uses a vector. Nothing for stable sorting,
++ and naive quicksort has bad worst-case behaviour.
++
++Bigloo, Scheme->C
++ Couldn't find anything -- but maybe I didn't search for the right
++ thing, since the Bigloo names are French. (I invite correction from
++ the Bigloo implementors.)
++
++Gambit
++ sort-list list <
++ Nothing for vectors. Simple, slow, unstable merge sort for lists.
++
++Elk
++ Another naive quicksort. Lists handled by converting to vector.
++ sort vector-or-list <
++ sort! vector-or-list <
++
++Chez Scheme
++ merge < list1 list2
++ merge! < list1 list2
++ sort < list
++ sort! < list
++
++ These are stable. I have not seen the source code.
++
++Common Lisp
++ sort sequence < [key]
++ stable-sort sequence < [key]
++ merge result-type sequence1 sequence2 < [key]
++
++ The sort procedures are allowed, but not required, to be destructive.
++
++SML/NJ
++ sort: ('a*'a -> bool) -> 'a list -> 'a list
++ "Smooth applicative merge sort," which is stable.
++ There is also a highly bummed quicksort for vectors.
++
++The right solution: Implement a full toolbox of carefully written standard sort
++routines.
++
++Having the source available for all of these above-cited Schemes made
++life a lot easier writing this code. I appreciate the authors making their
++source available under such open terms.
++
++
++-------------------------------------------------------------------------------
++* Design rules
++--------------
++
++** What vs. how
++===============
++There are two different interfaces: "what" (simple) & "how" (detailed).
++
++ - Simple: you specify semantics: datatype (list or vector),
++ mutability, and stability.
++
++ - Detailed: you specify the actual algorithm (quick, heap,
++ insert, merge). Different algorithms have different properties,
++ both semantic & pragmatic, so these exports are necessary.
++
++ It is necessarily the case that the specifications of these procedures
++ make statements about execution "pragmatics." For example, the sole
++ distinction between heap sort and quick sort -- both of which are
++ provided by this library -- is one of execution time, which is not a
++ "semantic" distinction. Similar resource-use statements are made about
++ "iterative" procedures, meaning that they can execute on input of
++ arbitrary size without needing to allocate an unbounded number of stack
++ frames.
++
++** Consistency across function signatures
++=========================================
++The two interfaces share common function signatures wherever
++possible, to facilitate switching a given call from one procedure
++to another.
++
++** Data parameter first, less-than parameter after
++==================================================
++These procedures uniformly observe the following parameter order:
++the data to be sorted come before the the comparison function.
++That is, we write
++ (sort lis <)
++not
++ (sort < lis).
++This is consistent with every single implementation out there, with
++the sole exception of Chez Scheme.
++
++In my opinion, it would be more consistent with other Scheme libraries
++to put the ordering function first -- the "operation currying" convention.
++(E.g., consider FOR-EACH or MAP or FIND.) I decided to leave things as they
++are in favor of near-total backwards compatibility with existing practice.
++
++[Perhaps this should be discussed.]
++
++** Ordering, comparison functions & stability
++=============================================
++These routines take a < comparison function, not a <= comparison
++function, and they sort into increasing order. The difference between
++a < spec and a <= spec comes up in two places:
++ - the definition of an ordered or sorted data set, and
++ - the definition of a stable sorting algorithm.
++
+++ We say that a data set (a list or vector) is *sorted* or *ordered*
++ if it contains no adjacent pair of values ... X Y ... such that Y < X.
++
++ In other words, scanning across the data never takes a "downwards" step.
++
++ If you use a <= procedure where these algorithms expect a <
++ procedure, you may not get the answers you expect. For example,
++ the LIST-SORTED? function will return false if you pass it a <= comparison
++ function and an ordered list containing adjacent equal elements.
++
+++ A "stable" sort is one that preserves the pre-existing order of equal
++ elements. Suppose, for example, that we sort a list of numbers by
++ comparing their absolute values, i.e., using comparison function
++ (lambda (x y) (< (abs x) (abs y)))
++ If we sort a list that contains both 3 and -3:
++ ... 3 ... -3 ...
++ then a stable sort is an algorithm that will not swap the order
++ of these two elements, that is, the answer will look like
++ ... 3 -3 ...
++ not
++ ... -3 3 ...
++
++ Choosing < for the comparison function instead of <= affects how stability
++ is coded. Given an adjacent pair X Y, (< y x) means "Y should be moved in
++ front of X" -- otherwise, leave things as they are. So using a <= function
++ where a < function is expected will *invert* stability.
++
++ This is due to the definition of equality, given a < comparator:
++ (and (not (< x y))
++ (not (< y x)))
++ The definition is rather different, given a <= comparator:
++ (and (<= x y)
++ (<= x y))
++
+++ A "stable" merge is one that reliably favors one of its data sets
++ when equal items appear in both data sets. *All merge operations in
++ this library are stable*, breaking ties between data sets in favor
++ of the first data set -- elements of the first list come before equal
++ elements in the second list.
++
++ So, if we are merging two lists of numbers ordered by absolute value
++ using the stable merge operation LIST-MERGE
++ (list-merge '(0 -2 4 8 -10) '(-1 3 -4 7)
++ (lambda (x y) (< (abs x) (abs y))))
++ reliably places the 4 of the first list before the equal-comparing -4
++ of the second list:
++ (0 -1 -2 4 -4 7 8 -10)
++
++In short, if your comparison function F answers true to (F x x), then
++using a stable sorting or merging algorithm will not give you a stable sort
++or merge, and LIST-SORTED? may surprise you. Note that you can synthesize a <
++function from a <= function with
++ (lambda (x y) (not (<= y x)))
++if need be.
++
++Precise definitions give sharp edges to tools, but require care
++in use. "Measure twice, cut once."
++
++I have adopted the choice of < from Common Lisp. I assume they
++had a good reason for adopting < instead of <=. I'd love to know
++what this reason is; send me email if you can explain it, please.
++
++** All vector operations accept optional subrange parameters
++============================================================
++The vector operations specified below all take optional START/END arguments
++indicating a selected subrange of a vector's elements. If a START parameter or
++START/END parameter pair is given to such a procedure, they must be exact,
++non-negative integers, such that
++ 0 <= START <= END <= (VECTOR-LENGTH V)
++where V is the related vector parameter. If not specified, they default to 0
++and the length of the vector, respectively. They are interpreted to select the
++range [START,END), that is, all elements from index START (inclusive) up to,
++but not including, index END.
++
++** Required vs. allowed side-effects
++====================================
++LIST-SORT! and LIST-STABLE-SORT! are allowed, but not required,
++to alter their arguments' cons cells to construct the result list. This is
++consistent with the what-not-how character of the group of procedures
++to which they belong (the "sort-lib" package).
++
++The LIST-DELETE-NEIGHBOR-DUPS!, LIST-MERGE! and LIST-MERGE-SORT! procedures,
++on the other hand, provide specific algorithms, and, as such, explicitly
++commit to the use of side-effects on their input lists in order to guarantee
++their key algorithmic properties (e.g., linear-time operation, constant-space
++stack use).
++
++-------------------------------------------------------------------------------
++* Procedure specification
++-------------------------
++The procedures are split into several packages. In a Scheme system that has a
++module or package system, these procedures should be contained in modules
++named as follows:
++ Package name Functionality
++ ------------ -------------
++ sort-lib General sorting for lists & vectors
++ sorted?-lib Sorted predicates for lists & vectors
++ list-merge-sort-lib List merge sort
++ vector-merge-sort-lib Vector merge sort
++ vector-heap-sort-lib Vector heap sort
++ vector-quick-sort-lib Vector quick sort
++ vector-insert-sort-lib Vector insertion sort
++ delndup-lib List and vector delete neighbor duplicates
++
++A Scheme system without a module system should provide all of the bindings
++defined in all of these modules as components of the "SRFI-32" package.
++
++Note that there is no list insert sort package, as you might as well always
++use list merge sort. The reference implementation's destructive list merge
++sort will do fewer SET-CDR!s than a destructive insert sort.
++
++** Procedure naming and functionality
++=====================================
++Almost all of the procedures described below are variants of two basic
++operations: sorting and merging. These procedures are consistently named
++by composing a set of basic lexemes to indicate what they do.
++
++Lexeme Meaning
++------ -------
++"sort" The procedure sorts its input data set by some < comparison function.
++
++"merge" The procedure merges two ordered data sets into a single ordered
++ result.
++
++"stable" This lexeme indicates that the sort is a stable one.
++
++"vector" The procedure operates upon vectors.
++
++"list" The procedure operates upon lists.
++
++"!" Procedures that end in "!" are allowed, and sometimes required,
++ to reuse their input storage to construct their answer.
++
++** Types of parameters and return values
++========================================
++In the procedures specified below,
++ - A LIS parameter is a list;
++
++ - A V parameter is a vector;
++
++ - A < or = parameter is a procedure accepting two arguments taken from the
++ specified procedure's data set(s), and returning a boolean;
++
++ - START and END parameters are exact, non-negative integers that
++ serve as vector indices selecting a subrange of some associated vector.
++ When specified, they must satisfy the relation
++ 0 <= start <= end <= (vector-length v)
++ where V is the associated vector.
++
++Passing values to procedures with these parameters that do not satisfy these
++types is an error.
++
++If a procedure is said to return "unspecified," this means that nothing at all
++is said about what the procedure returns, not even the number of return
++values. Such a procedure is not even required to be consistent from call to
++call in the nature or number of its return values. It is simply required to
++return a value (or values) that may be passed to a command continuation, e.g.
++as the value of an expression appearing as a non-terminal subform of a BEGIN
++expression. Note that in R5RS, this restricts such a procedure to returning a
++single value; non-R5RS systems may not even provide this restriction.
++
++** sort-lib - general sorting package
++=====================================
++This library provides basic sorting and merging functionality suitable for
++general programming. The procedures are named by their semantic properties,
++i.e., what they do to the data (sort, stable sort, merge, and so forth).
++
++ Procedure Suggested algorithm
++ -------------------------------------------------------------------------
++ list-sorted? lis < -> boolean
++ list-merge lis1 lis2 < -> list
++ list-merge! lis1 lis2 < -> list
++ list-sort lis < -> list (vector heap or quick)
++ list-sort! lis < -> list (list merge sort)
++ list-stable-sort lis < -> list (vector merge sort)
++ list-stable-sort! lis < -> list (list merge sort)
++ list-delete-neighbor-dups lis = -> list
++ list-delete-neighbor-dups! lis = -> list
++
++ vector-sorted? v < [start end] -> boolean
++ vector-merge v1 v2 < [start1 end1 start2 end2] -> vector
++ vector-merge! v v1 v2 < [start start1 end1 start2 end2] -> unspecific
++ vector-sort v < [start end] -> vector (heap or quick sort)
++ vector-sort! v < [start end] -> unspecific (heap or quick sort)
++ vector-stable-sort v < [start end] -> vector (vector merge sort)
++ vector-stable-sort! v < [start end] -> unspecific (vector merge sort)
++ vector-delete-neighbor-dups v = [start end] -> vector
++ vector-delete-neighbor-dups! v = [start end] -> end'
++
++ LIST-SORTED? and VECTOR-SORTED? return true if their input list or vector
++ is in sorted order, as determined by their < comparison parameter.
++
++ All four merge operations are stable: an element of the initial list LIS1
++ or vector V1 will come before an equal-comparing element in the second
++ list LIS2 or vector V2 in the result.
++
++ The procedures
++ LIST-MERGE
++ LIST-SORT
++ LIST-STABLE-SORT
++ LIST-DELETE-NEIGHBOR-DUPS
++ do not alter their inputs and are allowed to return a value that shares
++ a common tail with a list argument.
++
++ The procedures
++ LIST-SORT!
++ LIST-STABLE-SORT!
++ are "linear update" operators -- they are allowed, but not required, to
++ alter the cons cells of their arguments to produce their results.
++
++ On the other hand, the procedures
++ LIST-DELETE-NEIGHBOR-DUPS!
++ LIST-MERGE!
++ make only a single, iterative, linear-time pass over their argument lists,
++ using SET-CDR!s to rearrange the cells of the lists into the final result
++ -- they work "in place." Hence, any cons cell appearing in the result must
++ have originally appeared in an input. The intent of this
++ iterative-algorithm commitment is to allow the programmer to be sure that
++ if, for example, LIST-MERGE! is asked to merge two ten-million-element
++ lists, the operation will complete without performing some extremely
++ (possibly twenty-million) deep recursion.
++
++ The vector procedures
++ VECTOR-SORT
++ VECTOR-STABLE-SORT
++ VECTOR-DELETE-NEIGHBOR-DUPS
++ do not alter their inputs, but allocate a fresh vector for their result,
++ of length END-START.
++
++ The vector procedures
++ VECTOR-SORT!
++ VECTOR-STABLE-SORT!
++ sort their data in-place. (But note that VECTOR-STABLE-SORT! may
++ allocate temporary storage proportional to the size of the input --
++ I am not aware of O(n lg n) stable vector sorting algorithms that
++ run in constant space.)
++
++ VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
++
++ VECTOR-MERGE! writes its result into vector V, beginning at index START0,
++ for indices less than END0 = START0 + (END1-START1) + (END2-START2). The
++ target subvector
++ V[start0,end0)
++ may not overlap either source subvector
++ V1[start1,end1)
++ V2[start2,end2).
++
++ The DELETE-NEIGHBOR-DUP-... procedures:
++ These procedures delete adjacent duplicate elements from a list or a
++ vector, using a given element-equality procedure. The first/leftmost
++ element of a run of equal elements is the one that survives. The list or
++ vector is not otherwise disordered.
++
++ These procedures are linear time -- much faster than the O(n^2) general
++ duplicate-element deletors that do not assume any "bunching" of elements
++ (such as the ones provided by SRFI-1). If you want to delete duplicate
++ elements from a large list or vector, sort the elements to bring equal
++ items together, then use one of these procedures, for a total time of
++ O(n lg n).
++
++ The comparison function = passed to these procedures is always applied
++ (= x y)
++ where X comes before Y in the containing list or vector.
++
++ - LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
++ may share storage with the input list.
++
++ - VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
++ rather allocates a fresh vector to hold the result.
++
++ - LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
++ mutate its input list in order to construct its answer.
++
++ - VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
++ answer, packing its answer into the index range [start,end'), where
++ END' is the non-negative exact integer returned as its value. It
++ returns END' as its result. The vector is not altered outside the range
++ [start,end').
++
++ [Maybe this procedure should take a "target" vector to write?]
++
++ - Examples:
++ (list-delete-neighbor-dups '(1 1 2 7 7 7 0 -2 -2) =)
++ => (1 2 7 0 -2)
++
++ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) =)
++ => #(1 2 7 0 -2)
++
++ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) = 3 7)
++ => #(7 0 -2)
++
++ ;; Result left in v[3,9):
++ (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
++ (cons (vector-delete-neighbor-dups! v = 3)
++ v))
++ => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
++
++
++** Algorithm-specific sorting packages
++======================================
++These packages provide more specific sorting functionality, that is,
++specific committment to particular algorithms that have particular
++pragmatic consequences (such as memory locality, asymptotic running time)
++beyond their semantic behaviour (sorting, stable sorting, merging, etc.).
++Programmers that need a particular algorithm can use one of these packages.
++
++sorted?-lib - sorted predicates
++ list-sorted? lis < -> boolean
++ vector-sorted? v < [start end] -> boolean
++
++ Return #f iff there is an adjacent pair ... X Y ... in the input
++ list or vector such that Y < X. The optional START/END range
++ arguments restrict VECTOR-SORTED? to the indicated subvector.
++
++list-merge-sort-lib - list merge sort
++ list-merge-sort lis < -> list
++ list-merge-sort! lis < -> list
++ list-merge lis1 lis2 < -> list
++ list-merge! lis1 lis2 < -> list
++
++ The sort procedures sort their data using a list merge sort, which is
++ stable. (The reference implementation is, additionally, a "natural" sort.
++ See below for the properties of this algorithm.)
++
++ The ! procedures are destructive -- they use SET-CDR!s to rearrange the
++ cells of the lists into the proper order. As such, they do not allocate
++ any extra cons cells -- they are "in place" sorts. Additionally,
++ LIST-MERGE! is iterative, not recursive -- it can operate on arguments of
++ arbitrary size without requiring an unbounded amount of stack space.
++
++ The merge operations are stable: an element of LIS1 will come before an
++ equal-comparing element in LIS2 in the result list.
++
++vector-merge-sort-lib - vector merge sort
++ vector-merge-sort v < [start end temp] -> vector
++ vector-merge-sort! v < [start end temp] -> unspecific
++ vector-merge v1 v2 < [start1 end1 start2 end2] -> vector
++ vector-merge! v v1 v2 < [start0 start1 end1 start2 end2] -> unspecific
++
++ The sort procedures sort their data using vector merge sort, which is
++ stable. (The reference implementation is, additionally, a "natural" sort.
++ See below for the properties of this algorithm.)
++
++ The optional START/END arguments provide for sorting of subranges, and
++ default to 0 and the length of the corresponding vector.
++
++ Merge-sorting a vector requires the allocation of a temporary "scratch"
++ work vector for the duration of the sort. This scratch vector can be
++ passed in by the client as the optional TEMP argument; if so, the supplied
++ vector must be of size >= END, and will not be altered outside the range
++ [start,end). If not supplied, the sort routines allocate one themselves.
++
++ The merge operations are stable: an element of V1 will come before an
++ equal-comparing element in V2 in the result vector.
++
++ VECTOR-MERGE-SORT! leaves its result in V[start,end).
++
++ VECTOR-MERGE-SORT returns a vector of length END-START.
++
++ VECTOR-MERGE returns a vector of length (END1-START1)+(END2-START2).
++
++ VECTOR-MERGE! writes its result into vector V, beginning at index START0,
++ for indices less than END0 = START0 + (END1-START1) + (END2-START2). The
++ target subvector
++ V[start0,end0)
++ may not overlap either source subvector
++ V1[start1,end1)
++ V2[start2,end2).
++
++vector-heap-sort-lib - vector heap sort
++ heap-sort v < [start end] -> vector
++ heap-sort! v < [start end] -> unspecific
++
++ These procedures sort their data using heap sort,
++ which is not a stable sorting algorithm.
++
++ HEAP-SORT returns a vector of length END-START.
++ HEAP-SORT! is in-place, leaving its result in V[start,end).
++
++vector-quick-sort-lib - vector quick sort
++ quick-sort v < [start end] -> vector
++ quick-sort! v < [start end] -> unspecific
++
++ These procedures sort their data using quick sort,
++ which is not a stable sorting algorithm.
++
++ QUICK-SORT returns a vector of length END-START.
++ QUICK-SORT! is in-place, leaving its result in V[start,end).
++
++vector-insert-sort-lib - vector insertion sort
++ insert-sort v < [start end] -> vector
++ insert-sort! v < [start end] -> unspecific
++
++ These procedures stably sort their data using insertion sort.
++
++ INSERT-SORT returns a vector of length END-START.
++ INSERT-SORT! is in-place, leaving its result in V[start,end).
++
++delndup-lib - list and vector delete neighbor duplicates
++ list-delete-neighbor-dups lis = -> list
++ list-delete-neighbor-dups! lis = -> list
++
++ vector-delete-neighbor-dups v = [start end] -> vector
++ vector-delete-neighbor-dups! v = [start end] -> end'
++
++ These procedures delete adjacent duplicate elements from a list or
++ a vector, using a given element-equality procedure =. The first/leftmost
++ element of a run of equal elements is the one that survives. The list
++ or vector is not otherwise disordered.
++
++ These procedures are linear time -- much faster than the O(n^2) general
++ duplicate-element deletors that do not assume any "bunching" of elements
++ (such as the ones provided by SRFI-1). If you want to delete duplicate
++ elements from a large list or vector, sort the elements to bring equal
++ items together, then use one of these procedures, for a total time of
++ O(n lg n).
++
++ The comparison function = passed to these procedures is always applied
++ (= x y)
++ where X comes before Y in the containing list or vector.
++
++ LIST-DELETE-NEIGHBOR-DUPS does not alter its input list; its answer
++ may share storage with the input list.
++
++ VECTOR-DELETE-NEIGHBOR-DUPS does not alter its input vector, but
++ rather allocates a fresh vector to hold the result.
++
++ LIST-DELETE-NEIGHBOR-DUPS! is permitted, but not required, to
++ mutate its input list in order to construct its answer.
++
++ VECTOR-DELETE-NEIGHBOR-DUPS! reuses its input vector to hold the
++ answer, packing its answer into the index range [start,end'), where
++ END' is the non-negative exact integer returned as its value. It
++ returns END' as its result. The vector is not altered outside the range
++ [start,end').
++
++ Examples:
++ (list-delete-neighbor-dups '(1 1 2 7 7 7 0 -2 -2) =)
++ => (1 2 7 0 -2)
++
++ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) =)
++ => #(1 2 7 0 -2)
++
++ (vector-delete-neighbor-dups '#(1 1 2 7 7 7 0 -2 -2) = 3 7)
++ => #(7 0 -2)
++
++ ;; Result left in v[3,9):
++ (let ((v (vector 0 0 0 1 1 2 2 3 3 4 4 5 5 6 6)))
++ (cons (vector-delete-neighbor-dups! v = 3)
++ v))
++ => (9 . #(0 0 0 1 2 3 4 5 6 4 4 5 5 6 6))
++
++
++-------------------------------------------------------------------------------
++* Algorithmic properties
++------------------------
++Different sort and merge algorithms have different properties.
++Choose the algorithm that matches your needs:
++
++Vector insert sort
++ Stable, but only suitable for small vectors -- O(n^2).
++
++Vector quick sort
++ Not stable. Is fast on average -- O(n lg n) -- but has bad worst-case
++ behaviour. Has good memory locality for big vectors (unlike heap sort).
++ A clever pivot-picking trick (median of three samples) helps avoid
++ worst-case behaviour, but pathological cases can still blow up.
++
++Vector heap sort
++ Not stable. Guaranteed fast -- O(n lg n) *worst* case. Poor locality
++ on large vectors. A very reliable workhorse.
++
++Vector merge sort
++ Stable. Not in-place -- requires a temporary buffer of equal size.
++ Fast -- O(n lg n) -- and has good memory locality for large vectors.
++
++ The implementation of vector merge sort provided by this SRFI's reference
++ implementation is, additionally, a "natural" sort, meaning that it
++ exploits existing order in the input data, providing O(n) best case.
++
++Destructive list merge sort
++ Stable, fast and in-place (i.e., allocates no new cons cells). "Fast"
++ means O(n lg n) worse-case, and substantially better if the data
++ is already mostly ordered, all the way down to linear time for
++ a completely-ordered input list (i.e., it is a "natural" sort).
++
++ Note that sorting lists involves chasing pointers through memory, which
++ can be a loser on modern machine architectures because of poor cache &
++ page locality. Pointer *writing*, which is what the SET-CDR!s of a
++ destructive list-sort algorithm do, is even worse, especially if your
++ Scheme has a generational GC -- the writes will thrash the write-barrier.
++ Sorting vectors has inherently better locality.
++
++ This SRFIs destructive list merge and merge sort implementations are
++ opportunistic -- they avoid redundant SET-CDR!s, and try to take long
++ already-ordered runs of list structure as-is when doing the merges.
++
++Pure list merge sort
++ Stable and fast -- O(n lg n) worst-case, and possibly better, depending
++ upon the input list (see above).
++
++
++Algorithm Stable? Worst case Average case In-place
++------------------------------------------------------
++V insert Yes O(n^2) O(n^2) Yes
++V quick No O(n^2) O(n lg n) Yes
++V heap No O(n lg n) O(n lg n) Yes
++V merge Yes O(n lg n) O(n lg n) No
++L merge Yes O(n lg n) O(n lg n) Either
++
++
++-------------------------------------------------------------------------------
++* Topics to be resolved during discussion phase
++-----------------------------------------------
++I particularly solicit comments about the following topics.
++
++- Include VECTOR-BINARY-SEARCH ?
++ Should we include
++ (VECTOR-BINARY-SEARCH v key< elt->key key [start end])
++ in the SRFI? It sort of goes with sorting; it's exactly ten lines of code.
++
++- Comparison function before or after the list/vector argument?
++ Should it be
++ (list-sort < lis)
++ or
++ (list-sort lis <)
++ There is overwhelming consistency among the implementations: data first,
++ < after. Only Chez does it differently.
++
++ I have done it in the backwards-compatible way. But I prefer the < first,
++ data after way.
++
++
++-------------------------------------------------------------------------------
++* Porting and optimisation
++--------------------------
++This package should be trivial to port. There are only four non-R4RS bits
++in the code:
++- Use of multiple-value return, with the R5RS VALUES procedure, and the
++ simple (RECEIVE (var ...) mv-exp body ...) multiple-value binding macro.
++
++- A VECTOR-COPY procedure. This is a tiny little procedure:
++ (vector-copy v [start end])
++
++- Use of the LET-OPTIONALS macro from scsh to parse and default optional
++ arguments to three routines. Again, easy to port the macro or rewrite
++ the code to parse, default, and error check the args by hand.
++
++- Calls to an ERROR function for complaining about bad arguments.
++
++This code is tightly bummed, as far as I can go in portable Scheme.
++
++You could speed up the vector code a lot by error-checking the procedure
++parameters and then shifting over to fixnum-specific arithmetic and
++dangerous vector-indexing and vector-setting primitives. The comments
++in the code indicate where the initial error checks would have to be
++added. There are several (QUOTIENT N 2)'s that could be changed to a
++fixnum right-shift, as well, in both the list and vector code. The code
++is designed to enable this -- each file usually exports one or two "safe"
++procedures that end up calling an internal "dangerous" primitive. The
++little exported cover procedures are where you move the error checks.
++
++This should provide *big* speedups. In fact, all the code bumming I've done
++pretty much disappears in the noise unless you have a good compiler and also
++can dump the vector-index checks and generic arithmetic -- so I've really just
++set things up for you to exploit.
++
++The optional-arg parsing, defaulting, and error checking is done with a
++portable R4RS macro. But if your Scheme has a faster mechanism (e.g., Chez),
++you should definitely port over to it. Note that argument defaulting and
++error-checking are interleaved -- you don't have to error-check defaulted
++START/END args to see if they are fixnums that are legal vector indices for
++the corresponding vector, etc.
++
++
++-------------------------------------------------------------------------------
++* References & Links
++--------------------
++
++This document, in HTML:
++ http://srfi.schemers.org/srfi-32/srfi-32.html
++ [This link may not be valid while the SRFI is in draft form.]
++
++This document, in simple text format:
++ http://srfi.schemers.org/srfi-32/srfi-32.txt
++
++Archive of SRFI-32 discussion-list email:
++ http://srfi.schemers.org/srfi-32/mail-archive/maillist.html
++
++SRFI web site:
++ http://srfi.schemers.org/
++
++[CommonLisp]
++ Common Lisp: the Language
++ Guy L. Steele Jr. (editor).
++ Digital Press, Maynard, Mass., second edition 1990.
++ Available at http://www.elwood.com/alu/table/references.htm#cltl2
++
++ The Common Lisp "HyperSpec," produced by Kent Pitman, is essentially
++ the ANSI spec for Common Lisp:
++ http://www.xanalys.com/software_tools/reference/HyperSpec/
++
++[R5RS]
++ Revised^5 Report on the Algorithmic Language Scheme,
++ R. Kelsey, W. Clinger, J. Rees (editors).
++ Higher-Order and Symbolic Computation, Vol. 11, No. 1, September, 1998.
++ and ACM SIGPLAN Notices, Vol. 33, No. 9, October, 1998.
++
++ Available at http://www.schemers.org/Documents/Standards/
++
++
++-------------------------------------------------------------------------------
++* Acknowledgements
++------------------
++
++I thank the authors of the open source I consulted when designing this
++library, particularly Richard O'Keefe, Donovan Kolby and the MIT Scheme Team.
++
++
++-------------------------------------------------------------------------------
++* Copyright
++-----------
++
++** SRFI text
++============
++This document is copyright (C) Olin Shivers (1998, 1999).
++All Rights Reserved.
++
++Permission is hereby granted, free of charge, to any person obtaining
++a copy of this software and associated documentation files (the
++"Software"), to deal in the Software without restriction, including
++without limitation the rights to use, copy, modify, merge, publish,
++distribute, sublicense, and/or sell copies of the Software, and to
++permit persons to whom the Software is furnished to do so, subject to
++the following conditions:
++
++The above copyright notice and this permission notice shall be
++included in all copies or substantial portions of the Software.
++
++THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
++EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
++MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
++NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
++LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
++OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
++WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
++
++** Reference implementation
++===========================
++Short summary: no restrictions.
++
++While I wrote all of this code myself, I read a lot of code before I began
++writing. However, all such code is, itself, either open source or public
++domain, rendering irrelevant any issue of "copyright taint."
++
++The natural merge sorts (pure list, destructive list, and vector) are not only
++my own code, but are implementations of an algorithm of my own devising. They
++run in O(n lg n) worst case, O(n) best case, and require only a logarithmic
++number of stack frames. And they are stable. And the destructive-list variant
++allocates zero cons cells; it simply rearranges the cells of the input list.
++
++Hence the reference implementation is
++ Copyright (c) 1998 by Olin Shivers.
++and made available under the same copyright as the SRFI text (see above).
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib/srfi/%3a5/let.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5/let.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1 +0,0 @@
+-#lang s-exp srfi/provider srfi/%3a5
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib/srfi/%3a5.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/%3a5.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,3 +0,0 @@
+-#lang s-exp srfi/provider srfi/5
+-
+-;; FIXME: "rest arguments" need to generate mutable lists
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib/srfi/5/let.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/5/let.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,86 +0,0 @@
+-;;;
+-;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
+-;;; Time-stamp: <03/04/08 09:56:06 solsona>
+-;;;
+-;;; Usually, I would add a copyright notice, and the announce that
+-;;; this code is under the LGPL licence. Nevertheless, I only did the
+-;;; port to PLT Scheme v200, and here is the copyright notice,
+-;;; comments, and licence from the original source:
+-;;;
+-;;; Copyright (C) Andy Gaynor (1999-2003)
+-;;;
+-;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
+-;;; Renamed to s:let by Eli Barzilay
+-
+-#lang scheme/base
+-(provide s:let)
+-
+-(define-syntax s:let
+- (syntax-rules ()
+- ;; standard
+- ((s:let () body ...)
+- (let () body ...))
+- ((s:let ((var val) ...) body ...)
+- (let ((var val) ...) body ...))
+-
+- ;; rest style
+- ((s:let ((var val) . bindings) body ...)
+- (let-loop #f bindings (var) (val) (body ...)))
+-
+- ;; signature style
+- ((s:let (name bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ;; standard named style
+- ((s:let name (bindings ...) body ...)
+- (let-loop name (bindings ...) () () (body ...)))
+-
+- ))
+-
+-;; A loop to walk down the list of bindings.
+-
+-(define-syntax let-loop
+- (syntax-rules ()
+-
+- ;; No more bindings - make a LETREC.
+- ((let-loop name () (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ...) . body)))
+- name)
+- vals ...))
+-
+- ;; Rest binding, no name
+- ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
+- (let ((var val) ... (rest-var (list rest-val ...))) . body))
+-
+- ;; Process a (var val) pair.
+- ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
+- (let-loop name (more ...) (vars ... var) (vals ... val) body))
+-
+- ;; End with a rest variable - make a LETREC.
+- ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
+- ((letrec ((name (lambda (vars ... . rest-var) . body)))
+- name)
+- vals ... rest-vals ...))))
+-
+-;; Four loops - normal and `signature-style', each with and without a rest
+-;; binding.
+-;;
+-;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
+-;; (if (= i n)
+-;; f0
+-;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
+-;;
+-;;(let fibonacci ((n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+-;;
+-;;(let (fibonacci (n 10) (i 0) . (f 0 1))
+-;; (if (= i n)
+-;; (car f)
+-;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib/srfi/5.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib/srfi/5.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
+@@ -1,2 +0,0 @@
+-;; module loader for SRFI-5
+-#lang s-exp srfi/provider srfi/5/let #:unprefix s:
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/info.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/info.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/info.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1,14 @@
++#lang info
++
++(define collection 'multi)
++
++(define deps '("scheme-lib"
++ "base"
++ "r6rs-lib"
++ "srfi-lib"
++ "compatibility-lib"))
++
++
++(define pkg-desc "implementation (no documentation) part of \"srfi nonfree\"")
++
++(define pkg-authors '(mflatt noel chongkai jay))
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5/let.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1 @@
++#lang s-exp srfi/provider srfi/%3a5
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/%3a5.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1,3 @@
++#lang s-exp srfi/provider srfi/5
++
++;; FIXME: "rest arguments" need to generate mutable lists
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5/let.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1,86 @@
++;;;
++;;; <let.rkt> ---- SRFI 5 A compatible let form with signatures and rest arguments
++;;; Time-stamp: <03/04/08 09:56:06 solsona>
++;;;
++;;; Usually, I would add a copyright notice, and the announce that
++;;; this code is under the LGPL licence. Nevertheless, I only did the
++;;; port to PLT Scheme v200, and here is the copyright notice,
++;;; comments, and licence from the original source:
++;;;
++;;; Copyright (C) Andy Gaynor (1999-2003)
++;;;
++;;; The version of my-let here was cleaned up by: Paul Schlie <schlie(a)attbi.com>.
++;;; Renamed to s:let by Eli Barzilay
++
++#lang scheme/base
++(provide s:let)
++
++(define-syntax s:let
++ (syntax-rules ()
++ ;; standard
++ ((s:let () body ...)
++ (let () body ...))
++ ((s:let ((var val) ...) body ...)
++ (let ((var val) ...) body ...))
++
++ ;; rest style
++ ((s:let ((var val) . bindings) body ...)
++ (let-loop #f bindings (var) (val) (body ...)))
++
++ ;; signature style
++ ((s:let (name bindings ...) body ...)
++ (let-loop name (bindings ...) () () (body ...)))
++
++ ;; standard named style
++ ((s:let name (bindings ...) body ...)
++ (let-loop name (bindings ...) () () (body ...)))
++
++ ))
++
++;; A loop to walk down the list of bindings.
++
++(define-syntax let-loop
++ (syntax-rules ()
++
++ ;; No more bindings - make a LETREC.
++ ((let-loop name () (vars ...) (vals ...) body)
++ ((letrec ((name (lambda (vars ...) . body)))
++ name)
++ vals ...))
++
++ ;; Rest binding, no name
++ ((let-loop #f (rest-var rest-val ...) (var ...) (val ...) body)
++ (let ((var val) ... (rest-var (list rest-val ...))) . body))
++
++ ;; Process a (var val) pair.
++ ((let-loop name ((var val) more ...) (vars ...) (vals ...) body)
++ (let-loop name (more ...) (vars ... var) (vals ... val) body))
++
++ ;; End with a rest variable - make a LETREC.
++ ((let-loop name (rest-var rest-vals ...) (vars ...) (vals ...) body)
++ ((letrec ((name (lambda (vars ... . rest-var) . body)))
++ name)
++ vals ... rest-vals ...))))
++
++;; Four loops - normal and `signature-style', each with and without a rest
++;; binding.
++;;
++;;(let fibonacci ((n 10) (i 0) (f0 0) (f1 1))
++;; (if (= i n)
++;; f0
++;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
++;;
++;;(let (fibonacci (n 10) (i 0) (f0 0) (f1 1))
++;; (if (= i n)
++;; f0
++;; (fibonacci n (+ i 1) f1 (+ f0 f1))))
++;;
++;;(let fibonacci ((n 10) (i 0) . (f 0 1))
++;; (if (= i n)
++;; (car f)
++;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
++;;
++;;(let (fibonacci (n 10) (i 0) . (f 0 1))
++;; (if (= i n)
++;; (car f)
++;; (fibonacci n (+ i 1) (cadr f) (+ (car f) (cadr f)))))
+diff -urN racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 1969-12-31 19:00:00.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lib-nonfree/srfi/5.rkt 2018-01-31 12:22:58.159501944 -0500
+@@ -0,0 +1,2 @@
++;; module loader for SRFI-5
++#lang s-exp srfi/provider srfi/5/let #:unprefix s:
+diff -urN racket-6.12-orig/share/pkgs/srfi-lite-lib/info.rkt racket-6.12/share/pkgs/srfi-lite-lib/info.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lite-lib/info.rkt 2018-01-26 16:10:04.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lite-lib/info.rkt 2018-01-31 12:22:58.160501940 -0500
+@@ -1 +1,9 @@
+-(module info setup/infotab (#%module-begin (define package-content-state (quote (built "6.12"))) (define collection (quote multi)) (define deps (quote ("base"))) (define pkg-desc "implementation of the most widely used \"srfi\" libraries") (define pkg-authors (quote (mflatt)))))
++#lang info
++
++(define collection 'multi)
++
++(define deps '("base"))
++
++(define pkg-desc "implementation of the most widely used \"srfi\" libraries")
++
++(define pkg-authors '(mflatt))
+Binary files racket-6.12-orig/share/pkgs/srfi-lite-lib/.info.rkt.swp and racket-6.12/share/pkgs/srfi-lite-lib/.info.rkt.swp differ
+diff -urN racket-6.12-orig/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt racket-6.12/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt
+--- racket-6.12-orig/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt 2018-01-26 15:34:52.000000000 -0500
++++ racket-6.12/share/pkgs/srfi-lite-lib/srfi/29/localization.rkt 2018-01-31 12:22:58.162501933 -0500
+@@ -2,6 +2,7 @@
+
+ (require racket/contract/base
+ racket/file
++ racket/list
+ (only-in racket/runtime-path define-runtime-path)
+ racket/string racket/format
+ syntax/modread
+@@ -113,10 +114,6 @@
+ (lambda ()
+ (with-input-from-file path read))))
+ #t))))
+- (define (rdc ls)
+- (if (null? (cdr ls))
+- '()
+- (cons (car ls) (rdc (cdr ls)))))
+
+ ;;Retrieve a localized template given its package name and a template name
+ (define (localized-template package-name template-name)
+@@ -127,5 +124,5 @@
+ (let ((bundle (hash-ref *localization-bundles* specifier #f)))
+ (cond ((and bundle (assq template-name bundle)) => cdr)
+ ((null? (cdr specifier)) #f)
+- (else (loop (rdc specifier))))))))
++ (else (loop (drop-right specifier 1))))))))
+ )
diff --git a/racket.spec b/racket.spec
new file mode 100644
index 0000000..d3dc907
--- /dev/null
+++ b/racket.spec
@@ -0,0 +1,369 @@
+Name: racket
+Version: 6.12
+Release: 8%{?dist}
+Summary: General purpose programming language
+
+License: GPLv3 and LGPLv3 and MIT
+URL: https://racket-lang.org
+Source0: https://mirror.racket-lang.org/installers/%{version}/%{name}-%{version}-src…
+
+# Update SRFI libraries to include upstream PR 5.
+# See: https://github.com/racket/srfi/pull/5 (merged)
+Patch0: racket-6.12-update-srfi.patch
+
+# Remove SRFI library and docs with restrictive licensing.
+# See: https://github.com/racket/srfi/issues/4 (open)
+# Note: Upstream maintainers have confirmed this
+# is safe, since the removed components are
+# extra elements which nothing else in the
+# package depends on.
+Patch1: racket-6.12-remove-nonfree.patch
+
+# Backport upstream fix of rpaths in racket/racket
+# which break check-buildroot
+# https://github.com/racket/racket/pull/1947 (merged)
+Patch2: racket-6.12-fix-rpaths.patch
+
+# Backport upstream fix of rpaths in racket/web-server
+# which break check-buildroot
+# https://github.com/racket/web-server/issues/36 (closed)
+Patch3: racket-6.12-fix-webserver-rpaths.patch
+
+# Update racket-doc/search to use https://docs.racket-lang.org
+# when no docs exist locally
+# https://github.com/racket/scribble/pull/164 (open)
+Patch4: racket-6.12-fix-doc-open-url.patch
+
+# Issue Building for s390x and armv7hl in koji
+ExcludeArch: s390x armv7hl
+
+# To compile the program
+BuildRequires: gcc
+
+# To fix rpath issue with executables.
+BuildRequires: chrpath
+
+# Racket heavily utilizes the system ffi library.
+BuildRequires: libffi-devel
+
+# For the racket/gui library (via libffi)
+# https://github.com/racket/gui/blob/master/gui-lib/mred/private/wx/gtk/gtk3.…
+BuildRequires: gtk3
+
+# For the racket/draw library (via libffi)
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/cair…
+BuildRequires: cairo
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/pang…
+BuildRequires: pango
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/png.…
+BuildRequires: libpng
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/jpeg…
+BuildRequires: libjpeg-turbo
+# https://github.com/racket/draw/blob/master/draw-lib/racket/draw/unsafe/glib…
+BuildRequires: glib2
+
+# To validate desktop file
+BuildRequires: desktop-file-utils
+
+BuildRequires: git
+
+# Require the subpackages
+Requires: racket-minimal%{?_isa} = %{version}-%{release}
+Requires: racket-pkgs = %{version}-%{release}
+
+%description
+Racket is a general-purpose programming language as well as
+the world's first ecosystem for developing and deploying new
+languages. Make your dream language, or use one of the dozens
+already available.
+
+%prep
+%autosetup -v -p1
+# Remove bundled libffi
+rm -r src/foreign/libffi
+
+%build
+cd src
+%configure \
+ --enable-pthread \
+ --enable-shared \
+ --enable-libffi \
+ --disable-strip
+%make_build
+
+%install
+cd src
+%make_install
+
+# Delete mred binaries and replace them with links.
+rm -vf ${RPM_BUILD_ROOT}%{_bindir}/mred
+rm -vf ${RPM_BUILD_ROOT}%{_bindir}/mred-text
+ln -vs %{_bindir}/gracket ${RPM_BUILD_ROOT}%{_bindir}/mred
+ln -vs %{_bindir}/gracket-text ${RPM_BUILD_ROOT}%{_bindir}/mred-text
+
+# Delete static library. Apperently --disable-libs does not stop it.
+rm -vf ${RPM_BUILD_ROOT}%{_libdir}/libracket3m.a
+
+# Delete duplicate license files
+rm -rf %{buildroot}%{_datadir}/racket/COPYING*txt
+
+# Fix the rpath error.
+chrpath --delete ${RPM_BUILD_ROOT}%{_bindir}/racket
+chrpath --delete ${RPM_BUILD_ROOT}%{_libdir}/racket/gracket
+
+# Remove the libtool files.
+rm -f ${RPM_BUILD_ROOT}%{_libdir}/*.la
+
+# Fix paths in the desktop files.
+sed -i "s#${RPM_BUILD_ROOT}##g" \
+ ${RPM_BUILD_ROOT}/%{_datadir}/applications/*.desktop
+
+# Validate desktop files
+desktop-file-validate %{buildroot}/%{_datadir}/applications/*.desktop
+
+# Fix paths in html docs
+DOCS_TO_FIX="
+syntax/module-helpers.html
+rackunit/api.html
+reference/collects.html"
+for i in $DOCS_TO_FIX; do
+ sed -i "s#${RPM_BUILD_ROOT}##g" \
+ ${RPM_BUILD_ROOT}/%{_datadir}/doc/racket/$i
+done
+
+chmod -x ${RPM_BUILD_ROOT}%{_libdir}/racket/starter-sh
+
+%ldconfig_scriptlets
+
+# Equivalent to upstream's minimal-racket release
+%package minimal
+Summary: A minimal Racket installation
+Requires: racket-collects = %{version}-%{release}
+%description minimal
+Racket's core runtime
+
+%package collects
+Summary: Racket's core collections libraries
+BuildArch: noarch
+%description collects
+Libraries providing Racket's core functionality
+
+# Arch independent source and bytecode files
+%package pkgs
+Summary: Racket package collections
+# See BuildRequires section for details on dependencies
+Requires: gtk3
+Requires: cairo
+Requires: pango
+Requires: libpng
+Requires: glib2
+Requires: libjpeg-turbo
+Requires: racket-minimal = %{version}-%{release}
+BuildArch: noarch
+%description pkgs
+Additional packages and libraries for Racket
+
+# Development headers and links
+%package devel
+Summary: Development files for Racket
+Requires: racket-minimal%{?_isa} = %{version}-%{release}
+%description devel
+Files needed to link against Racket.
+
+# HTML documentation
+%package doc
+Summary: Documentation files for Racket
+BuildArch: noarch
+%description doc
+A local installation of the Racket documentation system.
+
+%files
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_bindir}/drracket
+%{_bindir}/gracket
+%{_bindir}/gracket-text
+%{_bindir}/mred-text
+%{_bindir}/mred
+%{_bindir}/mzc
+%{_bindir}/mzpp
+%{_bindir}/mzscheme
+%{_bindir}/mztext
+%{_bindir}/pdf-slatex
+%{_bindir}/plt-games
+%{_bindir}/plt-help
+%{_bindir}/plt-r5rs
+%{_bindir}/plt-r6rs
+%{_bindir}/plt-web-server
+%{_bindir}/scribble
+%{_bindir}/setup-plt
+%{_bindir}/slatex
+%{_bindir}/slideshow
+%{_bindir}/swindle
+%{_datadir}/applications/
+
+%files collects
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_datadir}/racket/collects
+
+%files minimal
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_bindir}/racket
+%{_bindir}/raco
+%{_libdir}/racket
+%{_libdir}/libracket3m-%{version}.so
+%{_datadir}/racket/links.rktd
+%{_datadir}/racket/pkgs/racket-lib
+%{_datadir}/man/man1/racket*
+%{_datadir}/man/man1/raco*
+%dir %{_datadir}/racket
+%dir %{_sysconfdir}/racket/
+%config %{_sysconfdir}/racket/config.rktd
+%exclude %{_libdir}/libracket3m.so
+
+%files pkgs
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_datadir}/racket
+%{_datadir}/man/man1/drracket*
+%{_datadir}/man/man1/gracket*
+%{_datadir}/man/man1/mred*
+%{_datadir}/man/man1/mzc*
+%{_datadir}/man/man1/mzscheme*
+%{_datadir}/man/man1/plt-help*
+%{_datadir}/man/man1/setup-plt*
+%exclude %{_datadir}/racket/links.rktd
+%exclude %dir %{_datadir}/racket/pkgs/racket-lib
+%exclude %dir %{_datadir}/racket/collects
+
+%files devel
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_includedir}/racket
+%{_libdir}/libracket3m.so
+
+%files doc
+%license src/COPYING.txt src/COPYING_LESSER.txt src/COPYING-libscheme.txt
+%{_datadir}/doc/racket
+
+%changelog
+* Mon Jul 30 2018 David Benoit <dbenoit(a)redhat.com> 6.12-8
+- Annotate dependencies with links to source code
+- Move dependencies to racket-pkgs, since they are only used
+ by that subpackage
+- Update mred symbolic links
+- Fix ownership of directories
+- Remove executable bit from starter-sh
+
+* Thu Jul 12 2018 David Benoit <dbenoit(a)redhat.com> 6.12-7
+- Remove hardened build since it is enabled by default
+- Add gcc to BuildRequires
+- Remove wildcards from directory listings in files section
+
+* Fri Apr 13 2018 David Benoit <dbenoit(a)redhat.com> 6.12-6
+- Remove license wildcard and add license field to each subpackage
+
+* Fri Apr 6 2018 David Benoit <dbenoit(a)redhat.com> 6.12-5
+- remove update-database post scripts
+- move libracket3m.so link into -devel
+- add ldconfig_scriptlets after install
+- remove disable debug_package and configure
+ with --disable-strip instead
+- add license to files section and update
+ license header field
+- validate desktop files
+- change ownership of /etc/racket
+- update changelog with release info
+- use specific man directory man/man1/*
+- refactor racket into subpackages
+ racket-minimal, racket-collects, and racket-pkgs
+
+* Wed Apr 4 2018 David Benoit <dbenoit(a)redhat.com> 6.12-4
+- noarch -docs subpackage
+
+* Tue Mar 20 2018 David Benoit <dbenoit(a)redhat.com> 6.12-3
+- fix text encoding issue in description section
+- remove doc-open-url scriptlets
+- add scriptlet to fix paths in html docs
+- add patch2 to backport rpaths fix in compiled .zo files
+- add patch3 to backport rpaths fix in web-server-lib
+- add patch4 to configure doc open url dynamically at runtime
+- remove override of __arch_install_post to allow full
+ checking of buildroot.
+
+* Thu Feb 1 2018 David Benoit <dbenoit(a)redhat.com> - 6.12-2
+- Fix duplication of object files
+- Add version to racket-devel requirements
+- Remove base package as a dependency of racket-doc
+- Remove Groups tag
+
+* Wed Jan 31 2018 David Benoit <dbenoit(a)redhat.com> - 6.12-1
+- Update to current stable version
+- Add patch0 to update SRFIs to latest upstream
+- Add patch1 to remove nonfree SRFI components
+
+* Thu Oct 26 2017 David Benoit <dbenoit(a)redhat.com> - 6.10-1
+- Update to current stable version
+- Remove libedit readline patch (fixed upstream)
+- Break docs into separate package
+- Add scriptlets to set doc-open-url based on
+ whether docs are installed
+- Exclude armv7hl and s390x as target arches
+- Update description to match website
+- Change URLs to use https instead of http
+
+
+* Thu Jul 6 2017 David Benoit <dbenoit(a)redhat.com> - 6.9-1
+- Update to current stable version
+- Patch libedit readline error
+- Remove ExclusiveArch to test all builds in koji
+
+* Fri Jan 22 2016 Brandon Thomas <bthomaszx(a)gmail.com> - 6.3-1
+- Update to current stable version.
+- Updated description to match website.
+- Removed build requirement "racket-packaging".
+- Updated to gtk+3.
+- Let Autoprovides determine provides.
+- Debuginfo package is empty and preventing the package from building.
+- Removed uneeded file copies.
+- Remove possible extra static library.
+
+* Sun Dec 14 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.1.6-1
+- Update to current snapshot to fix match hash-table expander.
+
+* Mon Dec 01 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.1-1
+- Update to current stable version.
+
+* Fri Sep 05 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.5-4
+- Use racket-packaging to capture module dependencies.
+
+* Tue Aug 19 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.5-1
+- Updated to 6.1.0.5
+- Merged the -doc package back in.
+
+* Fri Aug 08 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.4-2
+- Do not alter .zo files, prevent check-buildroot from being run instead.
+
+* Thu Aug 07 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.4-1
+- Updated to 6.1.0.4
+- Split-off -doc package.
+
+* Fri Jul 25 2014 Jan Dvorak <mordae(a)anilinux.org> - 6.1.0.3-4
+- Updated to 6.1.0.3
+- Dropped the unnecessary static library.
+- Dropped mred programs to enable debug package.
+
+* Sat Jun 22 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.5-1
+- Changed to use 5.3.5 version of Racket.
+- Created static package for developers who may need static libraries.
+- Added RPM optimization options to CFLAGS for build.
+- Added macro to use SMP build options in make.
+
+* Thu May 16 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.4-1
+- Changed to use 5.3.4 version of Racket.
+
+* Tue May 14 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.3-3
+- Moved documentation to /usr/doc directory.
+
+* Mon May 13 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.3-2
+- Remove bundled libffi from racket before building program.
+
+* Thu May 9 2013 Daniel E. Wilson <danw(a)bureau-13.org> - 5.3.3-1
+- Initial Revision.
1
0
05 Sep '18
The package rpms/racket.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/racket.git/commit/?id=b8569a44d31ac….
Change:
+%ifarch %{ix86}
Thanks.
Full change:
============
commit b8569a44d31acfcec836302dff9ae178d3957c0b
Author: dbenoit <dbenoit(a)redhat.com>
Date: Wed Sep 5 15:37:42 2018 -0400
fix i686 and remove ppc64
diff --git a/racket.spec b/racket.spec
index 5fb06c9..bb31acd 100644
--- a/racket.spec
+++ b/racket.spec
@@ -1,6 +1,6 @@
Name: racket
Version: 7.0
-Release: 1%{?dist}
+Release: 2%{?dist}
Summary: General purpose programming language
License: GPLv3 and LGPLv3 and MIT
@@ -21,7 +21,8 @@ Patch0: racket-7.0-remove-nonfree.patch
Patch1: racket-7.0-fix-doc-open-url.patch
# Issue Building for s390x and armv7hl in koji
-ExcludeArch: s390x armv7hl
+#ExcludeArch: s390x armv7hl
+ExcludeArch: s390x armv7hl ppc64
# To compile the program
BuildRequires: gcc
@@ -70,11 +71,20 @@ rm -r src/foreign/libffi
%build
cd src
+
+# Disable SSE on i686 until fixed upstream
+# https://github.com/racket/racket/issues/2245
+%ifarch %{ix86}
+ %set_build_flags
+ export CFLAGS=$(echo $CFLAGS | sed -e "s/-mfpmath=sse *//")
+%endif
+
%configure \
--enable-pthread \
--enable-shared \
--enable-libffi \
--disable-strip
+
%make_build
%install
@@ -231,7 +241,13 @@ A local installation of the Racket documentation system.
%{_datadir}/doc/racket
%changelog
-* Fri Aug 3 2018 David Benoit <dbenoit(a)redhat.com> 7.0-1
+* Wed Sep 5 2018 David Benoit <dbenoit(a)redhat.com> 7.0-2
+- Disable SSE math on i686 until issue is fixed upstream
+- Exclude ppc due to issue building Racket v7.0 and
+ arch being deprecated in next release
+
+* Fri Aug 17 2018 David Benoit <dbenoit(a)redhat.com> 7.0-1
+- Update sources to Racket v7.0
- Remove 6.12 patches and add update remove nonfree
srfi patch to 7.0
1
0
Architecture specific change in rpms/webkit2gtk3.git
by githook-noreply＠fedoraproject.org 05 Sep '18
by githook-noreply＠fedoraproject.org 05 Sep '18
05 Sep '18
The package rpms/webkit2gtk3.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/webkit2gtk3.git/commit/?id=d7f13138…
https://src.fedoraproject.org/cgit/rpms/webkit2gtk3.git/commit/?id=026c18c6…
https://src.fedoraproject.org/cgit/rpms/webkit2gtk3.git/commit/?id=9ee7a89d….
Change:
-%ifarch aarch64
+%ifarch s390 s390x %{power64}
+%ifarch s390x %{power64}
Thanks.
Full change:
============
commit d7f13138badbfd9419b1b2dcb8525bb406accf48
Author: Tomas Popela <tpopela(a)redhat.com>
Date: Wed Sep 5 14:30:03 2018 +0200
Enable ld.gold on aarch64
diff --git a/webkit2gtk3.spec b/webkit2gtk3.spec
index 1f62ae0..2b1813b 100644
--- a/webkit2gtk3.spec
+++ b/webkit2gtk3.spec
@@ -173,12 +173,6 @@ rm -rf Source/ThirdParty/qunit/
# is only needed for x86_64.
%global _dwz_max_die_limit_x86_64 250000000
-%ifarch aarch64
-# Use linker flags to reduce memory consumption - on other arches the ld.gold is
-# used and also it doesn't have the --reduce-memory-overheads option
-%global optflags %{optflags} -Wl,--no-keep-memory -Wl,--reduce-memory-overheads
-%endif
-
# Decrease debuginfo even on ix86 because of:
# https://bugs.webkit.org/show_bug.cgi?id=140176
%ifarch s390x %{arm} %{ix86} %{power64} %{mips}
@@ -190,7 +184,6 @@ rm -rf Source/ThirdParty/qunit/
%global optflags %{optflags} -DUSER_AGENT_GTK_DISTRIBUTOR_NAME=\'\\"Fedora\\"\'
%endif
-# Disable ld.gold on aarch64 as the support is in upstream, but not packaged in Fedora.
mkdir -p %{_target_platform}
pushd %{_target_platform}
%cmake \
@@ -198,9 +191,6 @@ pushd %{_target_platform}
-DCMAKE_BUILD_TYPE=Release \
-DENABLE_GTKDOC=ON \
-DENABLE_MINIBROWSER=ON \
-%ifarch aarch64
- -DUSE_LD_GOLD=OFF \
-%endif
%ifarch s390x %{power64}
-DENABLE_JIT=OFF \
-DUSE_SYSTEM_MALLOC=ON \
commit 9ee7a89d4f0f05f4b2deba0335c74766d6e1319e
Author: Tomas Popela <tpopela(a)redhat.com>
Date: Wed Sep 5 13:48:45 2018 +0200
Remove support for s390
diff --git a/webkit2gtk3.spec b/webkit2gtk3.spec
index ff591fd..1f62ae0 100644
--- a/webkit2gtk3.spec
+++ b/webkit2gtk3.spec
@@ -173,7 +173,7 @@ rm -rf Source/ThirdParty/qunit/
# is only needed for x86_64.
%global _dwz_max_die_limit_x86_64 250000000
-%ifarch s390 aarch64
+%ifarch aarch64
# Use linker flags to reduce memory consumption - on other arches the ld.gold is
# used and also it doesn't have the --reduce-memory-overheads option
%global optflags %{optflags} -Wl,--no-keep-memory -Wl,--reduce-memory-overheads
@@ -181,7 +181,7 @@ rm -rf Source/ThirdParty/qunit/
# Decrease debuginfo even on ix86 because of:
# https://bugs.webkit.org/show_bug.cgi?id=140176
-%ifarch s390 s390x %{arm} %{ix86} %{power64} %{mips}
+%ifarch s390x %{arm} %{ix86} %{power64} %{mips}
# Decrease debuginfo verbosity to reduce memory consumption even more
%global optflags %(echo %{optflags} | sed 's/-g /-g1 /')
%endif
@@ -190,8 +190,7 @@ rm -rf Source/ThirdParty/qunit/
%global optflags %{optflags} -DUSER_AGENT_GTK_DISTRIBUTOR_NAME=\'\\"Fedora\\"\'
%endif
-# Disable ld.gold on s390 as it does not have it.
-# Also for aarch64 as the support is in upstream, but not packaged in Fedora.
+# Disable ld.gold on aarch64 as the support is in upstream, but not packaged in Fedora.
mkdir -p %{_target_platform}
pushd %{_target_platform}
%cmake \
@@ -199,10 +198,10 @@ pushd %{_target_platform}
-DCMAKE_BUILD_TYPE=Release \
-DENABLE_GTKDOC=ON \
-DENABLE_MINIBROWSER=ON \
-%ifarch s390 aarch64
+%ifarch aarch64
-DUSE_LD_GOLD=OFF \
%endif
-%ifarch s390 s390x %{power64}
+%ifarch s390x %{power64}
-DENABLE_JIT=OFF \
-DUSE_SYSTEM_MALLOC=ON \
%endif
commit 026c18c6f654ab1d68819ead7edafb85239ea88b
Author: Tomas Popela <tpopela(a)redhat.com>
Date: Wed Sep 5 13:45:07 2018 +0200
Remove support for ppc
diff --git a/webkit2gtk3.spec b/webkit2gtk3.spec
index 62307c6..ff591fd 100644
--- a/webkit2gtk3.spec
+++ b/webkit2gtk3.spec
@@ -181,16 +181,11 @@ rm -rf Source/ThirdParty/qunit/
# Decrease debuginfo even on ix86 because of:
# https://bugs.webkit.org/show_bug.cgi?id=140176
-%ifarch s390 s390x %{arm} %{ix86} ppc %{power64} %{mips}
+%ifarch s390 s390x %{arm} %{ix86} %{power64} %{mips}
# Decrease debuginfo verbosity to reduce memory consumption even more
%global optflags %(echo %{optflags} | sed 's/-g /-g1 /')
%endif
-%ifarch ppc
-# Use linker flag -relax to get WebKit build under ppc(32) with JIT disabled
-%global optflags %{optflags} -Wl,-relax
-%endif
-
%if 0%{?fedora}
%global optflags %{optflags} -DUSER_AGENT_GTK_DISTRIBUTOR_NAME=\'\\"Fedora\\"\'
%endif
@@ -207,7 +202,7 @@ pushd %{_target_platform}
%ifarch s390 aarch64
-DUSE_LD_GOLD=OFF \
%endif
-%ifarch s390 s390x ppc %{power64}
+%ifarch s390 s390x %{power64}
-DENABLE_JIT=OFF \
-DUSE_SYSTEM_MALLOC=ON \
%endif
1
0
Architecture specific change in rpms/webkit2gtk3.git
by githook-noreply＠fedoraproject.org 05 Sep '18
by githook-noreply＠fedoraproject.org 05 Sep '18
05 Sep '18
The package rpms/webkit2gtk3.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/webkit2gtk3.git/commit/?id=b7ddce8c…
https://src.fedoraproject.org/cgit/rpms/webkit2gtk3.git/commit/?id=3c9a469b…
https://src.fedoraproject.org/cgit/rpms/webkit2gtk3.git/commit/?id=e9a4a11d….
Change:
+%ifarch s390 s390x %{power64}
-%ifarch aarch64
+%ifarch s390x %{power64}
Thanks.
Full change:
============
commit 3c9a469b8eaa25167b9581ff35b3706622666092
Author: Tomas Popela <tpopela(a)redhat.com>
Date: Wed Sep 5 14:30:03 2018 +0200
Enable ld.gold on aarch64
diff --git a/webkit2gtk3.spec b/webkit2gtk3.spec
index 1f62ae0..2b1813b 100644
--- a/webkit2gtk3.spec
+++ b/webkit2gtk3.spec
@@ -173,12 +173,6 @@ rm -rf Source/ThirdParty/qunit/
# is only needed for x86_64.
%global _dwz_max_die_limit_x86_64 250000000
-%ifarch aarch64
-# Use linker flags to reduce memory consumption - on other arches the ld.gold is
-# used and also it doesn't have the --reduce-memory-overheads option
-%global optflags %{optflags} -Wl,--no-keep-memory -Wl,--reduce-memory-overheads
-%endif
-
# Decrease debuginfo even on ix86 because of:
# https://bugs.webkit.org/show_bug.cgi?id=140176
%ifarch s390x %{arm} %{ix86} %{power64} %{mips}
@@ -190,7 +184,6 @@ rm -rf Source/ThirdParty/qunit/
%global optflags %{optflags} -DUSER_AGENT_GTK_DISTRIBUTOR_NAME=\'\\"Fedora\\"\'
%endif
-# Disable ld.gold on aarch64 as the support is in upstream, but not packaged in Fedora.
mkdir -p %{_target_platform}
pushd %{_target_platform}
%cmake \
@@ -198,9 +191,6 @@ pushd %{_target_platform}
-DCMAKE_BUILD_TYPE=Release \
-DENABLE_GTKDOC=ON \
-DENABLE_MINIBROWSER=ON \
-%ifarch aarch64
- -DUSE_LD_GOLD=OFF \
-%endif
%ifarch s390x %{power64}
-DENABLE_JIT=OFF \
-DUSE_SYSTEM_MALLOC=ON \
commit e9a4a11dc9073d1abc46bb6ca2502b2126f45b2c
Author: Tomas Popela <tpopela(a)redhat.com>
Date: Wed Sep 5 13:48:45 2018 +0200
Remove support for s390
diff --git a/webkit2gtk3.spec b/webkit2gtk3.spec
index ff591fd..1f62ae0 100644
--- a/webkit2gtk3.spec
+++ b/webkit2gtk3.spec
@@ -173,7 +173,7 @@ rm -rf Source/ThirdParty/qunit/
# is only needed for x86_64.
%global _dwz_max_die_limit_x86_64 250000000
-%ifarch s390 aarch64
+%ifarch aarch64
# Use linker flags to reduce memory consumption - on other arches the ld.gold is
# used and also it doesn't have the --reduce-memory-overheads option
%global optflags %{optflags} -Wl,--no-keep-memory -Wl,--reduce-memory-overheads
@@ -181,7 +181,7 @@ rm -rf Source/ThirdParty/qunit/
# Decrease debuginfo even on ix86 because of:
# https://bugs.webkit.org/show_bug.cgi?id=140176
-%ifarch s390 s390x %{arm} %{ix86} %{power64} %{mips}
+%ifarch s390x %{arm} %{ix86} %{power64} %{mips}
# Decrease debuginfo verbosity to reduce memory consumption even more
%global optflags %(echo %{optflags} | sed 's/-g /-g1 /')
%endif
@@ -190,8 +190,7 @@ rm -rf Source/ThirdParty/qunit/
%global optflags %{optflags} -DUSER_AGENT_GTK_DISTRIBUTOR_NAME=\'\\"Fedora\\"\'
%endif
-# Disable ld.gold on s390 as it does not have it.
-# Also for aarch64 as the support is in upstream, but not packaged in Fedora.
+# Disable ld.gold on aarch64 as the support is in upstream, but not packaged in Fedora.
mkdir -p %{_target_platform}
pushd %{_target_platform}
%cmake \
@@ -199,10 +198,10 @@ pushd %{_target_platform}
-DCMAKE_BUILD_TYPE=Release \
-DENABLE_GTKDOC=ON \
-DENABLE_MINIBROWSER=ON \
-%ifarch s390 aarch64
+%ifarch aarch64
-DUSE_LD_GOLD=OFF \
%endif
-%ifarch s390 s390x %{power64}
+%ifarch s390x %{power64}
-DENABLE_JIT=OFF \
-DUSE_SYSTEM_MALLOC=ON \
%endif
commit b7ddce8c9adae2a07b3afe7f6f0365f8d3d65398
Author: Tomas Popela <tpopela(a)redhat.com>
Date: Wed Sep 5 13:45:07 2018 +0200
Remove support for ppc
diff --git a/webkit2gtk3.spec b/webkit2gtk3.spec
index 62307c6..ff591fd 100644
--- a/webkit2gtk3.spec
+++ b/webkit2gtk3.spec
@@ -181,16 +181,11 @@ rm -rf Source/ThirdParty/qunit/
# Decrease debuginfo even on ix86 because of:
# https://bugs.webkit.org/show_bug.cgi?id=140176
-%ifarch s390 s390x %{arm} %{ix86} ppc %{power64} %{mips}
+%ifarch s390 s390x %{arm} %{ix86} %{power64} %{mips}
# Decrease debuginfo verbosity to reduce memory consumption even more
%global optflags %(echo %{optflags} | sed 's/-g /-g1 /')
%endif
-%ifarch ppc
-# Use linker flag -relax to get WebKit build under ppc(32) with JIT disabled
-%global optflags %{optflags} -Wl,-relax
-%endif
-
%if 0%{?fedora}
%global optflags %{optflags} -DUSER_AGENT_GTK_DISTRIBUTOR_NAME=\'\\"Fedora\\"\'
%endif
@@ -207,7 +202,7 @@ pushd %{_target_platform}
%ifarch s390 aarch64
-DUSE_LD_GOLD=OFF \
%endif
-%ifarch s390 s390x ppc %{power64}
+%ifarch s390 s390x %{power64}
-DENABLE_JIT=OFF \
-DUSE_SYSTEM_MALLOC=ON \
%endif
1
0
Architecture specific change in rpms/webkit2gtk3.git
by githook-noreply＠fedoraproject.org 05 Sep '18
by githook-noreply＠fedoraproject.org 05 Sep '18
05 Sep '18
The package rpms/webkit2gtk3.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/webkit2gtk3.git/commit/?id=993499dd…
https://src.fedoraproject.org/cgit/rpms/webkit2gtk3.git/commit/?id=b3878d23…
https://src.fedoraproject.org/cgit/rpms/webkit2gtk3.git/commit/?id=f740aa52….
Change:
+%ifarch s390x %{power64}
+%ifarch s390 s390x %{power64}
-%ifarch aarch64
Thanks.
Full change:
============
commit f740aa52956f956a058efe296ef43497ea74d979
Author: Tomas Popela <tpopela(a)redhat.com>
Date: Wed Sep 5 14:30:03 2018 +0200
Enable ld.gold on aarch64
diff --git a/webkit2gtk3.spec b/webkit2gtk3.spec
index d29eb8b..f278f77 100644
--- a/webkit2gtk3.spec
+++ b/webkit2gtk3.spec
@@ -172,12 +172,6 @@ rm -rf Source/ThirdParty/qunit/
# is only needed for x86_64.
%global _dwz_max_die_limit_x86_64 250000000
-%ifarch aarch64
-# Use linker flags to reduce memory consumption - on other arches the ld.gold is
-# used and also it doesn't have the --reduce-memory-overheads option
-%global optflags %{optflags} -Wl,--no-keep-memory -Wl,--reduce-memory-overheads
-%endif
-
# Decrease debuginfo even on ix86 because of:
# https://bugs.webkit.org/show_bug.cgi?id=140176
%ifarch s390x %{arm} %{ix86} %{power64} %{mips}
@@ -189,7 +183,6 @@ rm -rf Source/ThirdParty/qunit/
%global optflags %{optflags} -DUSER_AGENT_GTK_DISTRIBUTOR_NAME=\'\\"Fedora\\"\'
%endif
-# Disable ld.gold on aarch64 as the support is in upstream, but not packaged in Fedora.
mkdir -p %{_target_platform}
pushd %{_target_platform}
%cmake \
@@ -197,9 +190,6 @@ pushd %{_target_platform}
-DCMAKE_BUILD_TYPE=Release \
-DENABLE_GTKDOC=ON \
-DENABLE_MINIBROWSER=ON \
-%ifarch aarch64
- -DUSE_LD_GOLD=OFF \
-%endif
%ifarch s390x %{power64}
-DENABLE_JIT=OFF \
-DUSE_SYSTEM_MALLOC=ON \
commit 993499ddd0e9b45e48cbe8b212a7135d3059faff
Author: Tomas Popela <tpopela(a)redhat.com>
Date: Wed Sep 5 13:48:45 2018 +0200
Remove support for s390
diff --git a/webkit2gtk3.spec b/webkit2gtk3.spec
index a11821c..d29eb8b 100644
--- a/webkit2gtk3.spec
+++ b/webkit2gtk3.spec
@@ -172,7 +172,7 @@ rm -rf Source/ThirdParty/qunit/
# is only needed for x86_64.
%global _dwz_max_die_limit_x86_64 250000000
-%ifarch s390 aarch64
+%ifarch aarch64
# Use linker flags to reduce memory consumption - on other arches the ld.gold is
# used and also it doesn't have the --reduce-memory-overheads option
%global optflags %{optflags} -Wl,--no-keep-memory -Wl,--reduce-memory-overheads
@@ -180,7 +180,7 @@ rm -rf Source/ThirdParty/qunit/
# Decrease debuginfo even on ix86 because of:
# https://bugs.webkit.org/show_bug.cgi?id=140176
-%ifarch s390 s390x %{arm} %{ix86} %{power64} %{mips}
+%ifarch s390x %{arm} %{ix86} %{power64} %{mips}
# Decrease debuginfo verbosity to reduce memory consumption even more
%global optflags %(echo %{optflags} | sed 's/-g /-g1 /')
%endif
@@ -189,8 +189,7 @@ rm -rf Source/ThirdParty/qunit/
%global optflags %{optflags} -DUSER_AGENT_GTK_DISTRIBUTOR_NAME=\'\\"Fedora\\"\'
%endif
-# Disable ld.gold on s390 as it does not have it.
-# Also for aarch64 as the support is in upstream, but not packaged in Fedora.
+# Disable ld.gold on aarch64 as the support is in upstream, but not packaged in Fedora.
mkdir -p %{_target_platform}
pushd %{_target_platform}
%cmake \
@@ -198,10 +197,10 @@ pushd %{_target_platform}
-DCMAKE_BUILD_TYPE=Release \
-DENABLE_GTKDOC=ON \
-DENABLE_MINIBROWSER=ON \
-%ifarch s390 aarch64
+%ifarch aarch64
-DUSE_LD_GOLD=OFF \
%endif
-%ifarch s390 s390x %{power64}
+%ifarch s390x %{power64}
-DENABLE_JIT=OFF \
-DUSE_SYSTEM_MALLOC=ON \
%endif
commit b3878d239f3669b3b89f85c2797db7217d4a0d83
Author: Tomas Popela <tpopela(a)redhat.com>
Date: Wed Sep 5 13:45:07 2018 +0200
Remove support for ppc
diff --git a/webkit2gtk3.spec b/webkit2gtk3.spec
index 55283d0..a11821c 100644
--- a/webkit2gtk3.spec
+++ b/webkit2gtk3.spec
@@ -71,9 +71,6 @@ BuildRequires: rubygems
BuildRequires: sqlite-devel
BuildRequires: upower-devel
BuildRequires: woff2-devel
-%ifarch ppc
-BuildRequires: libatomic
-%endif
Requires: geoclue2
@@ -183,16 +180,11 @@ rm -rf Source/ThirdParty/qunit/
# Decrease debuginfo even on ix86 because of:
# https://bugs.webkit.org/show_bug.cgi?id=140176
-%ifarch s390 s390x %{arm} %{ix86} ppc %{power64} %{mips}
+%ifarch s390 s390x %{arm} %{ix86} %{power64} %{mips}
# Decrease debuginfo verbosity to reduce memory consumption even more
%global optflags %(echo %{optflags} | sed 's/-g /-g1 /')
%endif
-%ifarch ppc
-# Use linker flag -relax to get WebKit build under ppc(32) with JIT disabled
-%global optflags %{optflags} -Wl,-relax
-%endif
-
%if 0%{?fedora}
%global optflags %{optflags} -DUSER_AGENT_GTK_DISTRIBUTOR_NAME=\'\\"Fedora\\"\'
%endif
@@ -209,7 +201,7 @@ pushd %{_target_platform}
%ifarch s390 aarch64
-DUSE_LD_GOLD=OFF \
%endif
-%ifarch s390 s390x ppc %{power64}
+%ifarch s390 s390x %{power64}
-DENABLE_JIT=OFF \
-DUSE_SYSTEM_MALLOC=ON \
%endif
1
0
05 Sep '18
The package rpms/ceph.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/ceph.git/commit/?id=b79737d3dfe8949…
https://src.fedoraproject.org/cgit/rpms/ceph.git/commit/?id=9b4d3225ea28de8…
https://src.fedoraproject.org/cgit/rpms/ceph.git/commit/?id=9094835645403e6….
Change:
-%ifnarch armv7hl
+%ifnarch %{arm}
+%ifarch %{arm}
Thanks.
Full change:
============
commit 28151a4d3fe04fc08c19e255071be288e3a6726f
Merge: 38c140c 2d75781
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Wed Sep 5 08:51:21 2018 -0400
New release (1:12.2.8-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --cc ceph.spec
index c0fc897,8cd60d0..1c6f82f
--- a/ceph.spec
+++ b/ceph.spec
@@@ -1795,7 -1797,11 +1796,10 @@@ exit
# We need an empty %%files list for python-ceph-compat, to tell rpmbuild to
# actually build this meta package.
-
%changelog
+ * Fri Aug 31 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.8-1
+ - New release (1:12.2.8-1)
+
* Wed Jul 18 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.7-1
- New release (1:12.2.7-1)
commit 2d75781cd8490c06b739c1e4b3a9efd6082c26af
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Fri Aug 31 13:38:49 2018 -0400
New release (1:12.2.8-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index ebcf901..8cd60d0 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -85,7 +85,7 @@
# main package definition
#################################################################################
Name: ceph
-Version: 12.2.7
+Version: 12.2.8
Release: 1%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
@@ -171,6 +171,7 @@ BuildRequires: python
BuildRequires: python2-devel
BuildRequires: python-nose
BuildRequires: python-requests
+BuildRequires: python-six
BuildRequires: python-virtualenv
BuildRequires: snappy-devel
BuildRequires: udev
@@ -326,6 +327,7 @@ Summary: Ceph Metadata Server Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
+Requires: python-six
%description mds
ceph-mds is the metadata server daemon for the Ceph distributed file system.
One or more instances of ceph-mds collectively manage the file system
@@ -868,7 +870,6 @@ cmake .. \
-DCMAKE_INSTALL_MANDIR=%{_mandir} \
-DCMAKE_INSTALL_DOCDIR=%{_docdir}/ceph \
-DCMAKE_INSTALL_INCLUDEDIR=%{_includedir} \
- -DWITH_EMBEDDED=OFF \
-DWITH_MANPAGE=ON \
-DWITH_PYTHON3=ON \
-DWITH_SYSTEMD=ON \
@@ -1492,7 +1493,7 @@ if [ $1 -eq 1 ] ; then
/usr/bin/systemctl start ceph-osd.target >/dev/null 2>&1 || :
fi
# work around https://tracker.ceph.com/issues/24903
-chown -h ceph:ceph /var/lib/ceph/osd/*/block* 2>&1 > /dev/null || :
+chown -f -h ceph:ceph /var/lib/ceph/osd/*/block* 2>&1 > /dev/null || :
%preun osd
%if 0%{?suse_version}
@@ -1798,6 +1799,9 @@ exit 0
%changelog
+* Fri Aug 31 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.8-1
+- New release (1:12.2.8-1)
+
* Wed Jul 18 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.7-1
- New release (1:12.2.7-1)
diff --git a/sources b/sources
index 7eb8c41..e781e91 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-12.2.7.tar.gz) = dde75a41c31cca54d4a680b0b93e28680a23f912a348cccca65e55e600c315bbfe3ced008d758fe8ba1e801a9d6e5f8079cc157ec5eef8916b86f487f42064ea
+SHA512 (ceph-12.2.8.tar.gz) = 91c5c11aaeef701b3e31c9f8a1192b6b79fd34f3b96c61595c65173c98bfce0009c5b7c9b29764f5e4d4b1f5ac2ca7109fed5ae598c176aba717a34a132fabac
commit ce9ff30f1cfd96189ebaec8dca6466043a8b1326
Author: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
Date: Mon Jul 30 00:22:27 2018 +0200
%{python_sitearch} %{python2_sitearch}
Signed-off-by: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
diff --git a/ceph.spec b/ceph.spec
index 97367a7..ebcf901 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -1562,8 +1562,8 @@ fi
%{_mandir}/man8/librados-config.8*
%files -n python-rados
-%{python_sitearch}/rados.so
-%{python_sitearch}/rados-*.egg-info
+%{python2_sitearch}/rados.so
+%{python2_sitearch}/rados-*.egg-info
%files -n python%{python3_pkgversion}-rados
%{python3_sitearch}/rados.cpython*.so
@@ -1607,16 +1607,16 @@ fi
%{_libdir}/librgw.so
%files -n python-rgw
-%{python_sitearch}/rgw.so
-%{python_sitearch}/rgw-*.egg-info
+%{python2_sitearch}/rgw.so
+%{python2_sitearch}/rgw-*.egg-info
%files -n python%{python3_pkgversion}-rgw
%{python3_sitearch}/rgw.cpython*.so
%{python3_sitearch}/rgw-*.egg-info
%files -n python-rbd
-%{python_sitearch}/rbd.so
-%{python_sitearch}/rbd-*.egg-info
+%{python2_sitearch}/rbd.so
+%{python2_sitearch}/rbd-*.egg-info
%files -n python%{python3_pkgversion}-rbd
%{python3_sitearch}/rbd.cpython*.so
@@ -1633,8 +1633,8 @@ fi
%{_libdir}/libcephfs.so
%files -n python-cephfs
-%{python_sitearch}/cephfs.so
-%{python_sitearch}/cephfs-*.egg-info
+%{python2_sitearch}/cephfs.so
+%{python2_sitearch}/cephfs-*.egg-info
%{python2_sitelib}/ceph_volume_client.py*
%files -n python%{python3_pkgversion}-cephfs
commit 7a061682c0fa1bdb72bc77ff973948861c9303cb
Author: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
Date: Sun Jul 29 23:39:13 2018 +0200
python-devel python2-devel
Signed-off-by: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
diff --git a/ceph.spec b/ceph.spec
index d62f832..97367a7 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -168,7 +168,7 @@ BuildRequires: parted
BuildRequires: perl
BuildRequires: pkgconfig
BuildRequires: python
-BuildRequires: python-devel
+BuildRequires: python2-devel
BuildRequires: python-nose
BuildRequires: python-requests
BuildRequires: python-virtualenv
commit 2c2513ea411bdf57eaa4ffad4f3e6161b9cf7e8f
Author: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
Date: Sun Jul 29 23:11:50 2018 +0200
%{python_sitelib} %{python2_sitelib}
Signed-off-by: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
diff --git a/ceph.spec b/ceph.spec
index 0e5af31..d62f832 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -1018,11 +1018,11 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%endif
%{_unitdir}/ceph-disk@.service
%{_unitdir}/ceph.target
-%{python_sitelib}/ceph_detect_init*
-%{python_sitelib}/ceph_disk*
+%{python2_sitelib}/ceph_detect_init*
+%{python2_sitelib}/ceph_disk*
%dir %{python_sitelib}/ceph_volume
-%{python_sitelib}/ceph_volume/*
-%{python_sitelib}/ceph_volume-*
+%{python2_sitelib}/ceph_volume/*
+%{python2_sitelib}/ceph_volume-*
%{_mandir}/man8/ceph-deploy.8*
%{_mandir}/man8/ceph-detect-init.8*
%{_mandir}/man8/ceph-create-keys.8*
@@ -1139,8 +1139,8 @@ fi
%config %{_sysconfdir}/bash_completion.d/radosgw-admin
%config(noreplace) %{_sysconfdir}/ceph/rbdmap
%{_unitdir}/rbdmap.service
-%{python_sitelib}/ceph_argparse.py*
-%{python_sitelib}/ceph_daemon.py*
+%{python2_sitelib}/ceph_argparse.py*
+%{python2_sitelib}/ceph_daemon.py*
%dir %{_udevrulesdir}
%{_udevrulesdir}/50-rbd.rules
%attr(3770,ceph,ceph) %dir %{_localstatedir}/log/ceph/
@@ -1286,7 +1286,7 @@ fi
%{_bindir}/ceph-monstore-tool
%{_mandir}/man8/ceph-mon.8*
%{_mandir}/man8/ceph-rest-api.8*
-%{python_sitelib}/ceph_rest_api.py*
+%{python2_sitelib}/ceph_rest_api.py*
%{_unitdir}/ceph-mon@.service
%{_unitdir}/ceph-mon.target
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/mon
@@ -1635,7 +1635,7 @@ fi
%files -n python-cephfs
%{python_sitearch}/cephfs.so
%{python_sitearch}/cephfs-*.egg-info
-%{python_sitelib}/ceph_volume_client.py*
+%{python2_sitelib}/ceph_volume_client.py*
%files -n python%{python3_pkgversion}-cephfs
%{python3_sitearch}/cephfs.cpython*.so
commit a4054a83fc48875e02d2620ea8cc76b26dfea20c
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Wed Jul 18 13:09:25 2018 -0400
New release (1:12.2.7-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index c6a9ede..0e5af31 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -1798,7 +1798,7 @@ exit 0
%changelog
-* Wed Jul 28 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.7-1
+* Wed Jul 18 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.7-1
- New release (1:12.2.7-1)
* Thu Jul 12 2018 Fedora Release Engineering <releng(a)fedoraproject.org> - 1:12.2.6-2
commit 1b6e363bd1c1bfaa714c9d0ef43ed595cc2361e9
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Wed Jul 18 09:59:46 2018 -0400
New release (1:12.2.7-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index f5c4121..c6a9ede 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -85,8 +85,8 @@
# main package definition
#################################################################################
Name: ceph
-Version: 12.2.6
-Release: 2%{?dist}
+Version: 12.2.7
+Release: 1%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -1491,6 +1491,8 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl start ceph-osd.target >/dev/null 2>&1 || :
fi
+# work around https://tracker.ceph.com/issues/24903
+chown -h ceph:ceph /var/lib/ceph/osd/*/block* 2>&1 > /dev/null || :
%preun osd
%if 0%{?suse_version}
@@ -1796,6 +1798,9 @@ exit 0
%changelog
+* Wed Jul 28 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.7-1
+- New release (1:12.2.7-1)
+
* Thu Jul 12 2018 Fedora Release Engineering <releng(a)fedoraproject.org> - 1:12.2.6-2
- Rebuilt for https://fedoraproject.org/wiki/Fedora_29_Mass_Rebuild
@@ -1925,10 +1930,10 @@ exit 0
- Use newer -devel package format
- Sync up the spec file
-* Wed Oct 26 2016 Ken Dreyer <ktdreyer(a)ktdreyer.com> 1:10.2.3-4
+* Wed Oct 26 2016 Ken Dreyer <ktdreyer(a)ktdreyer.com> - 1:10.2.3-4
- librgw: add API version defines for librgw and rgw_file
-* Wed Oct 26 2016 Ken Dreyer <ktdreyer(a)ktdreyer.com> 1:10.2.3-3
+* Wed Oct 26 2016 Ken Dreyer <ktdreyer(a)ktdreyer.com> - 1:10.2.3-3
- update patches style for rdopkg
* Thu Sep 29 2016 Boris Ranto <branto(a)redhat.com> - 1:10.2.3-2
diff --git a/sources b/sources
index 2920e75..7eb8c41 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-12.2.6.tar.gz) = 7512623a9bf64b5da1989f62a6c5a3761ae21773bdac185f36a4dcd470f6489724807f28e98e9861ff34f86d2b0ce7af00eac2ddbaa5bba43b49786cb65ad7ff
+SHA512 (ceph-12.2.7.tar.gz) = dde75a41c31cca54d4a680b0b93e28680a23f912a348cccca65e55e600c315bbfe3ced008d758fe8ba1e801a9d6e5f8079cc157ec5eef8916b86f487f42064ea
commit 0fb6d3aa054e1c0996b84b670cc454a841c1460b
Author: Fedora Release Engineering <releng(a)fedoraproject.org>
Date: Thu Jul 12 21:41:00 2018 +0000
- Rebuilt for https://fedoraproject.org/wiki/Fedora_29_Mass_Rebuild
Signed-off-by: Fedora Release Engineering <releng(a)fedoraproject.org>
diff --git a/ceph.spec b/ceph.spec
index bc2111c..f5c4121 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -86,7 +86,7 @@
#################################################################################
Name: ceph
Version: 12.2.6
-Release: 1%{?dist}
+Release: 2%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -1796,6 +1796,9 @@ exit 0
%changelog
+* Thu Jul 12 2018 Fedora Release Engineering <releng(a)fedoraproject.org> - 1:12.2.6-2
+- Rebuilt for https://fedoraproject.org/wiki/Fedora_29_Mass_Rebuild
+
* Wed Jul 11 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.6-1
- New release (1:12.2.6-1)
commit 294576f608fc74be499777b70d41ed4444c5cb4a
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Wed Jul 11 09:01:29 2018 -0400
New release (1:12.2.6-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index e7a968c..bc2111c 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -85,8 +85,8 @@
# main package definition
#################################################################################
Name: ceph
-Version: 12.2.5
-Release: 3%{?dist}
+Version: 12.2.6
+Release: 1%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -190,7 +190,7 @@ BuildRequires: pkgconfig(systemd)
BuildRequires: systemd-rpm-macros
BuildRequires: systemd
%{?systemd_requires}
-PreReq: %fillup_prereq
+# PreReq: %%fillup_prereq
BuildRequires: net-tools
BuildRequires: libbz2-devel
BuildRequires: btrfsprogs
@@ -1796,6 +1796,9 @@ exit 0
%changelog
+* Wed Jul 11 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.6-1
+- New release (1:12.2.6-1)
+
* Mon Jul 2 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-3
- New release (1:12.2.5-3) w/ python-3.7
diff --git a/sources b/sources
index 021832c..2920e75 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-12.2.5.tar.gz) = bd78358590eb077afcc3c1e12864fadee906c95ac8afee25bdca6155f99b75f89d923159fcc13430abba21a89fee3e566177ff5f9ffabdd3b3f220d05602ab86
+SHA512 (ceph-12.2.6.tar.gz) = 7512623a9bf64b5da1989f62a6c5a3761ae21773bdac185f36a4dcd470f6489724807f28e98e9861ff34f86d2b0ce7af00eac2ddbaa5bba43b49786cb65ad7ff
commit 9aeaf53af545215050e285a73f59f7368dceb896
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Mon Jul 2 09:34:50 2018 -0400
New release (1:12.2.5-3) w/ python-3.7
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index 4e62552..e7a968c 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -86,7 +86,7 @@
#################################################################################
Name: ceph
Version: 12.2.5
-Release: 2%{?dist}
+Release: 3%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -1796,6 +1796,9 @@ exit 0
%changelog
+* Mon Jul 2 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-3
+- New release (1:12.2.5-3) w/ python-3.7
+
* Fri Jun 29 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-2
- New release (1:12.2.5-2)
commit 9b4d3225ea28de8647d7dae8a659f3ddfb793911
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Sun Jul 1 06:35:53 2018 -0400
New release (1:12.2.5-2)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/0001-src-rocksdb-util-murmurhash.patch b/0001-src-rocksdb-util-murmurhash.patch
new file mode 100644
index 0000000..f67a0ff
--- /dev/null
+++ b/0001-src-rocksdb-util-murmurhash.patch
@@ -0,0 +1,13 @@
+--- ceph-12.1.1.orig/src/rocksdb/util/murmurhash.cc 2017-04-27 01:13:46.000000000 +0100
++++ ceph-12.1.1.orig/src/rocksdb/util/murmurhash.cc 2017-07-25 11:37:28.910266684 +0100
+@@ -113,8 +113,8 @@ unsigned int MurmurHash2 ( const void *
+
+ switch(len)
+ {
+- case 3: h ^= data[2] << 16;
+- case 2: h ^= data[1] << 8;
++ case 3: h ^= data[2] << 16; // fallthrough
++ case 2: h ^= data[1] << 8; // fallthrough
+ case 1: h ^= data[0];
+ h *= m;
+ };
diff --git a/0002-cmake-Support-ppc64.patch b/0002-cmake-Support-ppc64.patch
new file mode 100644
index 0000000..6b09549
--- /dev/null
+++ b/0002-cmake-Support-ppc64.patch
@@ -0,0 +1,127 @@
+From 2f0a7153460acc3f21462236f470ec3471fa2ee1 Mon Sep 17 00:00:00 2001
+From: Boris Ranto <branto(a)redhat.com>
+Date: Mon, 31 Jul 2017 19:50:23 +0200
+Subject: [PATCH] cmake: Support ppc64
+
+The ppc64 support requires a couple of changes:
+- adding the ppc64 support to cmake
+- changing optimized crc32 code to compile on ppc64le only
+- moving ifdef condition before crc32_align to avoid defined but not
+used warning
+
+Signed-off-by: Boris Ranto <branto(a)redhat.com>
+---
+ cmake/modules/SIMDExt.cmake | 15 ++++++++++++++-
+ src/CMakeLists.txt | 4 +++-
+ src/arch/ppc.c | 8 ++++----
+ src/common/crc32c_ppc.c | 6 +++---
+ 4 files changed, 24 insertions(+), 9 deletions(-)
+
+diff --git a/cmake/modules/SIMDExt.cmake b/cmake/modules/SIMDExt.cmake
+index 5330835..c47667d 100644
+--- a/cmake/modules/SIMDExt.cmake
++++ b/cmake/modules/SIMDExt.cmake
+@@ -109,7 +109,20 @@ elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "i386|i686|amd64|x86_64|AMD64")
+ endif(CMAKE_SYSTEM_PROCESSOR MATCHES "i686|amd64|x86_64|AMD64")
+ elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "(powerpc|ppc)64le")
+ set(HAVE_PPC64LE 1)
+- message(STATUS " we are ppc64le")
++ message(STATUS " we are ppc64")
++ CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
++ if(HAS_ALTIVEC)
++ message(STATUS " HAS_ALTIVEC yes")
++ set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maltivec")
++ set(CMAKE_CXX_FLAGS "${CMAKE_C_FLAGS} -maltivec")
++ endif()
++ CHECK_C_COMPILER_FLAG("-mcpu=power8" HAVE_POWER8)
++ if(HAVE_POWER8)
++ message(STATUS " HAVE_POWER8 yes")
++ endif()
++elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "(power|ppc)64")
++ set(HAVE_PPC64 1)
++ message(STATUS " we are ppc64")
+ CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
+ if(HAS_ALTIVEC)
+ message(STATUS " HAS_ALTIVEC yes")
+diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
+index 66f0c14..38d1913 100644
+--- a/src/CMakeLists.txt
++++ b/src/CMakeLists.txt
+@@ -568,7 +568,9 @@ if(HAVE_INTEL)
+ endif(HAVE_GOOD_YASM_ELF64)
+ elseif(HAVE_POWER8)
+ list(APPEND libcommon_files
+- common/crc32c_ppc.c
++ common/crc32c_ppc.c)
++elseif(HAVE_PPC64LE)
++ list(APPEND libcommon_files
+ common/crc32c_ppc_asm.S
+ common/crc32c_ppc_fast_zero_asm.S)
+ endif(HAVE_INTEL)
+diff --git a/src/arch/ppc.c b/src/arch/ppc.c
+index f21e2fe..11d3a49 100644
+--- a/src/arch/ppc.c
++++ b/src/arch/ppc.c
+@@ -14,10 +14,10 @@ int ceph_arch_ppc_crc32 = 0;
+
+ #include <stdio.h>
+
+-#if __linux__ && __powerpc64__
++#ifdef HAVE_PPC64LE
+ #include <sys/auxv.h>
+ #include <asm/cputable.h>
+-#endif /* __linux__ && __powerpc64__ */
++#endif /* HAVE_PPC64LE */
+
+ #ifndef PPC_FEATURE2_VEC_CRYPTO
+ #define PPC_FEATURE2_VEC_CRYPTO 0x02000000
+@@ -31,9 +31,9 @@ int ceph_arch_ppc_probe(void)
+ {
+ ceph_arch_ppc_crc32 = 0;
+
+-#if __linux__ && __powerpc64__
++#ifdef HAVE_PPC64LE
+ if (getauxval(AT_HWCAP2) & PPC_FEATURE2_VEC_CRYPTO) ceph_arch_ppc_crc32 = 1;
+-#endif /* __linux__ && __powerpc64__ */
++#endif /* HAVE_PPC64LE */
+
+ return 0;
+ }
+diff --git a/src/common/crc32c_ppc.c b/src/common/crc32c_ppc.c
+index 43756e2..52fd1c4 100644
+--- a/src/common/crc32c_ppc.c
++++ b/src/common/crc32c_ppc.c
+@@ -20,6 +20,7 @@
+ #define VMX_ALIGN 16
+ #define VMX_ALIGN_MASK (VMX_ALIGN-1)
+
++#ifdef HAVE_PPC64LE
+ #ifdef REFLECT
+ static unsigned int crc32_align(unsigned int crc, unsigned char const *p,
+ unsigned long len)
+@@ -38,7 +39,6 @@ static unsigned int crc32_align(unsigned int crc, unsigned char const *p,
+ }
+ #endif
+
+-#ifdef HAVE_POWER8
+ static inline unsigned long polynomial_multiply(unsigned int a, unsigned int b) {
+ vector unsigned int va = {a, 0, 0, 0};
+ vector unsigned int vb = {b, 0, 0, 0};
+@@ -134,7 +134,7 @@ uint32_t ceph_crc32c_ppc(uint32_t crc, unsigned char const *data, unsigned len)
+ return crc;
+ }
+
+-#else /* HAVE_POWER8 */
++#else /* HAVE_PPC64LE */
+
+ /* This symbol has to exist on non-ppc architectures (and on legacy
+ * ppc systems using power7 or below) in order to compile properly
+@@ -145,4 +145,4 @@ uint32_t ceph_crc32c_ppc(uint32_t crc, unsigned char const *data, unsigned len)
+ return 0;
+ }
+
+-#endif /* HAVE_POWER8 */
++#endif /* HAVE_PPC64LE */
+--
+2.9.4
+
diff --git a/0003-librbd-Conditionally-import-TrimRequest.cc.patch b/0003-librbd-Conditionally-import-TrimRequest.cc.patch
new file mode 100644
index 0000000..2f02efb
--- /dev/null
+++ b/0003-librbd-Conditionally-import-TrimRequest.cc.patch
@@ -0,0 +1,31 @@
+From 74a754690736f6608b0d4d9c807df0bd777a129d Mon Sep 17 00:00:00 2001
+From: Boris Ranto <branto(a)redhat.com>
+Date: Fri, 8 Dec 2017 00:21:38 +0100
+Subject: [PATCH] librbd: Conditionally import TrimRequest.cc
+
+We include TrimRequest.cc in librbd tests at two places:
+ - operation/test_mock_TrimRequest.cc
+ - operation/test_mock_ResizeRequest.cc
+
+That causes linking errors when doing the builds because some of the
+structures are defined twice.
+
+Signed-off-by: Boris Ranto <branto(a)redhat.com>
+---
+ src/librbd/operation/TrimRequest.cc | 2 ++
+ 1 file changed, 2 insertions(+)
+
+diff --git a/src/librbd/operation/TrimRequest.cc b/src/librbd/operation/TrimRequest.cc
+index 28f2deb..929ca51 100644
+--- a/src/librbd/operation/TrimRequest.cc
++++ b/src/librbd/operation/TrimRequest.cc
+@@ -362,4 +362,6 @@ void TrimRequest<I>::send_finish(int r) {
+ } // namespace operation
+ } // namespace librbd
+
++#ifndef TEST_F
+ template class librbd::operation::TrimRequest<librbd::ImageCtx>;
++#endif
+--
+2.9.5
+
diff --git a/0005-src-rocksdb-table-block.h.patch b/0005-src-rocksdb-table-block.h.patch
new file mode 100644
index 0000000..8acb401
--- /dev/null
+++ b/0005-src-rocksdb-table-block.h.patch
@@ -0,0 +1,155 @@
+--- ceph-12.2.3/src/rocksdb/table/block.h.orig 2018-02-22 07:49:38.044899631 -0500
++++ ceph-12.2.3/src/rocksdb/table/block.h 2018-02-22 07:58:55.855899631 -0500
+@@ -65,7 +65,8 @@
+
+ // Create bitmap and set all the bits to 0
+ bitmap_ = new std::atomic<uint32_t>[bitmap_size];
+- memset(bitmap_, 0, bitmap_size * kBytesPersEntry);
++ // memset(bitmap_, 0, bitmap_size * kBytesPersEntry);
++ { unsigned i = 0; for (; i < bitmap_size;) bitmap_[i++] = 0; }
+
+ RecordTick(GetStatistics(), READ_AMP_TOTAL_READ_BYTES,
+ num_bits_needed << bytes_per_bit_pow_);
+--- ceph-12.2.3/src/rocksdb/db/c.cc.orig 2018-02-22 08:14:56.033899631 -0500
++++ ceph-12.2.3/src/rocksdb/db/c.cc 2018-02-22 10:06:39.759899631 -0500
+@@ -1322,11 +1322,6 @@
+ b->rep.PutLogData(Slice(blob, len));
+ }
+
+-void rocksdb_writebatch_iterate(
+- rocksdb_writebatch_t* b,
+- void* state,
+- void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
+- void (*deleted)(void*, const char* k, size_t klen)) {
+ class H : public WriteBatch::Handler {
+ public:
+ void* state_;
+@@ -1339,6 +1334,12 @@
+ (*deleted_)(state_, key.data(), key.size());
+ }
+ };
++
++void rocksdb_writebatch_iterate(
++ rocksdb_writebatch_t* b,
++ void* state,
++ void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
++ void (*deleted)(void*, const char* k, size_t klen)) {
+ H handler;
+ handler.state_ = state;
+ handler.put_ = put;
+@@ -1579,18 +1580,6 @@
+ void* state,
+ void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
+ void (*deleted)(void*, const char* k, size_t klen)) {
+- class H : public WriteBatch::Handler {
+- public:
+- void* state_;
+- void (*put_)(void*, const char* k, size_t klen, const char* v, size_t vlen);
+- void (*deleted_)(void*, const char* k, size_t klen);
+- virtual void Put(const Slice& key, const Slice& value) override {
+- (*put_)(state_, key.data(), key.size(), value.data(), value.size());
+- }
+- virtual void Delete(const Slice& key) override {
+- (*deleted_)(state_, key.data(), key.size());
+- }
+- };
+ H handler;
+ handler.state_ = state;
+ handler.put_ = put;
+@@ -2532,13 +2521,9 @@
+ delete filter;
+ }
+
+-rocksdb_filterpolicy_t* rocksdb_filterpolicy_create_bloom_format(int bits_per_key, bool original_format) {
+- // Make a rocksdb_filterpolicy_t, but override all of its methods so
+- // they delegate to a NewBloomFilterPolicy() instead of user
+- // supplied C functions.
+- struct Wrapper : public rocksdb_filterpolicy_t {
++ struct WrapperFP : public rocksdb_filterpolicy_t {
+ const FilterPolicy* rep_;
+- ~Wrapper() { delete rep_; }
++ ~WrapperFP() { delete rep_; }
+ const char* Name() const override { return rep_->Name(); }
+ void CreateFilter(const Slice* keys, int n,
+ std::string* dst) const override {
+@@ -2549,11 +2534,16 @@
+ }
+ static void DoNothing(void*) { }
+ };
+- Wrapper* wrapper = new Wrapper;
++
++rocksdb_filterpolicy_t* rocksdb_filterpolicy_create_bloom_format(int bits_per_key, bool original_format) {
++ // Make a rocksdb_filterpolicy_t, but override all of its methods so
++ // they delegate to a NewBloomFilterPolicy() instead of user
++ // supplied C functions.
++ WrapperFP* wrapper = new WrapperFP;
+ wrapper->rep_ = NewBloomFilterPolicy(bits_per_key, original_format);
+ wrapper->state_ = nullptr;
+ wrapper->delete_filter_ = nullptr;
+- wrapper->destructor_ = &Wrapper::DoNothing;
++ wrapper->destructor_ = &WrapperFP::DoNothing;
+ return wrapper;
+ }
+
+@@ -2889,10 +2879,9 @@
+ delete st;
+ }
+
+-rocksdb_slicetransform_t* rocksdb_slicetransform_create_fixed_prefix(size_t prefixLen) {
+- struct Wrapper : public rocksdb_slicetransform_t {
++ struct WrapperST : public rocksdb_slicetransform_t {
+ const SliceTransform* rep_;
+- ~Wrapper() { delete rep_; }
++ ~WrapperST() { delete rep_; }
+ const char* Name() const override { return rep_->Name(); }
+ Slice Transform(const Slice& src) const override {
+ return rep_->Transform(src);
+@@ -2903,31 +2892,20 @@
+ bool InRange(const Slice& src) const override { return rep_->InRange(src); }
+ static void DoNothing(void*) { }
+ };
+- Wrapper* wrapper = new Wrapper;
++
++rocksdb_slicetransform_t* rocksdb_slicetransform_create_fixed_prefix(size_t prefixLen) {
++ WrapperST* wrapper = new WrapperST;
+ wrapper->rep_ = rocksdb::NewFixedPrefixTransform(prefixLen);
+ wrapper->state_ = nullptr;
+- wrapper->destructor_ = &Wrapper::DoNothing;
++ wrapper->destructor_ = &WrapperST::DoNothing;
+ return wrapper;
+ }
+
+ rocksdb_slicetransform_t* rocksdb_slicetransform_create_noop() {
+- struct Wrapper : public rocksdb_slicetransform_t {
+- const SliceTransform* rep_;
+- ~Wrapper() { delete rep_; }
+- const char* Name() const override { return rep_->Name(); }
+- Slice Transform(const Slice& src) const override {
+- return rep_->Transform(src);
+- }
+- bool InDomain(const Slice& src) const override {
+- return rep_->InDomain(src);
+- }
+- bool InRange(const Slice& src) const override { return rep_->InRange(src); }
+- static void DoNothing(void*) { }
+- };
+- Wrapper* wrapper = new Wrapper;
++ WrapperST* wrapper = new WrapperST;
+ wrapper->rep_ = rocksdb::NewNoopTransform();
+ wrapper->state_ = nullptr;
+- wrapper->destructor_ = &Wrapper::DoNothing;
++ wrapper->destructor_ = &WrapperST::DoNothing;
+ return wrapper;
+ }
+
+--- ceph-12.2.3/src/rocksdb/memtable/inlineskiplist.h.orig 2018-02-22 10:34:06.918899631 -0500
++++ ceph-12.2.3/src/rocksdb/memtable/inlineskiplist.h 2018-02-22 10:34:44.145899631 -0500
+@@ -279,7 +279,7 @@
+ // next_[0]. This is used for passing data from AllocateKey to Insert.
+ void StashHeight(const int height) {
+ assert(sizeof(int) <= sizeof(next_[0]));
+- memcpy(&next_[0], &height, sizeof(int));
++ memcpy(static_cast<void*>(&next_[0]), &height, sizeof(int));
+ }
+
+ // Retrieves the value passed to StashHeight. Undefined after a call
diff --git a/ceph.spec b/ceph.spec
index 96ba41f..4e62552 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -16,59 +16,58 @@
#
%global _hardened_build 1
-%bcond_with python3
%bcond_without ocf
+%ifnarch armv7hl
+%bcond_without cephfs_java
+%else
+%bcond_with cephfs_java
+%endif
+%if 0%{?suse_version}
+%bcond_with ceph_test_package
+%else
+%bcond_without ceph_test_package
+%endif
%bcond_with make_check
%ifarch s390 s390x
%bcond_with tcmalloc
%else
%bcond_without tcmalloc
%endif
+%ifnarch armv7hl
+%bcond_with lowmem_builder
+%else
+%if 0%{?rhel}
+%ifnarch ppc64le
+%bcond_with lowmem_builder
+%else
+%bcond_without lowmem_builder
+%endif
+%else
+%bcond_without lowmem_builder
+%endif
+%endif
%if 0%{?fedora} || 0%{?rhel}
%bcond_without selinux
-%bcond_without ceph_test_package
-%bcond_without cephfs_java
-%bcond_without lttng
-%global _remote_tarball_prefix https://download.ceph.com/tarballs/
%endif
%if 0%{?suse_version}
%bcond_with selinux
-%bcond_with ceph_test_package
-%bcond_with cephfs_java
-#Compat macro for new _fillupdir macro introduced in Nov 2017
-%if ! %{defined _fillupdir}
-%global _fillupdir /var/adm/fillup-templates
%endif
-%if 0%{?is_opensuse}
-%bcond_without lttng
-%else
-%ifarch x86_64 aarch64
+
+# LTTng-UST enabled on Fedora, RHEL 6+, and SLE (not openSUSE)
+%if 0%{?fedora} || 0%{?rhel} >= 6 || 0%{?suse_version}
+%if ! 0%{?is_opensuse}
%bcond_without lttng
-%else
-%bcond_with lttng
%endif
%endif
-%endif
-%if 0%{?suse_version} >= 1500
-%bcond_with python2
-%else
-%bcond_without python2
-%endif
-%if 0%{without python2}
-%global _defined_if_python2_absent 1
-%endif
%if %{with selinux}
# get selinux policy version
-%{!?_selinux_policy_version: %global _selinux_policy_version 0.0.0}
+%{!?_selinux_policy_version: %global _selinux_policy_version %(sed -e 's,.*selinux-policy-\\([^/]*\\)/.*,\\1,' /usr/share/selinux/devel/policyhelp 2>/dev/null || echo 0.0.0)}
%endif
%{!?_udevrulesdir: %global _udevrulesdir /lib/udev/rules.d}
%{!?tmpfiles_create: %global tmpfiles_create systemd-tmpfiles --create}
%{!?python3_pkgversion: %global python3_pkgversion 3}
-# define _python_buildid macro which will expand to the empty string when
-# building with python2
-%global _python_buildid %{?_defined_if_python2_absent:%{python3_pkgversion}}
# unify libexec for all targets
%global _libexecdir %{_exec_prefix}/lib
@@ -77,23 +76,22 @@
%global _find_debuginfo_dwz_opts %{nil}
%if ( 0%{?rhel} && 0%{?rhel} < 7 )
-%global _rundir %{_localstatedir}/run/
+%global _rundir %{_localstatedir}/run
%else
-%global _rundir /run/
+%global _rundir /run
%endif
#################################################################################
# main package definition
#################################################################################
Name: ceph
-Version: 13.2.0
-Release: 3%{?dist}
+Version: 12.2.5
+Release: 2%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
-# define _epoch_prefix macro which will expand to the empty string if epoch is
-# undefined
+# define %%_epoch_prefix macro which will expand to the empty string if %%epoch is undefined
%global _epoch_prefix %{?epoch:%{epoch}:}
Summary: User space components of the Ceph file system
@@ -102,11 +100,14 @@ License: LGPL-2.1 and CC-BY-SA-3.0 and GPL-2.0 and BSL-1.0 and BSD-3-Clause and
Group: System/Filesystems
%endif
URL: http://ceph.com/
-Source0: %{?_remote_tarball_prefix}%{name}-%{version}.tar.gz
+Source0: http://download.ceph.com/tarballs/%{name}-%{version}.tar.gz
# https://bugzilla.redhat.com/show_bug.cgi?id=1474773
-Patch001: 0001-blobstore.patch
+Patch001: 0001-src-rocksdb-util-murmurhash.patch
+# https://bugzilla.redhat.com/show_bug.cgi?id=1474774
+Patch002: 0002-cmake-Support-ppc64.patch
+Patch003: 0003-librbd-Conditionally-import-TrimRequest.cc.patch
+Patch005: 0005-src-rocksdb-table-block.h.patch
%if 0%{?suse_version}
-# _insert_obs_source_lines_here
%if 0%{?is_opensuse}
ExclusiveArch: x86_64 aarch64 ppc64 ppc64le
%else
@@ -116,10 +117,10 @@ ExclusiveArch: x86_64 aarch64 ppc64le s390x
#################################################################################
# dependencies that apply across all distro families
#################################################################################
-Requires: ceph-osd = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mds = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mgr = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mon = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-osd = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mds = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mgr = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mon = %{_epoch_prefix}%{version}-%{release}
Requires(post): binutils
%if 0%{with cephfs_java}
BuildRequires: java-devel
@@ -128,48 +129,50 @@ BuildRequires: sharutils
%if 0%{with selinux}
BuildRequires: checkpolicy
BuildRequires: selinux-policy-devel
+BuildRequires: selinux-policy-doc
+%endif
+%if 0%{with make_check}
+%if 0%{?fedora} || 0%{?rhel}
+BuildRequires: python-cherrypy
+BuildRequires: python-werkzeug
+%endif
+%if 0%{?suse_version}
+BuildRequires: python-CherryPy
+BuildRequires: python-Werkzeug
+BuildRequires: python-numpy-devel
+%endif
+BuildRequires: python-coverage
+BuildRequires: python-pecan
+BuildRequires: socat
%endif
BuildRequires: bc
BuildRequires: gperf
-BuildRequires: cmake
+BuildRequires: cmake
BuildRequires: cryptsetup
BuildRequires: fuse-devel
-BuildRequires: cryptopp-devel
-%if 0%{?rhel} == 7
-# devtoolset offers newer make and valgrind-devel, but the old ones are good
-# enough.
-BuildRequires: devtoolset-7-gcc-c++
-%else
BuildRequires: gcc-c++
-%endif
BuildRequires: gdbm
%if 0%{with tcmalloc}
BuildRequires: gperftools-devel >= 2.4
%endif
-BuildRequires: jq
+BuildRequires: jq
BuildRequires: leveldb-devel > 1.2
BuildRequires: libaio-devel
BuildRequires: libblkid-devel >= 2.17
BuildRequires: libcurl-devel
BuildRequires: libudev-devel
-BuildRequires: liboath-devel
BuildRequires: libtool
BuildRequires: libxml2-devel
-BuildRequires: libuuid-devel
BuildRequires: make
BuildRequires: parted
BuildRequires: perl
BuildRequires: pkgconfig
-BuildRequires: procps
-BuildRequires: python%{_python_buildid}
-BuildRequires: python%{_python_buildid}-devel
-BuildRequires: python%{_python_buildid}-nose
-BuildRequires: python%{_python_buildid}-requests
-BuildRequires: python%{_python_buildid}-virtualenv
+BuildRequires: python
+BuildRequires: python-devel
+BuildRequires: python-nose
+BuildRequires: python-requests
+BuildRequires: python-virtualenv
BuildRequires: snappy-devel
-%if 0%{with make_check}
-BuildRequires: socat
-%endif
BuildRequires: udev
BuildRequires: util-linux
BuildRequires: valgrind-devel
@@ -183,49 +186,43 @@ BuildRequires: yasm
# distro-conditional dependencies
#################################################################################
%if 0%{?suse_version}
-BuildRequires: pkgconfig(systemd)
+BuildRequires: pkgconfig(systemd)
BuildRequires: systemd-rpm-macros
BuildRequires: systemd
%{?systemd_requires}
PreReq: %fillup_prereq
BuildRequires: net-tools
BuildRequires: libbz2-devel
-BuildRequires: btrfsprogs
+BuildRequires: btrfsprogs
BuildRequires: mozilla-nss-devel
BuildRequires: keyutils-devel
-BuildRequires: libopenssl-devel
-BuildRequires: lsb-release
-BuildRequires: openldap2-devel
-BuildRequires: cunit-devel
-BuildRequires: python%{_python_buildid}-base
-BuildRequires: python%{_python_buildid}-Cython
-BuildRequires: python%{_python_buildid}-PrettyTable
-BuildRequires: python%{_python_buildid}-Sphinx
-BuildRequires: rdma-core-devel
-BuildRequires: liblz4-devel >= 1.7
+BuildRequires: libopenssl-devel
+BuildRequires: lsb-release
+BuildRequires: openldap2-devel
+BuildRequires: python-Cython
+BuildRequires: python-PrettyTable
+BuildRequires: python-Sphinx
+BuildRequires: rdma-core-devel
%endif
%if 0%{?fedora} || 0%{?rhel}
Requires: systemd
-BuildRequires: boost-random
+BuildRequires: boost-random
BuildRequires: btrfs-progs
BuildRequires: nss-devel
BuildRequires: keyutils-libs-devel
-# RDMA is no longer built on 32-bit ARM: see rhbz#1484155
+# RDMA is no longer built on 32-bit ARM: see #1484155
%ifnarch %{arm}
BuildRequires: rdma-core-devel
%endif
-BuildRequires: openldap-devel
-BuildRequires: openssl-devel
-BuildRequires: redhat-lsb-core
-BuildRequires: CUnit-devel
-BuildRequires: Cython%{_python_buildid}
-BuildRequires: python%{_python_buildid}-prettytable
-BuildRequires: python%{_python_buildid}-sphinx
-BuildRequires: lz4-devel >= 1.7
+BuildRequires: openldap-devel
+BuildRequires: openssl-devel
+BuildRequires: redhat-lsb-core
+BuildRequires: Cython
+BuildRequires: python-prettytable
+BuildRequires: python-sphinx
%endif
# python34-... for RHEL, python3-... for all other supported distros
-%if %{with python3}
-%if 0%{?rhel}
+%if ( 0%{?rhel} && 0%{?rhel} <= 7 )
BuildRequires: python34-devel
BuildRequires: python34-setuptools
BuildRequires: python34-Cython
@@ -234,34 +231,6 @@ BuildRequires: python3-devel
BuildRequires: python3-setuptools
BuildRequires: python3-Cython
%endif
-%endif
-# distro-conditional make check dependencies
-%if 0%{with make_check}
-%if 0%{?fedora} || 0%{?rhel}
-BuildRequires: python%{_python_buildid}-cherrypy
-BuildRequires: python%{_python_buildid}-routes
-BuildRequires: python%{_python_buildid}-pecan
-BuildRequires: python%{_python_buildid}-werkzeug
-BuildRequires: python%{_python_buildid}-tox
-BuildRequires: python%{_python_buildid}-coverage
-%if 0%{?fedora}
-BuildRequires: python%{_python_buildid}-bcrypt
-%endif
-%if 0%{?rhel}
-BuildRequires: py-bcrypt
-%endif
-%endif
-%if 0%{?suse_version}
-BuildRequires: python%{_python_buildid}-CherryPy
-BuildRequires: python%{_python_buildid}-Routes
-BuildRequires: python%{_python_buildid}-Werkzeug
-BuildRequires: python%{_python_buildid}-pecan
-BuildRequires: python%{_python_buildid}-numpy-devel
-BuildRequires: python%{_python_buildid}-bcrypt
-BuildRequires: python%{_python_buildid}-tox
-BuildRequires: python%{_python_buildid}-coverage
-%endif
-%endif
# lttng and babeltrace for rbd-replay-prep
%if %{with lttng}
%if 0%{?fedora} || 0%{?rhel}
@@ -270,7 +239,7 @@ BuildRequires: libbabeltrace-devel
%endif
%if 0%{?suse_version}
BuildRequires: lttng-ust-devel
-BuildRequires: babeltrace-devel
+BuildRequires: babeltrace-devel
%endif
%endif
%if 0%{?suse_version}
@@ -281,7 +250,7 @@ BuildRequires: expat-devel
%endif
#hardened-cc1
%if 0%{?fedora} || 0%{?rhel}
-BuildRequires: redhat-rpm-config
+BuildRequires: redhat-rpm-config
%endif
%description
@@ -293,39 +262,33 @@ on commodity hardware and delivers object, block and file system storage.
# subpackages
#################################################################################
%package base
-Summary: Ceph Base Package
+Summary: Ceph Base Package
%if 0%{?suse_version}
-Group: System/Filesystems
+Group: System/Filesystems
%endif
-Requires: ceph-common = %{_epoch_prefix}%{version}-%{release}
-Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
-Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
-Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
-Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-common = %{_epoch_prefix}%{version}-%{release}
+Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
+Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
+Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
+Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
%if 0%{with selinux}
-Requires: ceph-selinux = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-selinux = %{_epoch_prefix}%{version}-%{release}
%endif
Requires(post):/sbin/ldconfig
Requires(postun):/sbin/ldconfig
-Requires: cryptsetup
-Requires: e2fsprogs
-Requires: findutils
-Requires: grep
-Requires: logrotate
-Requires: parted
-Requires: psmisc
-Requires: python%{_python_buildid}-requests
-Requires: python%{_python_buildid}-setuptools
-Requires: util-linux
-Requires: xfsprogs
-Requires: which
-%if 0%{?fedora} || 0%{?rhel}
-Requires: gdisk
-%endif
-%if 0%{?suse_version}
-Recommends: ntp-daemon
-Recommends: chrony
-Requires: gptfdisk
+Requires: python
+Requires: python-requests
+Requires: python-setuptools
+Requires: grep
+Requires: xfsprogs
+Requires: logrotate
+Requires: util-linux
+Requires: cryptsetup
+Requires: findutils
+Requires: psmisc
+Requires: which
+%if 0%{?suse_version}
+Recommends: ntp-daemon
%endif
%description base
Base is the package that includes all the files shared amongst ceph servers
@@ -338,18 +301,17 @@ Group: System/Filesystems
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
-Requires: python%{_python_buildid}-rados = %{_epoch_prefix}%{version}-%{release}
-Requires: python%{_python_buildid}-rbd = %{_epoch_prefix}%{version}-%{release}
-Requires: python%{_python_buildid}-cephfs = %{_epoch_prefix}%{version}-%{release}
-Requires: python%{_python_buildid}-rgw = %{_epoch_prefix}%{version}-%{release}
+Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
+Requires: python-rbd = %{_epoch_prefix}%{version}-%{release}
+Requires: python-cephfs = %{_epoch_prefix}%{version}-%{release}
+Requires: python-rgw = %{_epoch_prefix}%{version}-%{release}
%if 0%{?fedora} || 0%{?rhel}
-Requires: python%{_python_buildid}-prettytable
-Requires: python%{_python_buildid}-requests
+Requires: python-prettytable
%endif
%if 0%{?suse_version}
-Requires: python%{_python_buildid}-PrettyTable
-Requires: python%{_python_buildid}-requests
+Requires: python-PrettyTable
%endif
+Requires: python-requests
%{?systemd_requires}
%if 0%{?suse_version}
Requires(pre): pwdutils
@@ -375,6 +337,13 @@ Summary: Ceph Monitor Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
+# For ceph-rest-api
+%if 0%{?fedora} || 0%{?rhel}
+Requires: python-flask
+%endif
+%if 0%{?suse_version}
+Requires: python-Flask
+%endif
%description mon
ceph-mon is the cluster monitor daemon for the Ceph distributed file
system. One or more instances of ceph-mon form a Paxos part-time
@@ -382,35 +351,24 @@ parliament cluster that provides extremely reliable and durable storage
of cluster membership, configuration, and state.
%package mgr
-Summary: Ceph Manager Daemon
+Summary: Ceph Manager Daemon
%if 0%{?suse_version}
-Group: System/Filesystems
+Group: System/Filesystems
%endif
-Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
%if 0%{?fedora} || 0%{?rhel}
-Requires: python%{_python_buildid}-cherrypy
-Requires: python%{_python_buildid}-routes
-Requires: python%{_python_buildid}-jinja2
-Requires: python%{_python_buildid}-pecan
-Requires: python%{_python_buildid}-werkzeug
-Requires: pyOpenSSL%{_python_buildid}
-%if 0%{?fedora}
-Requires: python%{_python_buildid}-bcrypt
-%endif
-%if 0%{?rhel}
-Requires: py-bcrypt
-%endif
+Requires: python-cherrypy
+Requires: python-jinja2
+Requires: python-werkzeug
+Requires: pyOpenSSL
%endif
%if 0%{?suse_version}
-Requires: python%{_python_buildid}-CherryPy
-Requires: python%{_python_buildid}-Routes
-Requires: python%{_python_buildid}-Jinja2
-Requires: python%{_python_buildid}-Werkzeug
-Requires: python%{_python_buildid}-pecan
-Requires: python%{_python_buildid}-pyOpenSSL
-Requires: python%{_python_buildid}-bcrypt
-Recommends: python%{_python_buildid}-influxdb
+Requires: python-CherryPy
+Requires: python-jinja2
+Requires: python-Werkzeug
+Requires: python-pyOpenSSL
%endif
+Requires: python-pecan
%description mgr
ceph-mgr enables python modules that provide services (such as the REST
module derived from Calamari) and expose CLI hooks. ceph-mgr gathers
@@ -497,7 +455,15 @@ Summary: Ceph Object Storage Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
-Requires: lvm2
+# for sgdisk, used by ceph-disk
+%if 0%{?fedora} || 0%{?rhel}
+Requires: gdisk
+%endif
+%if 0%{?suse_version}
+Requires: gptfdisk
+%endif
+Requires: parted
+Requires: lvm2
%description osd
ceph-osd is the object storage daemon for the Ceph distributed file
system. It is responsible for storing objects on a local file system
@@ -554,7 +520,6 @@ Obsoletes: librgw2-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use RADOS gateway client library.
-%if 0%{with python2}
%package -n python-rgw
Summary: Python 2 libraries for the RADOS gateway
%if 0%{?suse_version}
@@ -566,22 +531,18 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rgw
This package contains Python 2 libraries for interacting with Cephs RADOS
gateway.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-rgw
Summary: Python 3 libraries for the RADOS gateway
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-rgw
This package contains Python 3 libraries for interacting with Cephs RADOS
gateway.
-%endif
-%if 0%{with python2}
%package -n python-rados
Summary: Python 2 libraries for the RADOS object store
%if 0%{?suse_version}
@@ -592,20 +553,17 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rados
This package contains Python 2 libraries for interacting with Cephs RADOS
object store.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-rados
Summary: Python 3 libraries for the RADOS object store
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: python%{python3_pkgversion}
Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-rados
This package contains Python 3 libraries for interacting with Cephs RADOS
object store.
-%endif
%package -n libradosstriper1
Summary: RADOS striping interface
@@ -665,11 +623,10 @@ Obsoletes: librbd1-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use RADOS block device.
-%if 0%{with python2}
%package -n python-rbd
Summary: Python 2 libraries for the RADOS block device
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
@@ -677,20 +634,17 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rbd
This package contains Python 2 libraries for interacting with Cephs RADOS
block device.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-rbd
Summary: Python 3 libraries for the RADOS block device
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-rbd
This package contains Python 3 libraries for interacting with Cephs RADOS
block device.
-%endif
%package -n libcephfs2
Summary: Ceph distributed file system client library
@@ -723,47 +677,41 @@ Obsoletes: libcephfs2-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use Cephs distributed file system.
-%if 0%{with python2}
%package -n python-cephfs
Summary: Python 2 libraries for Ceph distributed file system
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
%if 0%{?suse_version}
-Recommends: python-rados = %{_epoch_prefix}%{version}-%{release}
+Recommends: python-rados = %{_epoch_prefix}%{version}-%{release}
%endif
Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-cephfs
This package contains Python 2 libraries for interacting with Cephs distributed
file system.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-cephfs
Summary: Python 3 libraries for Ceph distributed file system
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-cephfs
This package contains Python 3 libraries for interacting with Cephs distributed
file system.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-ceph-argparse
Summary: Python 3 utility libraries for Ceph CLI
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
%description -n python%{python3_pkgversion}-ceph-argparse
This package contains types and routines for Python 3 used by the Ceph CLI as
well as the RESTful interface. These have to do with querying the daemons for
command-description information, validating user command input against those
descriptions, and submitting the command to the appropriate daemon.
-%endif
%if 0%{with ceph_test_package}
%package -n ceph-test
@@ -813,18 +761,17 @@ Group: System/Libraries
%endif
Requires: java
Requires: libcephfs_jni1 = %{_epoch_prefix}%{version}-%{release}
-BuildRequires: junit
+Requires: junit
+BuildRequires: junit
%description -n cephfs-java
This package contains the Java libraries for the Ceph File System.
%endif
%package -n rados-objclass-devel
-Summary: RADOS object class development kit
-%if 0%{?suse_version}
-Group: Development/Libraries/C and C++
-%endif
-Requires: librados2-devel = %{_epoch_prefix}%{version}-%{release}
+Summary: RADOS object class development kit
+Group: Development/Libraries
+Requires: librados2-devel = %{_epoch_prefix}%{version}-%{release}
%description -n rados-objclass-devel
This package contains libraries and headers needed to develop RADOS object
class plugins.
@@ -848,11 +795,10 @@ populated file-systems.
%endif
-%if 0%{with python2}
%package -n python-ceph-compat
Summary: Compatibility package for Cephs python libraries
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Obsoletes: python-ceph
Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
@@ -865,7 +811,6 @@ This is a compatibility package to accommodate python-ceph split into
python-rados, python-rbd, python-rgw and python-cephfs. Packages still
depending on python-ceph should be fixed to depend on python-rados,
python-rbd, python-rgw or python-cephfs instead.
-%endif
#################################################################################
# common
@@ -874,11 +819,6 @@ python-rbd, python-rgw or python-cephfs instead.
%autosetup -p1 -n %{name}-%{version}
%build
-
-%if 0%{?rhel} == 7
-. /opt/rh/devtoolset-7/enable
-%endif
-
%if 0%{with cephfs_java}
# Find jni.h
for i in /usr/{lib64,lib}/jvm/java/include{,/linux}; do
@@ -886,40 +826,37 @@ for i in /usr/{lib64,lib}/jvm/java/include{,/linux}; do
done
%endif
-%if 0%{?suse_version}
-# the following setting fixed an OOM condition we once encountered in the OBS
+%if %{with lowmem_builder}
RPM_OPT_FLAGS="$RPM_OPT_FLAGS --param ggc-min-expand=20 --param ggc-min-heapsize=32768"
%endif
+%ifnarch armv7hl
+export RPM_OPT_FLAGS=`echo $RPM_OPT_FLAGS | sed -e 's/i386/i486/'`
+%else
+export RPM_OPT_FLAGS=`echo $RPM_OPT_FLAGS | sed -e 's/i386/i486/' -e 's/-pipe//g'`
+%endif
export CPPFLAGS="$java_inc"
export CFLAGS="$RPM_OPT_FLAGS"
export CXXFLAGS="$RPM_OPT_FLAGS"
export LDFLAGS="$RPM_LD_FLAGS"
-# Parallel build settings ...
-CEPH_MFLAGS_JOBS="%{?_smp_mflags}"
-CEPH_SMP_NCPUS=$(echo "$CEPH_MFLAGS_JOBS" | sed 's/-j//')
-%if 0%{?__isa_bits} == 32
-# 32-bit builds can use 3G memory max, which is not enough even for -j2
-CEPH_SMP_NCPUS="1"
-%endif
-# do not eat all memory
-echo "Available memory:"
-free -h
-echo "System limits:"
-ulimit -a
-if test -n "$CEPH_SMP_NCPUS" -a "$CEPH_SMP_NCPUS" -gt 1 ; then
- mem_per_process=1800
- max_mem=$(LANG=C free -m | sed -n "s|^Mem: *\([0-9]*\).*$|\1|p")
- max_jobs="$(($max_mem / $mem_per_process))"
- test "$CEPH_SMP_NCPUS" -gt "$max_jobs" && CEPH_SMP_NCPUS="$max_jobs" && echo "Warning: Reducing build parallelism to -j$max_jobs because of memory limits"
- test "$CEPH_SMP_NCPUS" -le 0 && CEPH_SMP_NCPUS="1" && echo "Warning: Not using parallel build at all because of memory limits"
-fi
-export CEPH_SMP_NCPUS
-export CEPH_MFLAGS_JOBS="-j$CEPH_SMP_NCPUS"
-
env | sort
+%if %{with lowmem_builder}
+%ifnarch armv7hl
+%if 0%{?jobs} > 8
+%define _smp_mflags -j8
+%endif
+%else
+%define _smp_mflags -j1
+%endif
+%endif
+
+# unlimit _smp_mflags in system macro if not set above
+%define _smp_ncpus_max 0
+# extract the number of processors for use with cmake
+%define _smp_ncpus %(echo %{_smp_mflags} | sed 's/-j//')
+
mkdir build
cd build
cmake .. \
@@ -933,19 +870,9 @@ cmake .. \
-DCMAKE_INSTALL_INCLUDEDIR=%{_includedir} \
-DWITH_EMBEDDED=OFF \
-DWITH_MANPAGE=ON \
-%if %{with python3}
-DWITH_PYTHON3=ON \
-%else
- -DWITH_PYTHON3=OFF \
-%endif
- -DWITH_MGR_DASHBOARD_FRONTEND=OFF \
-%if %{with python2}
- -DWITH_PYTHON2=ON \
-%else
- -DWITH_PYTHON2=OFF \
- -DMGR_PYTHON_VERSION=3 \
-%endif
-%if ( ( 0%{?rhel} && 0%{?rhel} < 8) && ! 0%{?centos} )
+ -DWITH_SYSTEMD=ON \
+%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
-DWITH_SUBMAN=ON \
%endif
%if 0%{without ceph_test_package}
@@ -973,23 +900,24 @@ cmake .. \
%else
-DWITH_BOOST_CONTEXT=OFF \
%endif
-%ifarch %{arm}
+%ifnarch %{arm}
-DWITH_RDMA=OFF \
%endif
- -DWITH_DPDK=ON \
- -DBOOST_J=$CEPH_SMP_NCPUS
+ -DBOOST_J=%{_smp_ncpus}
-make "$CEPH_MFLAGS_JOBS"
+make %{?_smp_mflags}
%if 0%{with make_check}
%check
# run in-tree unittests
cd build
-ctest "$CEPH_MFLAGS_JOBS"
+ctest %{?_smp_mflags}
+
%endif
+
%install
pushd build
make DESTDIR=%{buildroot} install
@@ -1001,9 +929,10 @@ install -m 0644 -D src/etc-rbdmap %{buildroot}%{_sysconfdir}/ceph/rbdmap
install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_sysconfdir}/sysconfig/ceph
%endif
%if 0%{?suse_version}
-install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_fillupdir}/sysconfig.%{name}
+install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_localstatedir}/adm/fillup-templates/sysconfig.%{name}
%endif
install -m 0644 -D systemd/ceph.tmpfiles.d %{buildroot}%{_tmpfilesdir}/ceph-common.conf
+install -m 0755 -D systemd/ceph %{buildroot}%{_sbindir}/rcceph
install -m 0644 -D systemd/50-ceph.preset %{buildroot}%{_libexecdir}/systemd/system-preset/50-ceph.preset
mkdir -p %{buildroot}%{_sbindir}
install -m 0644 -D src/logrotate.conf %{buildroot}%{_sysconfdir}/logrotate.d/ceph
@@ -1026,7 +955,7 @@ install -m 0644 -D udev/95-ceph-osd.rules %{buildroot}%{_udevrulesdir}/95-ceph-o
#set up placeholder directories
mkdir -p %{buildroot}%{_sysconfdir}/ceph
-mkdir -p %{buildroot}%{_rundir}ceph
+mkdir -p %{buildroot}%{_rundir}/ceph
mkdir -p %{buildroot}%{_localstatedir}/log/ceph
mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/tmp
mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/mon
@@ -1045,9 +974,6 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%py3_compile %{buildroot}%{python3_sitelib}
%endif
-%clean
-rm -rf %{buildroot}
-
#################################################################################
# files and systemd scriptlets
#################################################################################
@@ -1063,6 +989,7 @@ rm -rf %{buildroot}
%{_libexecdir}/systemd/system-preset/50-ceph.preset
%{_sbindir}/ceph-create-keys
%{_sbindir}/ceph-disk
+%{_sbindir}/rcceph
%dir %{_libexecdir}/ceph
%{_libexecdir}/ceph/ceph_common.sh
%dir %{_libdir}/rados-classes
@@ -1085,30 +1012,17 @@ rm -rf %{buildroot}
%config(noreplace) %{_sysconfdir}/sysconfig/ceph
%endif
%if 0%{?suse_version}
-%{_fillupdir}/sysconfig.*
+%{_localstatedir}/adm/fillup-templates/sysconfig.*
%config %{_sysconfdir}/sysconfig/SuSEfirewall2.d/services/ceph-mon
%config %{_sysconfdir}/sysconfig/SuSEfirewall2.d/services/ceph-osd-mds
%endif
%{_unitdir}/ceph-disk@.service
%{_unitdir}/ceph.target
-%if 0%{with python2}
%{python_sitelib}/ceph_detect_init*
%{python_sitelib}/ceph_disk*
-%endif
-%if 0%{with python3}
-%{python3_sitelib}/ceph_detect_init*
-%{python3_sitelib}/ceph_disk*
-%endif
-%if 0%{with python2}
%dir %{python_sitelib}/ceph_volume
%{python_sitelib}/ceph_volume/*
%{python_sitelib}/ceph_volume-*
-%endif
-%if 0%{with python3}
-%dir %{python3_sitelib}/ceph_volume
-%{python3_sitelib}/ceph_volume/*
-%{python3_sitelib}/ceph_volume-*
-%endif
%{_mandir}/man8/ceph-deploy.8*
%{_mandir}/man8/ceph-detect-init.8*
%{_mandir}/man8/ceph-create-keys.8*
@@ -1127,7 +1041,6 @@ rm -rf %{buildroot}
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/bootstrap-rbd
%post base
-/sbin/ldconfig
%if 0%{?suse_version}
%fillup_only
if [ $1 -eq 1 ] ; then
@@ -1150,7 +1063,6 @@ fi
%endif
%postun base
-/sbin/ldconfig
test -n "$FIRST_ARG" || FIRST_ARG=$1
%if 0%{?suse_version}
DISABLE_RESTART_ON_UPDATE="yes"
@@ -1181,6 +1093,7 @@ fi
%{_bindir}/ceph-dencoder
%{_bindir}/ceph-rbdnamer
%{_bindir}/ceph-syn
+%{_bindir}/ceph-crush-location
%{_bindir}/cephfs-data-scan
%{_bindir}/cephfs-journal-tool
%{_bindir}/cephfs-table-tool
@@ -1198,6 +1111,7 @@ fi
%{_bindir}/rbd-replay-prep
%endif
%{_bindir}/ceph-post-file
+%{_bindir}/ceph-brag
%{_tmpfilesdir}/ceph-common.conf
%{_mandir}/man8/ceph-authtool.8*
%{_mandir}/man8/ceph-conf.8*
@@ -1225,16 +1139,8 @@ fi
%config %{_sysconfdir}/bash_completion.d/radosgw-admin
%config(noreplace) %{_sysconfdir}/ceph/rbdmap
%{_unitdir}/rbdmap.service
-%if 0%{with python2}
%{python_sitelib}/ceph_argparse.py*
%{python_sitelib}/ceph_daemon.py*
-%endif
-%if 0%{with python3}
-%{python3_sitelib}/ceph_argparse.py
-%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
-%{python3_sitelib}/ceph_daemon.py
-%{python3_sitelib}/__pycache__/ceph_daemon.cpython*.py*
-%endif
%dir %{_udevrulesdir}
%{_udevrulesdir}/50-rbd.rules
%attr(3770,ceph,ceph) %dir %{_localstatedir}/log/ceph/
@@ -1376,8 +1282,11 @@ fi
%files mon
%{_bindir}/ceph-mon
+%{_bindir}/ceph-rest-api
%{_bindir}/ceph-monstore-tool
%{_mandir}/man8/ceph-mon.8*
+%{_mandir}/man8/ceph-rest-api.8*
+%{python_sitelib}/ceph_rest_api.py*
%{_unitdir}/ceph-mon@.service
%{_unitdir}/ceph-mon.target
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/mon
@@ -1556,7 +1465,7 @@ fi
%{_mandir}/man8/ceph-bluestore-tool.8*
%{_mandir}/man8/ceph-volume.8*
%{_mandir}/man8/ceph-volume-systemd.8*
-%if ( ( 0%{?rhel} && 0%{?rhel} < 8) && ! 0%{?centos} )
+%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
%attr(0755,-,-) %{_sysconfdir}/cron.hourly/subman
%endif
%{_unitdir}/ceph-osd@.service
@@ -1570,6 +1479,11 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl preset ceph-osd(a)\*.service ceph-volume(a)\*.service ceph-osd.target >/dev/null 2>&1 || :
fi
+%if 0%{?sysctl_apply}
+ %sysctl_apply 90-ceph-osd.conf
+%else
+ /usr/lib/systemd/systemd-sysctl %{_sysctldir}/90-ceph-osd.conf > /dev/null 2>&1 || :
+%endif
%endif
%if 0%{?fedora} || 0%{?rhel}
%systemd_post ceph-osd(a)\*.service ceph-volume(a)\*.service ceph-osd.target
@@ -1577,11 +1491,6 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl start ceph-osd.target >/dev/null 2>&1 || :
fi
-%if 0%{?sysctl_apply}
- %sysctl_apply 90-ceph-osd.conf
-%else
- /usr/lib/systemd/systemd-sysctl %{_sysctldir}/90-ceph-osd.conf > /dev/null 2>&1 || :
-%endif
%preun osd
%if 0%{?suse_version}
@@ -1650,17 +1559,13 @@ fi
%{_bindir}/librados-config
%{_mandir}/man8/librados-config.8*
-%if 0%{with python2}
%files -n python-rados
%{python_sitearch}/rados.so
%{python_sitearch}/rados-*.egg-info
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-rados
%{python3_sitearch}/rados.cpython*.so
%{python3_sitearch}/rados-*.egg-info
-%endif
%ldconfig_scriptlets -n libradosstriper1
%files -n libradosstriper1
@@ -1692,10 +1597,6 @@ fi
%ldconfig_scriptlets -n librgw2
%files -n librgw2
%{_libdir}/librgw.so.*
-%if %{with lttng}
-%{_libdir}/librgw_op_tp.so*
-%{_libdir}/librgw_rados_tp.so*
-%endif
%files -n librgw-devel
%dir %{_includedir}/rados
@@ -1703,29 +1604,21 @@ fi
%{_includedir}/rados/rgw_file.h
%{_libdir}/librgw.so
-%if 0%{with python2}
%files -n python-rgw
%{python_sitearch}/rgw.so
%{python_sitearch}/rgw-*.egg-info
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-rgw
%{python3_sitearch}/rgw.cpython*.so
%{python3_sitearch}/rgw-*.egg-info
-%endif
-%if 0%{with python2}
%files -n python-rbd
%{python_sitearch}/rbd.so
%{python_sitearch}/rbd-*.egg-info
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-rbd
%{python3_sitearch}/rbd.cpython*.so
%{python3_sitearch}/rbd-*.egg-info
-%endif
%ldconfig_scriptlets -n libcephfs2
%files -n libcephfs2
@@ -1737,28 +1630,22 @@ fi
%{_includedir}/cephfs/ceph_statx.h
%{_libdir}/libcephfs.so
-%if 0%{with python2}
%files -n python-cephfs
%{python_sitearch}/cephfs.so
%{python_sitearch}/cephfs-*.egg-info
%{python_sitelib}/ceph_volume_client.py*
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-cephfs
%{python3_sitearch}/cephfs.cpython*.so
%{python3_sitearch}/cephfs-*.egg-info
%{python3_sitelib}/ceph_volume_client.py
%{python3_sitelib}/__pycache__/ceph_volume_client.cpython*.py*
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-ceph-argparse
%{python3_sitelib}/ceph_argparse.py
%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
%{python3_sitelib}/ceph_daemon.py
%{python3_sitelib}/__pycache__/ceph_daemon.cpython*.py*
-%endif
%if 0%{with ceph_test_package}
%files -n ceph-test
@@ -1780,7 +1667,13 @@ fi
%{_bindir}/ceph_rgw_multiparser
%{_bindir}/ceph_scratchtool
%{_bindir}/ceph_scratchtoolpp
+%{_bindir}/ceph_smalliobench
+%{_bindir}/ceph_smalliobenchdumb
+%{_bindir}/ceph_smalliobenchfs
+%{_bindir}/ceph_smalliobenchrbd
%{_bindir}/ceph_test_*
+%{_bindir}/ceph_tpbench
+%{_bindir}/ceph_xattr_bench
%{_bindir}/ceph-coverage
%{_bindir}/ceph-debugpack
%{_mandir}/man8/ceph-debugpack.8*
@@ -1850,7 +1743,7 @@ fi
rm -f ${FILE_CONTEXT}.pre
# The fixfiles command won't fix label for /var/run/ceph
-/usr/sbin/restorecon -R %{_rundir}ceph > /dev/null 2>&1
+/usr/sbin/restorecon -R %{_rundir}/ceph > /dev/null 2>&1
# Start the daemons iff they were running before
if test $STATUS -eq 0; then
@@ -1886,7 +1779,7 @@ if [ $1 -eq 0 ]; then
/usr/sbin/fixfiles -C ${FILE_CONTEXT}.pre restore 2> /dev/null
rm -f ${FILE_CONTEXT}.pre
# The fixfiles command won't fix label for /var/run/ceph
- /usr/sbin/restorecon -R %{_rundir}ceph > /dev/null 2>&1
+ /usr/sbin/restorecon -R %{_rundir}/ceph > /dev/null 2>&1
# Start the daemons if they were running before
if test $STATUS -eq 0; then
@@ -1897,28 +1790,14 @@ exit 0
%endif # with selinux
-%if 0%{with python2}
%files -n python-ceph-compat
# We need an empty %%files list for python-ceph-compat, to tell rpmbuild to
# actually build this meta package.
-%endif
%changelog
-* Thu Jun 28 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.2.0-3
-- New release (1:13.2.0-3)
-
-* Tue Jun 19 2018 Miro Hronok <mhroncok(a)redhat.com> - 1:13.1.0-2
-- Rebuilt for Python 3.7
-
-* Thu May 31 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.2.0-1
-- New release (1:13.2.0-1)
-
-* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-2
-- New release (1:13.1.0-2) +crypto_plugins
-
-* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-1
-- New release (1:13.1.0-1)
+* Fri Jun 29 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-2
+- New release (1:12.2.5-2)
* Fri Apr 27 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-1
- New release (1:12.2.5-1)
diff --git a/sources b/sources
index ae5b656..021832c 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-13.2.0.tar.gz) = 2524ad5a868024463e1a1edc5bd9c3f7bc1f58e47bf8d331902ed53206120abc6a427dd9c1d657131e67efd955d6ed8052b6c3fed8658aa905d58ac58ee62973
+SHA512 (ceph-12.2.5.tar.gz) = bd78358590eb077afcc3c1e12864fadee906c95ac8afee25bdca6155f99b75f89d923159fcc13430abba21a89fee3e566177ff5f9ffabdd3b3f220d05602ab86
commit 9094835645403e63f2e8a34a33e8fc9672d9f64d
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Thu Jun 28 02:04:51 2018 -0400
New release (1:13.2.0-3)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch b/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
deleted file mode 100644
index ffc38da..0000000
--- a/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
+++ /dev/null
@@ -1,165 +0,0 @@
---- ceph-13.1.0/src/mon/LogMonitor.cc.orig 2018-05-08 08:12:50.101933925 -0400
-+++ ceph-13.1.0/src/mon/LogMonitor.cc 2018-05-08 08:12:59.470933925 -0400
-@@ -62,7 +62,6 @@
- {
- dout(10) << "create_initial -- creating initial map" << dendl;
- LogEntry e;
-- memset(&e.who, 0, sizeof(e.who));
- e.name = g_conf->name;
- e.stamp = ceph_clock_now();
- e.prio = CLOG_INFO;
---- ceph-13.1.0/src/mon/MonMap.h.orig 2018-05-08 08:13:08.173933925 -0400
-+++ ceph-13.1.0/src/mon/MonMap.h 2018-05-08 08:13:17.391933925 -0400
-@@ -118,7 +118,6 @@
-
- MonMap()
- : epoch(0) {
-- memset(&fsid, 0, sizeof(fsid));
- }
-
- uuid_d& get_fsid() { return fsid; }
---- ceph-13.1.0/src/os/filestore/HashIndex.h.orig 2018-05-08 08:13:35.377933925 -0400
-+++ ceph-13.1.0/src/os/filestore/HashIndex.h 2018-05-08 08:13:59.853933925 -0400
-@@ -395,7 +395,7 @@
-
- struct CmpPairBitwise {
- bool operator()(const pair<string, ghobject_t>& l,
-- const pair<string, ghobject_t>& r)
-+ const pair<string, ghobject_t>& r) const
- {
- if (l.first < r.first)
- return true;
-@@ -408,7 +408,7 @@
- };
-
- struct CmpHexdigitStringBitwise {
-- bool operator()(const string& l, const string& r) {
-+ bool operator()(const string& l, const string& r) const {
- return reverse_hexdigit_bits_string(l) < reverse_hexdigit_bits_string(r);
- }
- };
---- ceph-13.1.0/src/os/filestore/LFNIndex.h.orig 2018-05-08 08:14:06.721933925 -0400
-+++ ceph-13.1.0/src/os/filestore/LFNIndex.h 2018-05-08 08:14:34.671933925 -0400
-@@ -63,7 +63,7 @@
- out: \
- complete_inject_failure(); \
- return r; \
-- } catch (RetryException) { \
-+ } catch (RetryException&) { \
- failed = true; \
- } catch (...) { \
- ceph_abort(); \
---- ceph-13.1.0/src/client/Inode.h.orig 2018-05-08 08:08:26.305933925 -0400
-+++ ceph-13.1.0/src/client/Inode.h 2018-05-08 08:08:38.029933925 -0400
-@@ -279,7 +279,6 @@
- _ref(0), ll_ref(0)
- {
- memset(&dir_layout, 0, sizeof(dir_layout));
-- memset("a, 0, sizeof(quota));
- }
- ~Inode();
-
---- ceph-13.1.0/src/osd/OSDMap.h.orig 2018-05-08 08:14:51.066933925 -0400
-+++ ceph-13.1.0/src/osd/OSDMap.h 2018-05-08 08:15:14.148933925 -0400
-@@ -427,7 +427,6 @@
- encode_features(0),
- epoch(e), new_pool_max(-1), new_flags(-1), new_max_osd(-1),
- have_crc(false), full_crc(0), inc_crc(0) {
-- memset(&fsid, 0, sizeof(fsid));
- }
- explicit Incremental(bufferlist &bl) {
- bufferlist::iterator p = bl.begin();
-@@ -607,7 +606,6 @@
- cached_up_osd_features(0),
- crc_defined(false), crc(0),
- crush(std::make_shared<CrushWrapper>()) {
-- memset(&fsid, 0, sizeof(fsid));
- }
-
- private:
---- ceph-13.1.0/src/common/cmdparse.h.orig 2018-05-08 08:09:17.772933925 -0400
-+++ ceph-13.1.0/src/common/cmdparse.h 2018-05-08 08:09:36.500933925 -0400
-@@ -54,7 +54,7 @@
- try {
- val = boost::get<T>(cmdmap.find(k)->second);
- return true;
-- } catch (boost::bad_get) {
-+ } catch (boost::bad_get&) {
- handle_bad_get(cct, k, typeid(T).name());
- }
- }
---- ceph-13.1.0/src/messages/MClientReply.h.orig 2018-05-08 08:10:22.281933925 -0400
-+++ ceph-13.1.0/src/messages/MClientReply.h 2018-05-08 08:10:46.660933925 -0400
-@@ -187,7 +187,7 @@
- if (features & CEPH_FEATURE_MDS_QUOTA)
- decode(quota, p);
- else
-- memset("a, 0, sizeof(quota));
-+ quota = quota_info_t{};
-
- if ((features & CEPH_FEATURE_FS_FILE_LAYOUT_V2))
- decode(layout.pool_ns, p);
---- ceph-13.1.0/src/messages/MMonSubscribeAck.h.orig 2018-05-08 08:10:58.478933925 -0400
-+++ ceph-13.1.0/src/messages/MMonSubscribeAck.h 2018-05-08 08:11:17.484933925 -0400
-@@ -23,7 +23,6 @@
-
- MMonSubscribeAck() : Message(CEPH_MSG_MON_SUBSCRIBE_ACK),
- interval(0) {
-- memset(&fsid, 0, sizeof(fsid));
- }
- MMonSubscribeAck(uuid_d& f, int i) : Message(CEPH_MSG_MON_SUBSCRIBE_ACK),
- interval(i), fsid(f) { }
---- ceph-13.1.0/src/mgr/DaemonState.h.orig 2018-05-08 08:12:16.521933925 -0400
-+++ ceph-13.1.0/src/mgr/DaemonState.h 2018-05-08 08:12:40.995933925 -0400
-@@ -130,7 +130,7 @@
- auto p = config_defaults_bl.begin();
- try {
- decode(config_defaults, p);
-- } catch (buffer::error e) {
-+ } catch (buffer::error& e) {
- }
- }
- return config_defaults;
---- ceph-13.1.0/cmake/modules/BuildDPDK.cmake.orig 2018-05-08 08:41:02.168933925 -0400
-+++ ceph-13.1.0/cmake/modules/BuildDPDK.cmake 2018-05-08 08:41:47.411933925 -0400
-@@ -71,7 +71,7 @@
- BUILD_IN_SOURCE 1
- INSTALL_COMMAND "true")
- ExternalProject_Add_Step(dpdk-ext patch-config
-- COMMAND ${CMAKE_MODULE_PATH}/patch-dpdk-conf.sh ${dpdk_dir} ${machine}
-+ COMMAND ${CMAKE_MODULE_PATH}/patch-dpdk-conf.sh ${dpdk_dir} ${machine} ${arch}
- DEPENDEES configure
- DEPENDERS build)
- # easier to adjust the config
-@@ -86,7 +86,7 @@
- # target
- file(MAKE_DIRECTORY ${DPDK_INCLUDE_DIR})
- foreach(c
-- pci bus_pci
-+ bus_pci pci
- eal
- mempool mempool_ring mempool_stack ring)
- add_library(dpdk::${c} STATIC IMPORTED)
---- ceph-13.1.0/cmake/modules/patch-dpdk-conf.sh.orig 2018-05-08 08:42:01.089933925 -0400
-+++ ceph-13.1.0/cmake/modules/patch-dpdk-conf.sh 2018-05-08 08:43:11.781933925 -0400
-@@ -15,8 +15,12 @@
- shift
- machine=$1
- shift
-+arch=$1
-+shift
-
- setconf CONFIG_RTE_MACHINE "${machine}"
-+setconf CONFIG_RTE_ARCH "${arch}"
-+
- # Disable experimental features
- setconf CONFIG_RTE_NEXT_ABI n
- setconf CONFIG_RTE_LIBRTE_MBUF_OFFLOAD n
-@@ -38,6 +42,7 @@
- setconf CONFIG_RTE_LIBRTE_VMXNET3_PMD n
- setconf CONFIG_RTE_LIBRTE_PMD_VHOST n
- setconf CONFIG_RTE_APP_EVENTDEV n
-+setconf CONFIG_RTE_MAX_VFIO_GROUPS 64
-
- # no test
- setconf CONFIG_RTE_APP_TEST n
diff --git a/0001-blobstore.patch b/0001-blobstore.patch
new file mode 100644
index 0000000..97462f0
--- /dev/null
+++ b/0001-blobstore.patch
@@ -0,0 +1,11 @@
+--- ceph-13.1.0/src/spdk/lib/blob/blobstore.c.orig 2018-05-08 16:04:30.329933925 -0400
++++ ceph-13.1.0/src/spdk/lib/blob/blobstore.c 2018-05-08 16:04:55.240933925 -0400
+@@ -2947,7 +2947,7 @@
+
+ /* START spdk_blob_resize */
+ int
+-spdk_blob_resize(struct spdk_blob *_blob, uint64_t sz)
++spdk_blob_resize(struct spdk_blob *_blob, size_t sz)
+ {
+ struct spdk_blob_data *blob = __blob_to_data(_blob);
+ int rc;
diff --git a/ceph.spec b/ceph.spec
index 8b0de11..96ba41f 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -16,6 +16,7 @@
#
%global _hardened_build 1
+%bcond_with python3
%bcond_without ocf
%bcond_with make_check
%ifarch s390 s390x
@@ -28,6 +29,7 @@
%bcond_without ceph_test_package
%bcond_without cephfs_java
%bcond_without lttng
+%global _remote_tarball_prefix https://download.ceph.com/tarballs/
%endif
%if 0%{?suse_version}
%bcond_with selinux
@@ -75,22 +77,23 @@
%global _find_debuginfo_dwz_opts %{nil}
%if ( 0%{?rhel} && 0%{?rhel} < 7 )
-%global _rundir %{_localstatedir}/run
+%global _rundir %{_localstatedir}/run/
%else
-%global _rundir /run
+%global _rundir /run/
%endif
#################################################################################
# main package definition
#################################################################################
Name: ceph
-Version: 13.1.0
-Release: 2%{?dist}
+Version: 13.2.0
+Release: 3%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
-# define %%_epoch_prefix macro which will expand to the empty string if %%epoch is undefined
+# define _epoch_prefix macro which will expand to the empty string if epoch is
+# undefined
%global _epoch_prefix %{?epoch:%{epoch}:}
Summary: User space components of the Ceph file system
@@ -99,14 +102,9 @@ License: LGPL-2.1 and CC-BY-SA-3.0 and GPL-2.0 and BSL-1.0 and BSD-3-Clause and
Group: System/Filesystems
%endif
URL: http://ceph.com/
-Source0: http://download.ceph.com/tarballs/%{name}-%{version}.tar.gz
+Source0: %{?_remote_tarball_prefix}%{name}-%{version}.tar.gz
# https://bugzilla.redhat.com/show_bug.cgi?id=1474773
-Patch001: 0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
-#Patch001: 0001-src-rocksdb-util-murmurhash.patch
-# https://bugzilla.redhat.com/show_bug.cgi?id=1474774
-#Patch002: 0002-cmake-Support-ppc64.patch
-#Patch003: 0003-librbd-Conditionally-import-TrimRequest.cc.patch
-#Patch005: 0005-src-rocksdb-table-block.h.patch
+Patch001: 0001-blobstore.patch
%if 0%{?suse_version}
# _insert_obs_source_lines_here
%if 0%{?is_opensuse}
@@ -136,6 +134,7 @@ BuildRequires: gperf
BuildRequires: cmake
BuildRequires: cryptsetup
BuildRequires: fuse-devel
+BuildRequires: cryptopp-devel
%if 0%{?rhel} == 7
# devtoolset offers newer make and valgrind-devel, but the old ones are good
# enough.
@@ -204,7 +203,6 @@ BuildRequires: python%{_python_buildid}-PrettyTable
BuildRequires: python%{_python_buildid}-Sphinx
BuildRequires: rdma-core-devel
BuildRequires: liblz4-devel >= 1.7
-BuildRequires: rdma-core-devel
%endif
%if 0%{?fedora} || 0%{?rhel}
Requires: systemd
@@ -212,7 +210,7 @@ BuildRequires: boost-random
BuildRequires: btrfs-progs
BuildRequires: nss-devel
BuildRequires: keyutils-libs-devel
-# RDMA is no longer built on 32-bit ARM: see #1484155
+# RDMA is no longer built on 32-bit ARM: see rhbz#1484155
%ifnarch %{arm}
BuildRequires: rdma-core-devel
%endif
@@ -226,7 +224,8 @@ BuildRequires: python%{_python_buildid}-sphinx
BuildRequires: lz4-devel >= 1.7
%endif
# python34-... for RHEL, python3-... for all other supported distros
-%if ( 0%{?rhel} && 0%{?rhel} <= 7 )
+%if %{with python3}
+%if 0%{?rhel}
BuildRequires: python34-devel
BuildRequires: python34-setuptools
BuildRequires: python34-Cython
@@ -235,6 +234,7 @@ BuildRequires: python3-devel
BuildRequires: python3-setuptools
BuildRequires: python3-Cython
%endif
+%endif
# distro-conditional make check dependencies
%if 0%{with make_check}
%if 0%{?fedora} || 0%{?rhel}
@@ -394,6 +394,12 @@ Requires: python%{_python_buildid}-jinja2
Requires: python%{_python_buildid}-pecan
Requires: python%{_python_buildid}-werkzeug
Requires: pyOpenSSL%{_python_buildid}
+%if 0%{?fedora}
+Requires: python%{_python_buildid}-bcrypt
+%endif
+%if 0%{?rhel}
+Requires: py-bcrypt
+%endif
%endif
%if 0%{?suse_version}
Requires: python%{_python_buildid}-CherryPy
@@ -562,6 +568,7 @@ This package contains Python 2 libraries for interacting with Cephs RADOS
gateway.
%endif
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-rgw
Summary: Python 3 libraries for the RADOS gateway
%if 0%{?suse_version}
@@ -572,6 +579,7 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
%description -n python%{python3_pkgversion}-rgw
This package contains Python 3 libraries for interacting with Cephs RADOS
gateway.
+%endif
%if 0%{with python2}
%package -n python-rados
@@ -586,6 +594,7 @@ This package contains Python 2 libraries for interacting with Cephs RADOS
object store.
%endif
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-rados
Summary: Python 3 libraries for the RADOS object store
%if 0%{?suse_version}
@@ -596,6 +605,7 @@ Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-rados
This package contains Python 3 libraries for interacting with Cephs RADOS
object store.
+%endif
%package -n libradosstriper1
Summary: RADOS striping interface
@@ -669,6 +679,7 @@ This package contains Python 2 libraries for interacting with Cephs RADOS
block device.
%endif
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-rbd
Summary: Python 3 libraries for the RADOS block device
%if 0%{?suse_version}
@@ -679,6 +690,7 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
%description -n python%{python3_pkgversion}-rbd
This package contains Python 3 libraries for interacting with Cephs RADOS
block device.
+%endif
%package -n libcephfs2
Summary: Ceph distributed file system client library
@@ -727,6 +739,7 @@ This package contains Python 2 libraries for interacting with Cephs distributed
file system.
%endif
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-cephfs
Summary: Python 3 libraries for Ceph distributed file system
%if 0%{?suse_version}
@@ -737,8 +750,9 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
%description -n python%{python3_pkgversion}-cephfs
This package contains Python 3 libraries for interacting with Cephs distributed
file system.
+%endif
-%if 0%{with python2}
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-ceph-argparse
Summary: Python 3 utility libraries for Ceph CLI
%if 0%{?suse_version}
@@ -919,7 +933,11 @@ cmake .. \
-DCMAKE_INSTALL_INCLUDEDIR=%{_includedir} \
-DWITH_EMBEDDED=OFF \
-DWITH_MANPAGE=ON \
+%if %{with python3}
-DWITH_PYTHON3=ON \
+%else
+ -DWITH_PYTHON3=OFF \
+%endif
-DWITH_MGR_DASHBOARD_FRONTEND=OFF \
%if %{with python2}
-DWITH_PYTHON2=ON \
@@ -927,7 +945,7 @@ cmake .. \
-DWITH_PYTHON2=OFF \
-DMGR_PYTHON_VERSION=3 \
%endif
-%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
+%if ( ( 0%{?rhel} && 0%{?rhel} < 8) && ! 0%{?centos} )
-DWITH_SUBMAN=ON \
%endif
%if 0%{without ceph_test_package}
@@ -955,13 +973,15 @@ cmake .. \
%else
-DWITH_BOOST_CONTEXT=OFF \
%endif
-%ifnarch %{arm}
+%ifarch %{arm}
-DWITH_RDMA=OFF \
%endif
+ -DWITH_DPDK=ON \
-DBOOST_J=$CEPH_SMP_NCPUS
make "$CEPH_MFLAGS_JOBS"
+
%if 0%{with make_check}
%check
# run in-tree unittests
@@ -1006,7 +1026,7 @@ install -m 0644 -D udev/95-ceph-osd.rules %{buildroot}%{_udevrulesdir}/95-ceph-o
#set up placeholder directories
mkdir -p %{buildroot}%{_sysconfdir}/ceph
-mkdir -p %{buildroot}%{_rundir}/ceph
+mkdir -p %{buildroot}%{_rundir}ceph
mkdir -p %{buildroot}%{_localstatedir}/log/ceph
mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/tmp
mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/mon
@@ -1025,6 +1045,9 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%py3_compile %{buildroot}%{python3_sitelib}
%endif
+%clean
+rm -rf %{buildroot}
+
#################################################################################
# files and systemd scriptlets
#################################################################################
@@ -1071,7 +1094,8 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%if 0%{with python2}
%{python_sitelib}/ceph_detect_init*
%{python_sitelib}/ceph_disk*
-%else
+%endif
+%if 0%{with python3}
%{python3_sitelib}/ceph_detect_init*
%{python3_sitelib}/ceph_disk*
%endif
@@ -1079,7 +1103,8 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%dir %{python_sitelib}/ceph_volume
%{python_sitelib}/ceph_volume/*
%{python_sitelib}/ceph_volume-*
-%else
+%endif
+%if 0%{with python3}
%dir %{python3_sitelib}/ceph_volume
%{python3_sitelib}/ceph_volume/*
%{python3_sitelib}/ceph_volume-*
@@ -1102,6 +1127,7 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/bootstrap-rbd
%post base
+/sbin/ldconfig
%if 0%{?suse_version}
%fillup_only
if [ $1 -eq 1 ] ; then
@@ -1124,6 +1150,7 @@ fi
%endif
%postun base
+/sbin/ldconfig
test -n "$FIRST_ARG" || FIRST_ARG=$1
%if 0%{?suse_version}
DISABLE_RESTART_ON_UPDATE="yes"
@@ -1171,7 +1198,6 @@ fi
%{_bindir}/rbd-replay-prep
%endif
%{_bindir}/ceph-post-file
-%{_bindir}/ceph-brag
%{_tmpfilesdir}/ceph-common.conf
%{_mandir}/man8/ceph-authtool.8*
%{_mandir}/man8/ceph-conf.8*
@@ -1202,7 +1228,8 @@ fi
%if 0%{with python2}
%{python_sitelib}/ceph_argparse.py*
%{python_sitelib}/ceph_daemon.py*
-%else
+%endif
+%if 0%{with python3}
%{python3_sitelib}/ceph_argparse.py
%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
%{python3_sitelib}/ceph_daemon.py
@@ -1529,7 +1556,7 @@ fi
%{_mandir}/man8/ceph-bluestore-tool.8*
%{_mandir}/man8/ceph-volume.8*
%{_mandir}/man8/ceph-volume-systemd.8*
-%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
+%if ( ( 0%{?rhel} && 0%{?rhel} < 8) && ! 0%{?centos} )
%attr(0755,-,-) %{_sysconfdir}/cron.hourly/subman
%endif
%{_unitdir}/ceph-osd@.service
@@ -1543,11 +1570,6 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl preset ceph-osd(a)\*.service ceph-volume(a)\*.service ceph-osd.target >/dev/null 2>&1 || :
fi
-%if 0%{?sysctl_apply}
- %sysctl_apply 90-ceph-osd.conf
-%else
- /usr/lib/systemd/systemd-sysctl %{_sysctldir}/90-ceph-osd.conf > /dev/null 2>&1 || :
-%endif
%endif
%if 0%{?fedora} || 0%{?rhel}
%systemd_post ceph-osd(a)\*.service ceph-volume(a)\*.service ceph-osd.target
@@ -1555,6 +1577,11 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl start ceph-osd.target >/dev/null 2>&1 || :
fi
+%if 0%{?sysctl_apply}
+ %sysctl_apply 90-ceph-osd.conf
+%else
+ /usr/lib/systemd/systemd-sysctl %{_sysctldir}/90-ceph-osd.conf > /dev/null 2>&1 || :
+%endif
%preun osd
%if 0%{?suse_version}
@@ -1629,9 +1656,11 @@ fi
%{python_sitearch}/rados-*.egg-info
%endif
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-rados
%{python3_sitearch}/rados.cpython*.so
%{python3_sitearch}/rados-*.egg-info
+%endif
%ldconfig_scriptlets -n libradosstriper1
%files -n libradosstriper1
@@ -1680,9 +1709,11 @@ fi
%{python_sitearch}/rgw-*.egg-info
%endif
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-rgw
%{python3_sitearch}/rgw.cpython*.so
%{python3_sitearch}/rgw-*.egg-info
+%endif
%if 0%{with python2}
%files -n python-rbd
@@ -1690,9 +1721,11 @@ fi
%{python_sitearch}/rbd-*.egg-info
%endif
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-rbd
%{python3_sitearch}/rbd.cpython*.so
%{python3_sitearch}/rbd-*.egg-info
+%endif
%ldconfig_scriptlets -n libcephfs2
%files -n libcephfs2
@@ -1711,13 +1744,15 @@ fi
%{python_sitelib}/ceph_volume_client.py*
%endif
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-cephfs
%{python3_sitearch}/cephfs.cpython*.so
%{python3_sitearch}/cephfs-*.egg-info
%{python3_sitelib}/ceph_volume_client.py
%{python3_sitelib}/__pycache__/ceph_volume_client.cpython*.py*
+%endif
-%if 0%{with python2}
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-ceph-argparse
%{python3_sitelib}/ceph_argparse.py
%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
@@ -1815,7 +1850,7 @@ fi
rm -f ${FILE_CONTEXT}.pre
# The fixfiles command won't fix label for /var/run/ceph
-/usr/sbin/restorecon -R %{_rundir}/ceph > /dev/null 2>&1
+/usr/sbin/restorecon -R %{_rundir}ceph > /dev/null 2>&1
# Start the daemons iff they were running before
if test $STATUS -eq 0; then
@@ -1851,7 +1886,7 @@ if [ $1 -eq 0 ]; then
/usr/sbin/fixfiles -C ${FILE_CONTEXT}.pre restore 2> /dev/null
rm -f ${FILE_CONTEXT}.pre
# The fixfiles command won't fix label for /var/run/ceph
- /usr/sbin/restorecon -R %{_rundir}/ceph > /dev/null 2>&1
+ /usr/sbin/restorecon -R %{_rundir}ceph > /dev/null 2>&1
# Start the daemons if they were running before
if test $STATUS -eq 0; then
@@ -1870,9 +1905,18 @@ exit 0
%changelog
+* Thu Jun 28 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.2.0-3
+- New release (1:13.2.0-3)
+
* Tue Jun 19 2018 Miro Hronok <mhroncok(a)redhat.com> - 1:13.1.0-2
- Rebuilt for Python 3.7
+* Thu May 31 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.2.0-1
+- New release (1:13.2.0-1)
+
+* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-2
+- New release (1:13.1.0-2) +crypto_plugins
+
* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-1
- New release (1:13.1.0-1)
diff --git a/sources b/sources
index e0ac550..ae5b656 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-13.1.0.tar.gz) = d4feb0b0bac5e30232eddcc3f69c1d939f0b5268491acecba9a58dcadbbdaab6367f415ede924ca84de9a89a686410613cdc631f8aff15aec0eb4531138eec7a
+SHA512 (ceph-13.2.0.tar.gz) = 2524ad5a868024463e1a1edc5bd9c3f7bc1f58e47bf8d331902ed53206120abc6a427dd9c1d657131e67efd955d6ed8052b6c3fed8658aa905d58ac58ee62973
commit 200f2e06d2a2cdc3ac4483467e9aff7ded9599c6
Author: Miro Hronok <miro(a)hroncok.cz>
Date: Tue Jun 19 10:41:00 2018 +0200
Rebuilt for Python 3.7
diff --git a/ceph.spec b/ceph.spec
index 1fa45d7..8b0de11 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -85,7 +85,7 @@
#################################################################################
Name: ceph
Version: 13.1.0
-Release: 1%{?dist}
+Release: 2%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -1870,6 +1870,9 @@ exit 0
%changelog
+* Tue Jun 19 2018 Miro Hronok <mhroncok(a)redhat.com> - 1:13.1.0-2
+- Rebuilt for Python 3.7
+
* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-1
- New release (1:13.1.0-1)
commit b79737d3dfe8949eee6764ee6298ee6a715a3042
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Tue May 8 11:57:37 2018 -0400
New release (1:13.1.0-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch b/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
new file mode 100644
index 0000000..ffc38da
--- /dev/null
+++ b/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
@@ -0,0 +1,165 @@
+--- ceph-13.1.0/src/mon/LogMonitor.cc.orig 2018-05-08 08:12:50.101933925 -0400
++++ ceph-13.1.0/src/mon/LogMonitor.cc 2018-05-08 08:12:59.470933925 -0400
+@@ -62,7 +62,6 @@
+ {
+ dout(10) << "create_initial -- creating initial map" << dendl;
+ LogEntry e;
+- memset(&e.who, 0, sizeof(e.who));
+ e.name = g_conf->name;
+ e.stamp = ceph_clock_now();
+ e.prio = CLOG_INFO;
+--- ceph-13.1.0/src/mon/MonMap.h.orig 2018-05-08 08:13:08.173933925 -0400
++++ ceph-13.1.0/src/mon/MonMap.h 2018-05-08 08:13:17.391933925 -0400
+@@ -118,7 +118,6 @@
+
+ MonMap()
+ : epoch(0) {
+- memset(&fsid, 0, sizeof(fsid));
+ }
+
+ uuid_d& get_fsid() { return fsid; }
+--- ceph-13.1.0/src/os/filestore/HashIndex.h.orig 2018-05-08 08:13:35.377933925 -0400
++++ ceph-13.1.0/src/os/filestore/HashIndex.h 2018-05-08 08:13:59.853933925 -0400
+@@ -395,7 +395,7 @@
+
+ struct CmpPairBitwise {
+ bool operator()(const pair<string, ghobject_t>& l,
+- const pair<string, ghobject_t>& r)
++ const pair<string, ghobject_t>& r) const
+ {
+ if (l.first < r.first)
+ return true;
+@@ -408,7 +408,7 @@
+ };
+
+ struct CmpHexdigitStringBitwise {
+- bool operator()(const string& l, const string& r) {
++ bool operator()(const string& l, const string& r) const {
+ return reverse_hexdigit_bits_string(l) < reverse_hexdigit_bits_string(r);
+ }
+ };
+--- ceph-13.1.0/src/os/filestore/LFNIndex.h.orig 2018-05-08 08:14:06.721933925 -0400
++++ ceph-13.1.0/src/os/filestore/LFNIndex.h 2018-05-08 08:14:34.671933925 -0400
+@@ -63,7 +63,7 @@
+ out: \
+ complete_inject_failure(); \
+ return r; \
+- } catch (RetryException) { \
++ } catch (RetryException&) { \
+ failed = true; \
+ } catch (...) { \
+ ceph_abort(); \
+--- ceph-13.1.0/src/client/Inode.h.orig 2018-05-08 08:08:26.305933925 -0400
++++ ceph-13.1.0/src/client/Inode.h 2018-05-08 08:08:38.029933925 -0400
+@@ -279,7 +279,6 @@
+ _ref(0), ll_ref(0)
+ {
+ memset(&dir_layout, 0, sizeof(dir_layout));
+- memset("a, 0, sizeof(quota));
+ }
+ ~Inode();
+
+--- ceph-13.1.0/src/osd/OSDMap.h.orig 2018-05-08 08:14:51.066933925 -0400
++++ ceph-13.1.0/src/osd/OSDMap.h 2018-05-08 08:15:14.148933925 -0400
+@@ -427,7 +427,6 @@
+ encode_features(0),
+ epoch(e), new_pool_max(-1), new_flags(-1), new_max_osd(-1),
+ have_crc(false), full_crc(0), inc_crc(0) {
+- memset(&fsid, 0, sizeof(fsid));
+ }
+ explicit Incremental(bufferlist &bl) {
+ bufferlist::iterator p = bl.begin();
+@@ -607,7 +606,6 @@
+ cached_up_osd_features(0),
+ crc_defined(false), crc(0),
+ crush(std::make_shared<CrushWrapper>()) {
+- memset(&fsid, 0, sizeof(fsid));
+ }
+
+ private:
+--- ceph-13.1.0/src/common/cmdparse.h.orig 2018-05-08 08:09:17.772933925 -0400
++++ ceph-13.1.0/src/common/cmdparse.h 2018-05-08 08:09:36.500933925 -0400
+@@ -54,7 +54,7 @@
+ try {
+ val = boost::get<T>(cmdmap.find(k)->second);
+ return true;
+- } catch (boost::bad_get) {
++ } catch (boost::bad_get&) {
+ handle_bad_get(cct, k, typeid(T).name());
+ }
+ }
+--- ceph-13.1.0/src/messages/MClientReply.h.orig 2018-05-08 08:10:22.281933925 -0400
++++ ceph-13.1.0/src/messages/MClientReply.h 2018-05-08 08:10:46.660933925 -0400
+@@ -187,7 +187,7 @@
+ if (features & CEPH_FEATURE_MDS_QUOTA)
+ decode(quota, p);
+ else
+- memset("a, 0, sizeof(quota));
++ quota = quota_info_t{};
+
+ if ((features & CEPH_FEATURE_FS_FILE_LAYOUT_V2))
+ decode(layout.pool_ns, p);
+--- ceph-13.1.0/src/messages/MMonSubscribeAck.h.orig 2018-05-08 08:10:58.478933925 -0400
++++ ceph-13.1.0/src/messages/MMonSubscribeAck.h 2018-05-08 08:11:17.484933925 -0400
+@@ -23,7 +23,6 @@
+
+ MMonSubscribeAck() : Message(CEPH_MSG_MON_SUBSCRIBE_ACK),
+ interval(0) {
+- memset(&fsid, 0, sizeof(fsid));
+ }
+ MMonSubscribeAck(uuid_d& f, int i) : Message(CEPH_MSG_MON_SUBSCRIBE_ACK),
+ interval(i), fsid(f) { }
+--- ceph-13.1.0/src/mgr/DaemonState.h.orig 2018-05-08 08:12:16.521933925 -0400
++++ ceph-13.1.0/src/mgr/DaemonState.h 2018-05-08 08:12:40.995933925 -0400
+@@ -130,7 +130,7 @@
+ auto p = config_defaults_bl.begin();
+ try {
+ decode(config_defaults, p);
+- } catch (buffer::error e) {
++ } catch (buffer::error& e) {
+ }
+ }
+ return config_defaults;
+--- ceph-13.1.0/cmake/modules/BuildDPDK.cmake.orig 2018-05-08 08:41:02.168933925 -0400
++++ ceph-13.1.0/cmake/modules/BuildDPDK.cmake 2018-05-08 08:41:47.411933925 -0400
+@@ -71,7 +71,7 @@
+ BUILD_IN_SOURCE 1
+ INSTALL_COMMAND "true")
+ ExternalProject_Add_Step(dpdk-ext patch-config
+- COMMAND ${CMAKE_MODULE_PATH}/patch-dpdk-conf.sh ${dpdk_dir} ${machine}
++ COMMAND ${CMAKE_MODULE_PATH}/patch-dpdk-conf.sh ${dpdk_dir} ${machine} ${arch}
+ DEPENDEES configure
+ DEPENDERS build)
+ # easier to adjust the config
+@@ -86,7 +86,7 @@
+ # target
+ file(MAKE_DIRECTORY ${DPDK_INCLUDE_DIR})
+ foreach(c
+- pci bus_pci
++ bus_pci pci
+ eal
+ mempool mempool_ring mempool_stack ring)
+ add_library(dpdk::${c} STATIC IMPORTED)
+--- ceph-13.1.0/cmake/modules/patch-dpdk-conf.sh.orig 2018-05-08 08:42:01.089933925 -0400
++++ ceph-13.1.0/cmake/modules/patch-dpdk-conf.sh 2018-05-08 08:43:11.781933925 -0400
+@@ -15,8 +15,12 @@
+ shift
+ machine=$1
+ shift
++arch=$1
++shift
+
+ setconf CONFIG_RTE_MACHINE "${machine}"
++setconf CONFIG_RTE_ARCH "${arch}"
++
+ # Disable experimental features
+ setconf CONFIG_RTE_NEXT_ABI n
+ setconf CONFIG_RTE_LIBRTE_MBUF_OFFLOAD n
+@@ -38,6 +42,7 @@
+ setconf CONFIG_RTE_LIBRTE_VMXNET3_PMD n
+ setconf CONFIG_RTE_LIBRTE_PMD_VHOST n
+ setconf CONFIG_RTE_APP_EVENTDEV n
++setconf CONFIG_RTE_MAX_VFIO_GROUPS 64
+
+ # no test
+ setconf CONFIG_RTE_APP_TEST n
diff --git a/0001-src-rocksdb-util-murmurhash.patch b/0001-src-rocksdb-util-murmurhash.patch
deleted file mode 100644
index f67a0ff..0000000
--- a/0001-src-rocksdb-util-murmurhash.patch
+++ /dev/null
@@ -1,13 +0,0 @@
---- ceph-12.1.1.orig/src/rocksdb/util/murmurhash.cc 2017-04-27 01:13:46.000000000 +0100
-+++ ceph-12.1.1.orig/src/rocksdb/util/murmurhash.cc 2017-07-25 11:37:28.910266684 +0100
-@@ -113,8 +113,8 @@ unsigned int MurmurHash2 ( const void *
-
- switch(len)
- {
-- case 3: h ^= data[2] << 16;
-- case 2: h ^= data[1] << 8;
-+ case 3: h ^= data[2] << 16; // fallthrough
-+ case 2: h ^= data[1] << 8; // fallthrough
- case 1: h ^= data[0];
- h *= m;
- };
diff --git a/0002-cmake-Support-ppc64.patch b/0002-cmake-Support-ppc64.patch
deleted file mode 100644
index 6b09549..0000000
--- a/0002-cmake-Support-ppc64.patch
+++ /dev/null
@@ -1,127 +0,0 @@
-From 2f0a7153460acc3f21462236f470ec3471fa2ee1 Mon Sep 17 00:00:00 2001
-From: Boris Ranto <branto(a)redhat.com>
-Date: Mon, 31 Jul 2017 19:50:23 +0200
-Subject: [PATCH] cmake: Support ppc64
-
-The ppc64 support requires a couple of changes:
-- adding the ppc64 support to cmake
-- changing optimized crc32 code to compile on ppc64le only
-- moving ifdef condition before crc32_align to avoid defined but not
-used warning
-
-Signed-off-by: Boris Ranto <branto(a)redhat.com>
----
- cmake/modules/SIMDExt.cmake | 15 ++++++++++++++-
- src/CMakeLists.txt | 4 +++-
- src/arch/ppc.c | 8 ++++----
- src/common/crc32c_ppc.c | 6 +++---
- 4 files changed, 24 insertions(+), 9 deletions(-)
-
-diff --git a/cmake/modules/SIMDExt.cmake b/cmake/modules/SIMDExt.cmake
-index 5330835..c47667d 100644
---- a/cmake/modules/SIMDExt.cmake
-+++ b/cmake/modules/SIMDExt.cmake
-@@ -109,7 +109,20 @@ elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "i386|i686|amd64|x86_64|AMD64")
- endif(CMAKE_SYSTEM_PROCESSOR MATCHES "i686|amd64|x86_64|AMD64")
- elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "(powerpc|ppc)64le")
- set(HAVE_PPC64LE 1)
-- message(STATUS " we are ppc64le")
-+ message(STATUS " we are ppc64")
-+ CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
-+ if(HAS_ALTIVEC)
-+ message(STATUS " HAS_ALTIVEC yes")
-+ set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maltivec")
-+ set(CMAKE_CXX_FLAGS "${CMAKE_C_FLAGS} -maltivec")
-+ endif()
-+ CHECK_C_COMPILER_FLAG("-mcpu=power8" HAVE_POWER8)
-+ if(HAVE_POWER8)
-+ message(STATUS " HAVE_POWER8 yes")
-+ endif()
-+elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "(power|ppc)64")
-+ set(HAVE_PPC64 1)
-+ message(STATUS " we are ppc64")
- CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
- if(HAS_ALTIVEC)
- message(STATUS " HAS_ALTIVEC yes")
-diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
-index 66f0c14..38d1913 100644
---- a/src/CMakeLists.txt
-+++ b/src/CMakeLists.txt
-@@ -568,7 +568,9 @@ if(HAVE_INTEL)
- endif(HAVE_GOOD_YASM_ELF64)
- elseif(HAVE_POWER8)
- list(APPEND libcommon_files
-- common/crc32c_ppc.c
-+ common/crc32c_ppc.c)
-+elseif(HAVE_PPC64LE)
-+ list(APPEND libcommon_files
- common/crc32c_ppc_asm.S
- common/crc32c_ppc_fast_zero_asm.S)
- endif(HAVE_INTEL)
-diff --git a/src/arch/ppc.c b/src/arch/ppc.c
-index f21e2fe..11d3a49 100644
---- a/src/arch/ppc.c
-+++ b/src/arch/ppc.c
-@@ -14,10 +14,10 @@ int ceph_arch_ppc_crc32 = 0;
-
- #include <stdio.h>
-
--#if __linux__ && __powerpc64__
-+#ifdef HAVE_PPC64LE
- #include <sys/auxv.h>
- #include <asm/cputable.h>
--#endif /* __linux__ && __powerpc64__ */
-+#endif /* HAVE_PPC64LE */
-
- #ifndef PPC_FEATURE2_VEC_CRYPTO
- #define PPC_FEATURE2_VEC_CRYPTO 0x02000000
-@@ -31,9 +31,9 @@ int ceph_arch_ppc_probe(void)
- {
- ceph_arch_ppc_crc32 = 0;
-
--#if __linux__ && __powerpc64__
-+#ifdef HAVE_PPC64LE
- if (getauxval(AT_HWCAP2) & PPC_FEATURE2_VEC_CRYPTO) ceph_arch_ppc_crc32 = 1;
--#endif /* __linux__ && __powerpc64__ */
-+#endif /* HAVE_PPC64LE */
-
- return 0;
- }
-diff --git a/src/common/crc32c_ppc.c b/src/common/crc32c_ppc.c
-index 43756e2..52fd1c4 100644
---- a/src/common/crc32c_ppc.c
-+++ b/src/common/crc32c_ppc.c
-@@ -20,6 +20,7 @@
- #define VMX_ALIGN 16
- #define VMX_ALIGN_MASK (VMX_ALIGN-1)
-
-+#ifdef HAVE_PPC64LE
- #ifdef REFLECT
- static unsigned int crc32_align(unsigned int crc, unsigned char const *p,
- unsigned long len)
-@@ -38,7 +39,6 @@ static unsigned int crc32_align(unsigned int crc, unsigned char const *p,
- }
- #endif
-
--#ifdef HAVE_POWER8
- static inline unsigned long polynomial_multiply(unsigned int a, unsigned int b) {
- vector unsigned int va = {a, 0, 0, 0};
- vector unsigned int vb = {b, 0, 0, 0};
-@@ -134,7 +134,7 @@ uint32_t ceph_crc32c_ppc(uint32_t crc, unsigned char const *data, unsigned len)
- return crc;
- }
-
--#else /* HAVE_POWER8 */
-+#else /* HAVE_PPC64LE */
-
- /* This symbol has to exist on non-ppc architectures (and on legacy
- * ppc systems using power7 or below) in order to compile properly
-@@ -145,4 +145,4 @@ uint32_t ceph_crc32c_ppc(uint32_t crc, unsigned char const *data, unsigned len)
- return 0;
- }
-
--#endif /* HAVE_POWER8 */
-+#endif /* HAVE_PPC64LE */
---
-2.9.4
-
diff --git a/0003-librbd-Conditionally-import-TrimRequest.cc.patch b/0003-librbd-Conditionally-import-TrimRequest.cc.patch
deleted file mode 100644
index 2f02efb..0000000
--- a/0003-librbd-Conditionally-import-TrimRequest.cc.patch
+++ /dev/null
@@ -1,31 +0,0 @@
-From 74a754690736f6608b0d4d9c807df0bd777a129d Mon Sep 17 00:00:00 2001
-From: Boris Ranto <branto(a)redhat.com>
-Date: Fri, 8 Dec 2017 00:21:38 +0100
-Subject: [PATCH] librbd: Conditionally import TrimRequest.cc
-
-We include TrimRequest.cc in librbd tests at two places:
- - operation/test_mock_TrimRequest.cc
- - operation/test_mock_ResizeRequest.cc
-
-That causes linking errors when doing the builds because some of the
-structures are defined twice.
-
-Signed-off-by: Boris Ranto <branto(a)redhat.com>
----
- src/librbd/operation/TrimRequest.cc | 2 ++
- 1 file changed, 2 insertions(+)
-
-diff --git a/src/librbd/operation/TrimRequest.cc b/src/librbd/operation/TrimRequest.cc
-index 28f2deb..929ca51 100644
---- a/src/librbd/operation/TrimRequest.cc
-+++ b/src/librbd/operation/TrimRequest.cc
-@@ -362,4 +362,6 @@ void TrimRequest<I>::send_finish(int r) {
- } // namespace operation
- } // namespace librbd
-
-+#ifndef TEST_F
- template class librbd::operation::TrimRequest<librbd::ImageCtx>;
-+#endif
---
-2.9.5
-
diff --git a/0004-cmake-modules-BuildBoost.cmake.patch b/0004-cmake-modules-BuildBoost.cmake.patch
deleted file mode 100644
index 644a6c4..0000000
--- a/0004-cmake-modules-BuildBoost.cmake.patch
+++ /dev/null
@@ -1,11 +0,0 @@
---- ceph-12.2.3/cmake/modules/BuildBoost.cmake.orig 2018-02-21 14:13:32.527899631 -0500
-+++ ceph-12.2.3/cmake/modules/BuildBoost.cmake 2018-02-21 14:14:04.796899631 -0500
-@@ -75,7 +75,7 @@
- set(install_command
- ${b2} install)
- set(boost_root_dir "${CMAKE_BINARY_DIR}/boost")
-- if(EXISTS "${PROJECT_SOURCE_DIR}/src/boost/libs/config/include/boost/config.hpp")
-+ if(EXISTS "${PROJECT_SOURCE_DIR}/src/boost/boost/config.hpp")
- message(STATUS "boost already in src")
- set(source_dir
- SOURCE_DIR "${PROJECT_SOURCE_DIR}/src/boost")
diff --git a/0005-src-rocksdb-table-block.h.patch b/0005-src-rocksdb-table-block.h.patch
deleted file mode 100644
index 8acb401..0000000
--- a/0005-src-rocksdb-table-block.h.patch
+++ /dev/null
@@ -1,155 +0,0 @@
---- ceph-12.2.3/src/rocksdb/table/block.h.orig 2018-02-22 07:49:38.044899631 -0500
-+++ ceph-12.2.3/src/rocksdb/table/block.h 2018-02-22 07:58:55.855899631 -0500
-@@ -65,7 +65,8 @@
-
- // Create bitmap and set all the bits to 0
- bitmap_ = new std::atomic<uint32_t>[bitmap_size];
-- memset(bitmap_, 0, bitmap_size * kBytesPersEntry);
-+ // memset(bitmap_, 0, bitmap_size * kBytesPersEntry);
-+ { unsigned i = 0; for (; i < bitmap_size;) bitmap_[i++] = 0; }
-
- RecordTick(GetStatistics(), READ_AMP_TOTAL_READ_BYTES,
- num_bits_needed << bytes_per_bit_pow_);
---- ceph-12.2.3/src/rocksdb/db/c.cc.orig 2018-02-22 08:14:56.033899631 -0500
-+++ ceph-12.2.3/src/rocksdb/db/c.cc 2018-02-22 10:06:39.759899631 -0500
-@@ -1322,11 +1322,6 @@
- b->rep.PutLogData(Slice(blob, len));
- }
-
--void rocksdb_writebatch_iterate(
-- rocksdb_writebatch_t* b,
-- void* state,
-- void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
-- void (*deleted)(void*, const char* k, size_t klen)) {
- class H : public WriteBatch::Handler {
- public:
- void* state_;
-@@ -1339,6 +1334,12 @@
- (*deleted_)(state_, key.data(), key.size());
- }
- };
-+
-+void rocksdb_writebatch_iterate(
-+ rocksdb_writebatch_t* b,
-+ void* state,
-+ void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
-+ void (*deleted)(void*, const char* k, size_t klen)) {
- H handler;
- handler.state_ = state;
- handler.put_ = put;
-@@ -1579,18 +1580,6 @@
- void* state,
- void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
- void (*deleted)(void*, const char* k, size_t klen)) {
-- class H : public WriteBatch::Handler {
-- public:
-- void* state_;
-- void (*put_)(void*, const char* k, size_t klen, const char* v, size_t vlen);
-- void (*deleted_)(void*, const char* k, size_t klen);
-- virtual void Put(const Slice& key, const Slice& value) override {
-- (*put_)(state_, key.data(), key.size(), value.data(), value.size());
-- }
-- virtual void Delete(const Slice& key) override {
-- (*deleted_)(state_, key.data(), key.size());
-- }
-- };
- H handler;
- handler.state_ = state;
- handler.put_ = put;
-@@ -2532,13 +2521,9 @@
- delete filter;
- }
-
--rocksdb_filterpolicy_t* rocksdb_filterpolicy_create_bloom_format(int bits_per_key, bool original_format) {
-- // Make a rocksdb_filterpolicy_t, but override all of its methods so
-- // they delegate to a NewBloomFilterPolicy() instead of user
-- // supplied C functions.
-- struct Wrapper : public rocksdb_filterpolicy_t {
-+ struct WrapperFP : public rocksdb_filterpolicy_t {
- const FilterPolicy* rep_;
-- ~Wrapper() { delete rep_; }
-+ ~WrapperFP() { delete rep_; }
- const char* Name() const override { return rep_->Name(); }
- void CreateFilter(const Slice* keys, int n,
- std::string* dst) const override {
-@@ -2549,11 +2534,16 @@
- }
- static void DoNothing(void*) { }
- };
-- Wrapper* wrapper = new Wrapper;
-+
-+rocksdb_filterpolicy_t* rocksdb_filterpolicy_create_bloom_format(int bits_per_key, bool original_format) {
-+ // Make a rocksdb_filterpolicy_t, but override all of its methods so
-+ // they delegate to a NewBloomFilterPolicy() instead of user
-+ // supplied C functions.
-+ WrapperFP* wrapper = new WrapperFP;
- wrapper->rep_ = NewBloomFilterPolicy(bits_per_key, original_format);
- wrapper->state_ = nullptr;
- wrapper->delete_filter_ = nullptr;
-- wrapper->destructor_ = &Wrapper::DoNothing;
-+ wrapper->destructor_ = &WrapperFP::DoNothing;
- return wrapper;
- }
-
-@@ -2889,10 +2879,9 @@
- delete st;
- }
-
--rocksdb_slicetransform_t* rocksdb_slicetransform_create_fixed_prefix(size_t prefixLen) {
-- struct Wrapper : public rocksdb_slicetransform_t {
-+ struct WrapperST : public rocksdb_slicetransform_t {
- const SliceTransform* rep_;
-- ~Wrapper() { delete rep_; }
-+ ~WrapperST() { delete rep_; }
- const char* Name() const override { return rep_->Name(); }
- Slice Transform(const Slice& src) const override {
- return rep_->Transform(src);
-@@ -2903,31 +2892,20 @@
- bool InRange(const Slice& src) const override { return rep_->InRange(src); }
- static void DoNothing(void*) { }
- };
-- Wrapper* wrapper = new Wrapper;
-+
-+rocksdb_slicetransform_t* rocksdb_slicetransform_create_fixed_prefix(size_t prefixLen) {
-+ WrapperST* wrapper = new WrapperST;
- wrapper->rep_ = rocksdb::NewFixedPrefixTransform(prefixLen);
- wrapper->state_ = nullptr;
-- wrapper->destructor_ = &Wrapper::DoNothing;
-+ wrapper->destructor_ = &WrapperST::DoNothing;
- return wrapper;
- }
-
- rocksdb_slicetransform_t* rocksdb_slicetransform_create_noop() {
-- struct Wrapper : public rocksdb_slicetransform_t {
-- const SliceTransform* rep_;
-- ~Wrapper() { delete rep_; }
-- const char* Name() const override { return rep_->Name(); }
-- Slice Transform(const Slice& src) const override {
-- return rep_->Transform(src);
-- }
-- bool InDomain(const Slice& src) const override {
-- return rep_->InDomain(src);
-- }
-- bool InRange(const Slice& src) const override { return rep_->InRange(src); }
-- static void DoNothing(void*) { }
-- };
-- Wrapper* wrapper = new Wrapper;
-+ WrapperST* wrapper = new WrapperST;
- wrapper->rep_ = rocksdb::NewNoopTransform();
- wrapper->state_ = nullptr;
-- wrapper->destructor_ = &Wrapper::DoNothing;
-+ wrapper->destructor_ = &WrapperST::DoNothing;
- return wrapper;
- }
-
---- ceph-12.2.3/src/rocksdb/memtable/inlineskiplist.h.orig 2018-02-22 10:34:06.918899631 -0500
-+++ ceph-12.2.3/src/rocksdb/memtable/inlineskiplist.h 2018-02-22 10:34:44.145899631 -0500
-@@ -279,7 +279,7 @@
- // next_[0]. This is used for passing data from AllocateKey to Insert.
- void StashHeight(const int height) {
- assert(sizeof(int) <= sizeof(next_[0]));
-- memcpy(&next_[0], &height, sizeof(int));
-+ memcpy(static_cast<void*>(&next_[0]), &height, sizeof(int));
- }
-
- // Retrieves the value passed to StashHeight. Undefined after a call
diff --git a/ceph.spec b/ceph.spec
index 9017056..1fa45d7 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -17,57 +17,56 @@
%global _hardened_build 1
%bcond_without ocf
-%ifnarch armv7hl
-%bcond_without cephfs_java
-%else
-%bcond_with cephfs_java
-%endif
-%if 0%{?suse_version}
-%bcond_with ceph_test_package
-%else
-%bcond_without ceph_test_package
-%endif
%bcond_with make_check
%ifarch s390 s390x
%bcond_with tcmalloc
%else
%bcond_without tcmalloc
%endif
-%ifnarch armv7hl
-%bcond_with lowmem_builder
-%else
-%if 0%{?rhel}
-%ifnarch ppc64le
-%bcond_with lowmem_builder
-%else
-%bcond_without lowmem_builder
-%endif
-%else
-%bcond_without lowmem_builder
-%endif
-%endif
%if 0%{?fedora} || 0%{?rhel}
%bcond_without selinux
+%bcond_without ceph_test_package
+%bcond_without cephfs_java
+%bcond_without lttng
%endif
%if 0%{?suse_version}
%bcond_with selinux
+%bcond_with ceph_test_package
+%bcond_with cephfs_java
+#Compat macro for new _fillupdir macro introduced in Nov 2017
+%if ! %{defined _fillupdir}
+%global _fillupdir /var/adm/fillup-templates
%endif
-
-# LTTng-UST enabled on Fedora, RHEL 6+, and SLE (not openSUSE)
-%if 0%{?fedora} || 0%{?rhel} >= 6 || 0%{?suse_version}
-%if ! 0%{?is_opensuse}
+%if 0%{?is_opensuse}
+%bcond_without lttng
+%else
+%ifarch x86_64 aarch64
%bcond_without lttng
+%else
+%bcond_with lttng
%endif
%endif
+%endif
+%if 0%{?suse_version} >= 1500
+%bcond_with python2
+%else
+%bcond_without python2
+%endif
+%if 0%{without python2}
+%global _defined_if_python2_absent 1
+%endif
%if %{with selinux}
# get selinux policy version
-%{!?_selinux_policy_version: %global _selinux_policy_version %(sed -e 's,.*selinux-policy-\\([^/]*\\)/.*,\\1,' /usr/share/selinux/devel/policyhelp 2>/dev/null || echo 0.0.0)}
+%{!?_selinux_policy_version: %global _selinux_policy_version 0.0.0}
%endif
%{!?_udevrulesdir: %global _udevrulesdir /lib/udev/rules.d}
%{!?tmpfiles_create: %global tmpfiles_create systemd-tmpfiles --create}
%{!?python3_pkgversion: %global python3_pkgversion 3}
+# define _python_buildid macro which will expand to the empty string when
+# building with python2
+%global _python_buildid %{?_defined_if_python2_absent:%{python3_pkgversion}}
# unify libexec for all targets
%global _libexecdir %{_exec_prefix}/lib
@@ -85,7 +84,7 @@
# main package definition
#################################################################################
Name: ceph
-Version: 12.2.5
+Version: 13.1.0
Release: 1%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
@@ -102,12 +101,14 @@ Group: System/Filesystems
URL: http://ceph.com/
Source0: http://download.ceph.com/tarballs/%{name}-%{version}.tar.gz
# https://bugzilla.redhat.com/show_bug.cgi?id=1474773
-Patch001: 0001-src-rocksdb-util-murmurhash.patch
+Patch001: 0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
+#Patch001: 0001-src-rocksdb-util-murmurhash.patch
# https://bugzilla.redhat.com/show_bug.cgi?id=1474774
-Patch002: 0002-cmake-Support-ppc64.patch
-Patch003: 0003-librbd-Conditionally-import-TrimRequest.cc.patch
-Patch005: 0005-src-rocksdb-table-block.h.patch
+#Patch002: 0002-cmake-Support-ppc64.patch
+#Patch003: 0003-librbd-Conditionally-import-TrimRequest.cc.patch
+#Patch005: 0005-src-rocksdb-table-block.h.patch
%if 0%{?suse_version}
+# _insert_obs_source_lines_here
%if 0%{?is_opensuse}
ExclusiveArch: x86_64 aarch64 ppc64 ppc64le
%else
@@ -117,10 +118,10 @@ ExclusiveArch: x86_64 aarch64 ppc64le s390x
#################################################################################
# dependencies that apply across all distro families
#################################################################################
-Requires: ceph-osd = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mds = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mgr = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mon = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-osd = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mds = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mgr = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mon = %{_epoch_prefix}%{version}-%{release}
Requires(post): binutils
%if 0%{with cephfs_java}
BuildRequires: java-devel
@@ -129,50 +130,47 @@ BuildRequires: sharutils
%if 0%{with selinux}
BuildRequires: checkpolicy
BuildRequires: selinux-policy-devel
-BuildRequires: selinux-policy-doc
-%endif
-%if 0%{with make_check}
-%if 0%{?fedora} || 0%{?rhel}
-BuildRequires: python-cherrypy
-BuildRequires: python-werkzeug
-%endif
-%if 0%{?suse_version}
-BuildRequires: python-CherryPy
-BuildRequires: python-Werkzeug
-BuildRequires: python-numpy-devel
-%endif
-BuildRequires: python-coverage
-BuildRequires: python-pecan
-BuildRequires: socat
%endif
BuildRequires: bc
BuildRequires: gperf
-BuildRequires: cmake
+BuildRequires: cmake
BuildRequires: cryptsetup
BuildRequires: fuse-devel
+%if 0%{?rhel} == 7
+# devtoolset offers newer make and valgrind-devel, but the old ones are good
+# enough.
+BuildRequires: devtoolset-7-gcc-c++
+%else
BuildRequires: gcc-c++
+%endif
BuildRequires: gdbm
%if 0%{with tcmalloc}
BuildRequires: gperftools-devel >= 2.4
%endif
-BuildRequires: jq
+BuildRequires: jq
BuildRequires: leveldb-devel > 1.2
BuildRequires: libaio-devel
BuildRequires: libblkid-devel >= 2.17
BuildRequires: libcurl-devel
BuildRequires: libudev-devel
+BuildRequires: liboath-devel
BuildRequires: libtool
BuildRequires: libxml2-devel
+BuildRequires: libuuid-devel
BuildRequires: make
BuildRequires: parted
BuildRequires: perl
BuildRequires: pkgconfig
-BuildRequires: python
-BuildRequires: python-devel
-BuildRequires: python-nose
-BuildRequires: python-requests
-BuildRequires: python-virtualenv
+BuildRequires: procps
+BuildRequires: python%{_python_buildid}
+BuildRequires: python%{_python_buildid}-devel
+BuildRequires: python%{_python_buildid}-nose
+BuildRequires: python%{_python_buildid}-requests
+BuildRequires: python%{_python_buildid}-virtualenv
BuildRequires: snappy-devel
+%if 0%{with make_check}
+BuildRequires: socat
+%endif
BuildRequires: udev
BuildRequires: util-linux
BuildRequires: valgrind-devel
@@ -186,27 +184,31 @@ BuildRequires: yasm
# distro-conditional dependencies
#################################################################################
%if 0%{?suse_version}
-BuildRequires: pkgconfig(systemd)
+BuildRequires: pkgconfig(systemd)
BuildRequires: systemd-rpm-macros
BuildRequires: systemd
%{?systemd_requires}
PreReq: %fillup_prereq
BuildRequires: net-tools
BuildRequires: libbz2-devel
-BuildRequires: btrfsprogs
+BuildRequires: btrfsprogs
BuildRequires: mozilla-nss-devel
BuildRequires: keyutils-devel
-BuildRequires: libopenssl-devel
-BuildRequires: lsb-release
-BuildRequires: openldap2-devel
-BuildRequires: python-Cython
-BuildRequires: python-PrettyTable
-BuildRequires: python-Sphinx
-BuildRequires: rdma-core-devel
+BuildRequires: libopenssl-devel
+BuildRequires: lsb-release
+BuildRequires: openldap2-devel
+BuildRequires: cunit-devel
+BuildRequires: python%{_python_buildid}-base
+BuildRequires: python%{_python_buildid}-Cython
+BuildRequires: python%{_python_buildid}-PrettyTable
+BuildRequires: python%{_python_buildid}-Sphinx
+BuildRequires: rdma-core-devel
+BuildRequires: liblz4-devel >= 1.7
+BuildRequires: rdma-core-devel
%endif
%if 0%{?fedora} || 0%{?rhel}
Requires: systemd
-BuildRequires: boost-random
+BuildRequires: boost-random
BuildRequires: btrfs-progs
BuildRequires: nss-devel
BuildRequires: keyutils-libs-devel
@@ -214,12 +216,14 @@ BuildRequires: keyutils-libs-devel
%ifnarch %{arm}
BuildRequires: rdma-core-devel
%endif
-BuildRequires: openldap-devel
-BuildRequires: openssl-devel
-BuildRequires: redhat-lsb-core
-BuildRequires: Cython
-BuildRequires: python-prettytable
-BuildRequires: python-sphinx
+BuildRequires: openldap-devel
+BuildRequires: openssl-devel
+BuildRequires: redhat-lsb-core
+BuildRequires: CUnit-devel
+BuildRequires: Cython%{_python_buildid}
+BuildRequires: python%{_python_buildid}-prettytable
+BuildRequires: python%{_python_buildid}-sphinx
+BuildRequires: lz4-devel >= 1.7
%endif
# python34-... for RHEL, python3-... for all other supported distros
%if ( 0%{?rhel} && 0%{?rhel} <= 7 )
@@ -231,6 +235,33 @@ BuildRequires: python3-devel
BuildRequires: python3-setuptools
BuildRequires: python3-Cython
%endif
+# distro-conditional make check dependencies
+%if 0%{with make_check}
+%if 0%{?fedora} || 0%{?rhel}
+BuildRequires: python%{_python_buildid}-cherrypy
+BuildRequires: python%{_python_buildid}-routes
+BuildRequires: python%{_python_buildid}-pecan
+BuildRequires: python%{_python_buildid}-werkzeug
+BuildRequires: python%{_python_buildid}-tox
+BuildRequires: python%{_python_buildid}-coverage
+%if 0%{?fedora}
+BuildRequires: python%{_python_buildid}-bcrypt
+%endif
+%if 0%{?rhel}
+BuildRequires: py-bcrypt
+%endif
+%endif
+%if 0%{?suse_version}
+BuildRequires: python%{_python_buildid}-CherryPy
+BuildRequires: python%{_python_buildid}-Routes
+BuildRequires: python%{_python_buildid}-Werkzeug
+BuildRequires: python%{_python_buildid}-pecan
+BuildRequires: python%{_python_buildid}-numpy-devel
+BuildRequires: python%{_python_buildid}-bcrypt
+BuildRequires: python%{_python_buildid}-tox
+BuildRequires: python%{_python_buildid}-coverage
+%endif
+%endif
# lttng and babeltrace for rbd-replay-prep
%if %{with lttng}
%if 0%{?fedora} || 0%{?rhel}
@@ -239,7 +270,7 @@ BuildRequires: libbabeltrace-devel
%endif
%if 0%{?suse_version}
BuildRequires: lttng-ust-devel
-BuildRequires: babeltrace-devel
+BuildRequires: babeltrace-devel
%endif
%endif
%if 0%{?suse_version}
@@ -250,7 +281,7 @@ BuildRequires: expat-devel
%endif
#hardened-cc1
%if 0%{?fedora} || 0%{?rhel}
-BuildRequires: redhat-rpm-config
+BuildRequires: redhat-rpm-config
%endif
%description
@@ -262,33 +293,39 @@ on commodity hardware and delivers object, block and file system storage.
# subpackages
#################################################################################
%package base
-Summary: Ceph Base Package
+Summary: Ceph Base Package
%if 0%{?suse_version}
-Group: System/Filesystems
+Group: System/Filesystems
%endif
-Requires: ceph-common = %{_epoch_prefix}%{version}-%{release}
-Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
-Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
-Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
-Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-common = %{_epoch_prefix}%{version}-%{release}
+Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
+Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
+Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
+Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
%if 0%{with selinux}
-Requires: ceph-selinux = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-selinux = %{_epoch_prefix}%{version}-%{release}
%endif
Requires(post):/sbin/ldconfig
Requires(postun):/sbin/ldconfig
-Requires: python
-Requires: python-requests
-Requires: python-setuptools
-Requires: grep
-Requires: xfsprogs
-Requires: logrotate
-Requires: util-linux
-Requires: cryptsetup
-Requires: findutils
-Requires: psmisc
-Requires: which
-%if 0%{?suse_version}
-Recommends: ntp-daemon
+Requires: cryptsetup
+Requires: e2fsprogs
+Requires: findutils
+Requires: grep
+Requires: logrotate
+Requires: parted
+Requires: psmisc
+Requires: python%{_python_buildid}-requests
+Requires: python%{_python_buildid}-setuptools
+Requires: util-linux
+Requires: xfsprogs
+Requires: which
+%if 0%{?fedora} || 0%{?rhel}
+Requires: gdisk
+%endif
+%if 0%{?suse_version}
+Recommends: ntp-daemon
+Recommends: chrony
+Requires: gptfdisk
%endif
%description base
Base is the package that includes all the files shared amongst ceph servers
@@ -301,17 +338,18 @@ Group: System/Filesystems
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
-Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
-Requires: python-rbd = %{_epoch_prefix}%{version}-%{release}
-Requires: python-cephfs = %{_epoch_prefix}%{version}-%{release}
-Requires: python-rgw = %{_epoch_prefix}%{version}-%{release}
+Requires: python%{_python_buildid}-rados = %{_epoch_prefix}%{version}-%{release}
+Requires: python%{_python_buildid}-rbd = %{_epoch_prefix}%{version}-%{release}
+Requires: python%{_python_buildid}-cephfs = %{_epoch_prefix}%{version}-%{release}
+Requires: python%{_python_buildid}-rgw = %{_epoch_prefix}%{version}-%{release}
%if 0%{?fedora} || 0%{?rhel}
-Requires: python-prettytable
+Requires: python%{_python_buildid}-prettytable
+Requires: python%{_python_buildid}-requests
%endif
%if 0%{?suse_version}
-Requires: python-PrettyTable
+Requires: python%{_python_buildid}-PrettyTable
+Requires: python%{_python_buildid}-requests
%endif
-Requires: python-requests
%{?systemd_requires}
%if 0%{?suse_version}
Requires(pre): pwdutils
@@ -337,13 +375,6 @@ Summary: Ceph Monitor Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
-# For ceph-rest-api
-%if 0%{?fedora} || 0%{?rhel}
-Requires: python-flask
-%endif
-%if 0%{?suse_version}
-Requires: python-Flask
-%endif
%description mon
ceph-mon is the cluster monitor daemon for the Ceph distributed file
system. One or more instances of ceph-mon form a Paxos part-time
@@ -351,24 +382,29 @@ parliament cluster that provides extremely reliable and durable storage
of cluster membership, configuration, and state.
%package mgr
-Summary: Ceph Manager Daemon
+Summary: Ceph Manager Daemon
%if 0%{?suse_version}
-Group: System/Filesystems
+Group: System/Filesystems
%endif
-Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
%if 0%{?fedora} || 0%{?rhel}
-Requires: python-cherrypy
-Requires: python-jinja2
-Requires: python-werkzeug
-Requires: pyOpenSSL
+Requires: python%{_python_buildid}-cherrypy
+Requires: python%{_python_buildid}-routes
+Requires: python%{_python_buildid}-jinja2
+Requires: python%{_python_buildid}-pecan
+Requires: python%{_python_buildid}-werkzeug
+Requires: pyOpenSSL%{_python_buildid}
+%endif
+%if 0%{?suse_version}
+Requires: python%{_python_buildid}-CherryPy
+Requires: python%{_python_buildid}-Routes
+Requires: python%{_python_buildid}-Jinja2
+Requires: python%{_python_buildid}-Werkzeug
+Requires: python%{_python_buildid}-pecan
+Requires: python%{_python_buildid}-pyOpenSSL
+Requires: python%{_python_buildid}-bcrypt
+Recommends: python%{_python_buildid}-influxdb
%endif
-%if 0%{?suse_version}
-Requires: python-CherryPy
-Requires: python-jinja2
-Requires: python-Werkzeug
-Requires: python-pyOpenSSL
-%endif
-Requires: python-pecan
%description mgr
ceph-mgr enables python modules that provide services (such as the REST
module derived from Calamari) and expose CLI hooks. ceph-mgr gathers
@@ -455,15 +491,7 @@ Summary: Ceph Object Storage Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
-# for sgdisk, used by ceph-disk
-%if 0%{?fedora} || 0%{?rhel}
-Requires: gdisk
-%endif
-%if 0%{?suse_version}
-Requires: gptfdisk
-%endif
-Requires: parted
-Requires: lvm2
+Requires: lvm2
%description osd
ceph-osd is the object storage daemon for the Ceph distributed file
system. It is responsible for storing objects on a local file system
@@ -520,6 +548,7 @@ Obsoletes: librgw2-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use RADOS gateway client library.
+%if 0%{with python2}
%package -n python-rgw
Summary: Python 2 libraries for the RADOS gateway
%if 0%{?suse_version}
@@ -531,11 +560,12 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rgw
This package contains Python 2 libraries for interacting with Cephs RADOS
gateway.
+%endif
%package -n python%{python3_pkgversion}-rgw
Summary: Python 3 libraries for the RADOS gateway
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
@@ -543,6 +573,7 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
This package contains Python 3 libraries for interacting with Cephs RADOS
gateway.
+%if 0%{with python2}
%package -n python-rados
Summary: Python 2 libraries for the RADOS object store
%if 0%{?suse_version}
@@ -553,11 +584,12 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rados
This package contains Python 2 libraries for interacting with Cephs RADOS
object store.
+%endif
%package -n python%{python3_pkgversion}-rados
Summary: Python 3 libraries for the RADOS object store
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: python%{python3_pkgversion}
Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
@@ -623,10 +655,11 @@ Obsoletes: librbd1-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use RADOS block device.
+%if 0%{with python2}
%package -n python-rbd
Summary: Python 2 libraries for the RADOS block device
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
@@ -634,11 +667,12 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rbd
This package contains Python 2 libraries for interacting with Cephs RADOS
block device.
+%endif
%package -n python%{python3_pkgversion}-rbd
Summary: Python 3 libraries for the RADOS block device
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
@@ -677,24 +711,26 @@ Obsoletes: libcephfs2-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use Cephs distributed file system.
+%if 0%{with python2}
%package -n python-cephfs
Summary: Python 2 libraries for Ceph distributed file system
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
%if 0%{?suse_version}
-Recommends: python-rados = %{_epoch_prefix}%{version}-%{release}
+Recommends: python-rados = %{_epoch_prefix}%{version}-%{release}
%endif
Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-cephfs
This package contains Python 2 libraries for interacting with Cephs distributed
file system.
+%endif
%package -n python%{python3_pkgversion}-cephfs
Summary: Python 3 libraries for Ceph distributed file system
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
@@ -702,16 +738,18 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
This package contains Python 3 libraries for interacting with Cephs distributed
file system.
+%if 0%{with python2}
%package -n python%{python3_pkgversion}-ceph-argparse
Summary: Python 3 utility libraries for Ceph CLI
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
%description -n python%{python3_pkgversion}-ceph-argparse
This package contains types and routines for Python 3 used by the Ceph CLI as
well as the RESTful interface. These have to do with querying the daemons for
command-description information, validating user command input against those
descriptions, and submitting the command to the appropriate daemon.
+%endif
%if 0%{with ceph_test_package}
%package -n ceph-test
@@ -761,17 +799,18 @@ Group: System/Libraries
%endif
Requires: java
Requires: libcephfs_jni1 = %{_epoch_prefix}%{version}-%{release}
-Requires: junit
-BuildRequires: junit
+BuildRequires: junit
%description -n cephfs-java
This package contains the Java libraries for the Ceph File System.
%endif
%package -n rados-objclass-devel
-Summary: RADOS object class development kit
-Group: Development/Libraries
-Requires: librados2-devel = %{_epoch_prefix}%{version}-%{release}
+Summary: RADOS object class development kit
+%if 0%{?suse_version}
+Group: Development/Libraries/C and C++
+%endif
+Requires: librados2-devel = %{_epoch_prefix}%{version}-%{release}
%description -n rados-objclass-devel
This package contains libraries and headers needed to develop RADOS object
class plugins.
@@ -795,10 +834,11 @@ populated file-systems.
%endif
+%if 0%{with python2}
%package -n python-ceph-compat
Summary: Compatibility package for Cephs python libraries
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Obsoletes: python-ceph
Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
@@ -811,6 +851,7 @@ This is a compatibility package to accommodate python-ceph split into
python-rados, python-rbd, python-rgw and python-cephfs. Packages still
depending on python-ceph should be fixed to depend on python-rados,
python-rbd, python-rgw or python-cephfs instead.
+%endif
#################################################################################
# common
@@ -819,6 +860,11 @@ python-rbd, python-rgw or python-cephfs instead.
%autosetup -p1 -n %{name}-%{version}
%build
+
+%if 0%{?rhel} == 7
+. /opt/rh/devtoolset-7/enable
+%endif
+
%if 0%{with cephfs_java}
# Find jni.h
for i in /usr/{lib64,lib}/jvm/java/include{,/linux}; do
@@ -826,36 +872,39 @@ for i in /usr/{lib64,lib}/jvm/java/include{,/linux}; do
done
%endif
-%if %{with lowmem_builder}
+%if 0%{?suse_version}
+# the following setting fixed an OOM condition we once encountered in the OBS
RPM_OPT_FLAGS="$RPM_OPT_FLAGS --param ggc-min-expand=20 --param ggc-min-heapsize=32768"
%endif
-%ifnarch armv7hl
-export RPM_OPT_FLAGS=`echo $RPM_OPT_FLAGS | sed -e 's/i386/i486/'`
-%else
-export RPM_OPT_FLAGS=`echo $RPM_OPT_FLAGS | sed -e 's/i386/i486/' -e 's/-pipe//g'`
-%endif
export CPPFLAGS="$java_inc"
export CFLAGS="$RPM_OPT_FLAGS"
export CXXFLAGS="$RPM_OPT_FLAGS"
export LDFLAGS="$RPM_LD_FLAGS"
-env | sort
-
-%if %{with lowmem_builder}
-%ifnarch armv7hl
-%if 0%{?jobs} > 8
-%define _smp_mflags -j8
-%endif
-%else
-%define _smp_mflags -j1
-%endif
-%endif
+# Parallel build settings ...
+CEPH_MFLAGS_JOBS="%{?_smp_mflags}"
+CEPH_SMP_NCPUS=$(echo "$CEPH_MFLAGS_JOBS" | sed 's/-j//')
+%if 0%{?__isa_bits} == 32
+# 32-bit builds can use 3G memory max, which is not enough even for -j2
+CEPH_SMP_NCPUS="1"
+%endif
+# do not eat all memory
+echo "Available memory:"
+free -h
+echo "System limits:"
+ulimit -a
+if test -n "$CEPH_SMP_NCPUS" -a "$CEPH_SMP_NCPUS" -gt 1 ; then
+ mem_per_process=1800
+ max_mem=$(LANG=C free -m | sed -n "s|^Mem: *\([0-9]*\).*$|\1|p")
+ max_jobs="$(($max_mem / $mem_per_process))"
+ test "$CEPH_SMP_NCPUS" -gt "$max_jobs" && CEPH_SMP_NCPUS="$max_jobs" && echo "Warning: Reducing build parallelism to -j$max_jobs because of memory limits"
+ test "$CEPH_SMP_NCPUS" -le 0 && CEPH_SMP_NCPUS="1" && echo "Warning: Not using parallel build at all because of memory limits"
+fi
+export CEPH_SMP_NCPUS
+export CEPH_MFLAGS_JOBS="-j$CEPH_SMP_NCPUS"
-# unlimit _smp_mflags in system macro if not set above
-%define _smp_ncpus_max 0
-# extract the number of processors for use with cmake
-%define _smp_ncpus %(echo %{_smp_mflags} | sed 's/-j//')
+env | sort
mkdir build
cd build
@@ -871,7 +920,13 @@ cmake .. \
-DWITH_EMBEDDED=OFF \
-DWITH_MANPAGE=ON \
-DWITH_PYTHON3=ON \
- -DWITH_SYSTEMD=ON \
+ -DWITH_MGR_DASHBOARD_FRONTEND=OFF \
+%if %{with python2}
+ -DWITH_PYTHON2=ON \
+%else
+ -DWITH_PYTHON2=OFF \
+ -DMGR_PYTHON_VERSION=3 \
+%endif
%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
-DWITH_SUBMAN=ON \
%endif
@@ -903,21 +958,18 @@ cmake .. \
%ifnarch %{arm}
-DWITH_RDMA=OFF \
%endif
- -DBOOST_J=%{_smp_ncpus}
-
-make %{?_smp_mflags}
+ -DBOOST_J=$CEPH_SMP_NCPUS
+make "$CEPH_MFLAGS_JOBS"
%if 0%{with make_check}
%check
# run in-tree unittests
cd build
-ctest %{?_smp_mflags}
-
+ctest "$CEPH_MFLAGS_JOBS"
%endif
-
%install
pushd build
make DESTDIR=%{buildroot} install
@@ -929,10 +981,9 @@ install -m 0644 -D src/etc-rbdmap %{buildroot}%{_sysconfdir}/ceph/rbdmap
install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_sysconfdir}/sysconfig/ceph
%endif
%if 0%{?suse_version}
-install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_localstatedir}/adm/fillup-templates/sysconfig.%{name}
+install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_fillupdir}/sysconfig.%{name}
%endif
install -m 0644 -D systemd/ceph.tmpfiles.d %{buildroot}%{_tmpfilesdir}/ceph-common.conf
-install -m 0755 -D systemd/ceph %{buildroot}%{_sbindir}/rcceph
install -m 0644 -D systemd/50-ceph.preset %{buildroot}%{_libexecdir}/systemd/system-preset/50-ceph.preset
mkdir -p %{buildroot}%{_sbindir}
install -m 0644 -D src/logrotate.conf %{buildroot}%{_sysconfdir}/logrotate.d/ceph
@@ -989,7 +1040,6 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%{_libexecdir}/systemd/system-preset/50-ceph.preset
%{_sbindir}/ceph-create-keys
%{_sbindir}/ceph-disk
-%{_sbindir}/rcceph
%dir %{_libexecdir}/ceph
%{_libexecdir}/ceph/ceph_common.sh
%dir %{_libdir}/rados-classes
@@ -1012,17 +1062,28 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%config(noreplace) %{_sysconfdir}/sysconfig/ceph
%endif
%if 0%{?suse_version}
-%{_localstatedir}/adm/fillup-templates/sysconfig.*
+%{_fillupdir}/sysconfig.*
%config %{_sysconfdir}/sysconfig/SuSEfirewall2.d/services/ceph-mon
%config %{_sysconfdir}/sysconfig/SuSEfirewall2.d/services/ceph-osd-mds
%endif
%{_unitdir}/ceph-disk@.service
%{_unitdir}/ceph.target
+%if 0%{with python2}
%{python_sitelib}/ceph_detect_init*
%{python_sitelib}/ceph_disk*
+%else
+%{python3_sitelib}/ceph_detect_init*
+%{python3_sitelib}/ceph_disk*
+%endif
+%if 0%{with python2}
%dir %{python_sitelib}/ceph_volume
%{python_sitelib}/ceph_volume/*
%{python_sitelib}/ceph_volume-*
+%else
+%dir %{python3_sitelib}/ceph_volume
+%{python3_sitelib}/ceph_volume/*
+%{python3_sitelib}/ceph_volume-*
+%endif
%{_mandir}/man8/ceph-deploy.8*
%{_mandir}/man8/ceph-detect-init.8*
%{_mandir}/man8/ceph-create-keys.8*
@@ -1093,7 +1154,6 @@ fi
%{_bindir}/ceph-dencoder
%{_bindir}/ceph-rbdnamer
%{_bindir}/ceph-syn
-%{_bindir}/ceph-crush-location
%{_bindir}/cephfs-data-scan
%{_bindir}/cephfs-journal-tool
%{_bindir}/cephfs-table-tool
@@ -1139,8 +1199,15 @@ fi
%config %{_sysconfdir}/bash_completion.d/radosgw-admin
%config(noreplace) %{_sysconfdir}/ceph/rbdmap
%{_unitdir}/rbdmap.service
+%if 0%{with python2}
%{python_sitelib}/ceph_argparse.py*
%{python_sitelib}/ceph_daemon.py*
+%else
+%{python3_sitelib}/ceph_argparse.py
+%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
+%{python3_sitelib}/ceph_daemon.py
+%{python3_sitelib}/__pycache__/ceph_daemon.cpython*.py*
+%endif
%dir %{_udevrulesdir}
%{_udevrulesdir}/50-rbd.rules
%attr(3770,ceph,ceph) %dir %{_localstatedir}/log/ceph/
@@ -1282,11 +1349,8 @@ fi
%files mon
%{_bindir}/ceph-mon
-%{_bindir}/ceph-rest-api
%{_bindir}/ceph-monstore-tool
%{_mandir}/man8/ceph-mon.8*
-%{_mandir}/man8/ceph-rest-api.8*
-%{python_sitelib}/ceph_rest_api.py*
%{_unitdir}/ceph-mon@.service
%{_unitdir}/ceph-mon.target
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/mon
@@ -1559,9 +1623,11 @@ fi
%{_bindir}/librados-config
%{_mandir}/man8/librados-config.8*
+%if 0%{with python2}
%files -n python-rados
%{python_sitearch}/rados.so
%{python_sitearch}/rados-*.egg-info
+%endif
%files -n python%{python3_pkgversion}-rados
%{python3_sitearch}/rados.cpython*.so
@@ -1597,6 +1663,10 @@ fi
%ldconfig_scriptlets -n librgw2
%files -n librgw2
%{_libdir}/librgw.so.*
+%if %{with lttng}
+%{_libdir}/librgw_op_tp.so*
+%{_libdir}/librgw_rados_tp.so*
+%endif
%files -n librgw-devel
%dir %{_includedir}/rados
@@ -1604,17 +1674,21 @@ fi
%{_includedir}/rados/rgw_file.h
%{_libdir}/librgw.so
+%if 0%{with python2}
%files -n python-rgw
%{python_sitearch}/rgw.so
%{python_sitearch}/rgw-*.egg-info
+%endif
%files -n python%{python3_pkgversion}-rgw
%{python3_sitearch}/rgw.cpython*.so
%{python3_sitearch}/rgw-*.egg-info
+%if 0%{with python2}
%files -n python-rbd
%{python_sitearch}/rbd.so
%{python_sitearch}/rbd-*.egg-info
+%endif
%files -n python%{python3_pkgversion}-rbd
%{python3_sitearch}/rbd.cpython*.so
@@ -1630,10 +1704,12 @@ fi
%{_includedir}/cephfs/ceph_statx.h
%{_libdir}/libcephfs.so
+%if 0%{with python2}
%files -n python-cephfs
%{python_sitearch}/cephfs.so
%{python_sitearch}/cephfs-*.egg-info
%{python_sitelib}/ceph_volume_client.py*
+%endif
%files -n python%{python3_pkgversion}-cephfs
%{python3_sitearch}/cephfs.cpython*.so
@@ -1641,11 +1717,13 @@ fi
%{python3_sitelib}/ceph_volume_client.py
%{python3_sitelib}/__pycache__/ceph_volume_client.cpython*.py*
+%if 0%{with python2}
%files -n python%{python3_pkgversion}-ceph-argparse
%{python3_sitelib}/ceph_argparse.py
%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
%{python3_sitelib}/ceph_daemon.py
%{python3_sitelib}/__pycache__/ceph_daemon.cpython*.py*
+%endif
%if 0%{with ceph_test_package}
%files -n ceph-test
@@ -1667,13 +1745,7 @@ fi
%{_bindir}/ceph_rgw_multiparser
%{_bindir}/ceph_scratchtool
%{_bindir}/ceph_scratchtoolpp
-%{_bindir}/ceph_smalliobench
-%{_bindir}/ceph_smalliobenchdumb
-%{_bindir}/ceph_smalliobenchfs
-%{_bindir}/ceph_smalliobenchrbd
%{_bindir}/ceph_test_*
-%{_bindir}/ceph_tpbench
-%{_bindir}/ceph_xattr_bench
%{_bindir}/ceph-coverage
%{_bindir}/ceph-debugpack
%{_mandir}/man8/ceph-debugpack.8*
@@ -1790,12 +1862,17 @@ exit 0
%endif # with selinux
+%if 0%{with python2}
%files -n python-ceph-compat
# We need an empty %%files list for python-ceph-compat, to tell rpmbuild to
# actually build this meta package.
+%endif
%changelog
+* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-1
+- New release (1:13.1.0-1)
+
* Fri Apr 27 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-1
- New release (1:12.2.5-1)
diff --git a/sources b/sources
index 021832c..e0ac550 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-12.2.5.tar.gz) = bd78358590eb077afcc3c1e12864fadee906c95ac8afee25bdca6155f99b75f89d923159fcc13430abba21a89fee3e566177ff5f9ffabdd3b3f220d05602ab86
+SHA512 (ceph-13.1.0.tar.gz) = d4feb0b0bac5e30232eddcc3f69c1d939f0b5268491acecba9a58dcadbbdaab6367f415ede924ca84de9a89a686410613cdc631f8aff15aec0eb4531138eec7a
1
0
05 Sep '18
The package rpms/ceph.git has added or updated architecture specific content in its
spec file (ExclusiveArch/ExcludeArch or %ifarch/%ifnarch) in commit(s):
https://src.fedoraproject.org/cgit/rpms/ceph.git/commit/?id=b79737d3dfe8949…
https://src.fedoraproject.org/cgit/rpms/ceph.git/commit/?id=9b4d3225ea28de8…
https://src.fedoraproject.org/cgit/rpms/ceph.git/commit/?id=9094835645403e6….
Change:
-%ifnarch armv7hl
+%ifnarch %{arm}
+%ifarch %{arm}
Thanks.
Full change:
============
commit 71beeeca2eb50c06591ca051e5b5cb41399d2fa5
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Wed Sep 5 07:10:02 2018 -0400
New release (1:12.2.8-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/sources b/sources
index 47f3ab5..e781e91 100644
--- a/sources
+++ b/sources
@@ -1,5 +1 @@
-<<<<<<< HEAD
-SHA512 (ceph-12.2.7.tar.gz) = dde75a41c31cca54d4a680b0b93e28680a23f912a348cccca65e55e600c315bbfe3ced008d758fe8ba1e801a9d6e5f8079cc157ec5eef8916b86f487f42064ea
-=======
SHA512 (ceph-12.2.8.tar.gz) = 91c5c11aaeef701b3e31c9f8a1192b6b79fd34f3b96c61595c65173c98bfce0009c5b7c9b29764f5e4d4b1f5ac2ca7109fed5ae598c176aba717a34a132fabac
->>>>>>> origin/master
commit 913d33f992f4523e7f00b3e7117e45f80a69a6e9
Merge: 25a7091 2d75781
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Wed Sep 5 07:09:37 2018 -0400
New release (1:12.2.8-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --cc sources
index 7eb8c41,e781e91..47f3ab5
--- a/sources
+++ b/sources
@@@ -1,1 -1,1 +1,5 @@@
++<<<<<<< HEAD
+SHA512 (ceph-12.2.7.tar.gz) = dde75a41c31cca54d4a680b0b93e28680a23f912a348cccca65e55e600c315bbfe3ced008d758fe8ba1e801a9d6e5f8079cc157ec5eef8916b86f487f42064ea
++=======
+ SHA512 (ceph-12.2.8.tar.gz) = 91c5c11aaeef701b3e31c9f8a1192b6b79fd34f3b96c61595c65173c98bfce0009c5b7c9b29764f5e4d4b1f5ac2ca7109fed5ae598c176aba717a34a132fabac
++>>>>>>> origin/master
commit 2d75781cd8490c06b739c1e4b3a9efd6082c26af
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Fri Aug 31 13:38:49 2018 -0400
New release (1:12.2.8-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index ebcf901..8cd60d0 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -85,7 +85,7 @@
# main package definition
#################################################################################
Name: ceph
-Version: 12.2.7
+Version: 12.2.8
Release: 1%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
@@ -171,6 +171,7 @@ BuildRequires: python
BuildRequires: python2-devel
BuildRequires: python-nose
BuildRequires: python-requests
+BuildRequires: python-six
BuildRequires: python-virtualenv
BuildRequires: snappy-devel
BuildRequires: udev
@@ -326,6 +327,7 @@ Summary: Ceph Metadata Server Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
+Requires: python-six
%description mds
ceph-mds is the metadata server daemon for the Ceph distributed file system.
One or more instances of ceph-mds collectively manage the file system
@@ -868,7 +870,6 @@ cmake .. \
-DCMAKE_INSTALL_MANDIR=%{_mandir} \
-DCMAKE_INSTALL_DOCDIR=%{_docdir}/ceph \
-DCMAKE_INSTALL_INCLUDEDIR=%{_includedir} \
- -DWITH_EMBEDDED=OFF \
-DWITH_MANPAGE=ON \
-DWITH_PYTHON3=ON \
-DWITH_SYSTEMD=ON \
@@ -1492,7 +1493,7 @@ if [ $1 -eq 1 ] ; then
/usr/bin/systemctl start ceph-osd.target >/dev/null 2>&1 || :
fi
# work around https://tracker.ceph.com/issues/24903
-chown -h ceph:ceph /var/lib/ceph/osd/*/block* 2>&1 > /dev/null || :
+chown -f -h ceph:ceph /var/lib/ceph/osd/*/block* 2>&1 > /dev/null || :
%preun osd
%if 0%{?suse_version}
@@ -1798,6 +1799,9 @@ exit 0
%changelog
+* Fri Aug 31 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.8-1
+- New release (1:12.2.8-1)
+
* Wed Jul 18 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.7-1
- New release (1:12.2.7-1)
diff --git a/sources b/sources
index 7eb8c41..e781e91 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-12.2.7.tar.gz) = dde75a41c31cca54d4a680b0b93e28680a23f912a348cccca65e55e600c315bbfe3ced008d758fe8ba1e801a9d6e5f8079cc157ec5eef8916b86f487f42064ea
+SHA512 (ceph-12.2.8.tar.gz) = 91c5c11aaeef701b3e31c9f8a1192b6b79fd34f3b96c61595c65173c98bfce0009c5b7c9b29764f5e4d4b1f5ac2ca7109fed5ae598c176aba717a34a132fabac
commit ce9ff30f1cfd96189ebaec8dca6466043a8b1326
Author: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
Date: Mon Jul 30 00:22:27 2018 +0200
%{python_sitearch} %{python2_sitearch}
Signed-off-by: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
diff --git a/ceph.spec b/ceph.spec
index 97367a7..ebcf901 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -1562,8 +1562,8 @@ fi
%{_mandir}/man8/librados-config.8*
%files -n python-rados
-%{python_sitearch}/rados.so
-%{python_sitearch}/rados-*.egg-info
+%{python2_sitearch}/rados.so
+%{python2_sitearch}/rados-*.egg-info
%files -n python%{python3_pkgversion}-rados
%{python3_sitearch}/rados.cpython*.so
@@ -1607,16 +1607,16 @@ fi
%{_libdir}/librgw.so
%files -n python-rgw
-%{python_sitearch}/rgw.so
-%{python_sitearch}/rgw-*.egg-info
+%{python2_sitearch}/rgw.so
+%{python2_sitearch}/rgw-*.egg-info
%files -n python%{python3_pkgversion}-rgw
%{python3_sitearch}/rgw.cpython*.so
%{python3_sitearch}/rgw-*.egg-info
%files -n python-rbd
-%{python_sitearch}/rbd.so
-%{python_sitearch}/rbd-*.egg-info
+%{python2_sitearch}/rbd.so
+%{python2_sitearch}/rbd-*.egg-info
%files -n python%{python3_pkgversion}-rbd
%{python3_sitearch}/rbd.cpython*.so
@@ -1633,8 +1633,8 @@ fi
%{_libdir}/libcephfs.so
%files -n python-cephfs
-%{python_sitearch}/cephfs.so
-%{python_sitearch}/cephfs-*.egg-info
+%{python2_sitearch}/cephfs.so
+%{python2_sitearch}/cephfs-*.egg-info
%{python2_sitelib}/ceph_volume_client.py*
%files -n python%{python3_pkgversion}-cephfs
commit 7a061682c0fa1bdb72bc77ff973948861c9303cb
Author: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
Date: Sun Jul 29 23:39:13 2018 +0200
python-devel python2-devel
Signed-off-by: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
diff --git a/ceph.spec b/ceph.spec
index d62f832..97367a7 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -168,7 +168,7 @@ BuildRequires: parted
BuildRequires: perl
BuildRequires: pkgconfig
BuildRequires: python
-BuildRequires: python-devel
+BuildRequires: python2-devel
BuildRequires: python-nose
BuildRequires: python-requests
BuildRequires: python-virtualenv
commit 2c2513ea411bdf57eaa4ffad4f3e6161b9cf7e8f
Author: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
Date: Sun Jul 29 23:11:50 2018 +0200
%{python_sitelib} %{python2_sitelib}
Signed-off-by: Igor Gnatenko <ignatenkobrain(a)fedoraproject.org>
diff --git a/ceph.spec b/ceph.spec
index 0e5af31..d62f832 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -1018,11 +1018,11 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%endif
%{_unitdir}/ceph-disk@.service
%{_unitdir}/ceph.target
-%{python_sitelib}/ceph_detect_init*
-%{python_sitelib}/ceph_disk*
+%{python2_sitelib}/ceph_detect_init*
+%{python2_sitelib}/ceph_disk*
%dir %{python_sitelib}/ceph_volume
-%{python_sitelib}/ceph_volume/*
-%{python_sitelib}/ceph_volume-*
+%{python2_sitelib}/ceph_volume/*
+%{python2_sitelib}/ceph_volume-*
%{_mandir}/man8/ceph-deploy.8*
%{_mandir}/man8/ceph-detect-init.8*
%{_mandir}/man8/ceph-create-keys.8*
@@ -1139,8 +1139,8 @@ fi
%config %{_sysconfdir}/bash_completion.d/radosgw-admin
%config(noreplace) %{_sysconfdir}/ceph/rbdmap
%{_unitdir}/rbdmap.service
-%{python_sitelib}/ceph_argparse.py*
-%{python_sitelib}/ceph_daemon.py*
+%{python2_sitelib}/ceph_argparse.py*
+%{python2_sitelib}/ceph_daemon.py*
%dir %{_udevrulesdir}
%{_udevrulesdir}/50-rbd.rules
%attr(3770,ceph,ceph) %dir %{_localstatedir}/log/ceph/
@@ -1286,7 +1286,7 @@ fi
%{_bindir}/ceph-monstore-tool
%{_mandir}/man8/ceph-mon.8*
%{_mandir}/man8/ceph-rest-api.8*
-%{python_sitelib}/ceph_rest_api.py*
+%{python2_sitelib}/ceph_rest_api.py*
%{_unitdir}/ceph-mon@.service
%{_unitdir}/ceph-mon.target
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/mon
@@ -1635,7 +1635,7 @@ fi
%files -n python-cephfs
%{python_sitearch}/cephfs.so
%{python_sitearch}/cephfs-*.egg-info
-%{python_sitelib}/ceph_volume_client.py*
+%{python2_sitelib}/ceph_volume_client.py*
%files -n python%{python3_pkgversion}-cephfs
%{python3_sitearch}/cephfs.cpython*.so
commit a4054a83fc48875e02d2620ea8cc76b26dfea20c
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Wed Jul 18 13:09:25 2018 -0400
New release (1:12.2.7-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index c6a9ede..0e5af31 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -1798,7 +1798,7 @@ exit 0
%changelog
-* Wed Jul 28 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.7-1
+* Wed Jul 18 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.7-1
- New release (1:12.2.7-1)
* Thu Jul 12 2018 Fedora Release Engineering <releng(a)fedoraproject.org> - 1:12.2.6-2
commit 1b6e363bd1c1bfaa714c9d0ef43ed595cc2361e9
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Wed Jul 18 09:59:46 2018 -0400
New release (1:12.2.7-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index f5c4121..c6a9ede 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -85,8 +85,8 @@
# main package definition
#################################################################################
Name: ceph
-Version: 12.2.6
-Release: 2%{?dist}
+Version: 12.2.7
+Release: 1%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -1491,6 +1491,8 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl start ceph-osd.target >/dev/null 2>&1 || :
fi
+# work around https://tracker.ceph.com/issues/24903
+chown -h ceph:ceph /var/lib/ceph/osd/*/block* 2>&1 > /dev/null || :
%preun osd
%if 0%{?suse_version}
@@ -1796,6 +1798,9 @@ exit 0
%changelog
+* Wed Jul 28 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.7-1
+- New release (1:12.2.7-1)
+
* Thu Jul 12 2018 Fedora Release Engineering <releng(a)fedoraproject.org> - 1:12.2.6-2
- Rebuilt for https://fedoraproject.org/wiki/Fedora_29_Mass_Rebuild
@@ -1925,10 +1930,10 @@ exit 0
- Use newer -devel package format
- Sync up the spec file
-* Wed Oct 26 2016 Ken Dreyer <ktdreyer(a)ktdreyer.com> 1:10.2.3-4
+* Wed Oct 26 2016 Ken Dreyer <ktdreyer(a)ktdreyer.com> - 1:10.2.3-4
- librgw: add API version defines for librgw and rgw_file
-* Wed Oct 26 2016 Ken Dreyer <ktdreyer(a)ktdreyer.com> 1:10.2.3-3
+* Wed Oct 26 2016 Ken Dreyer <ktdreyer(a)ktdreyer.com> - 1:10.2.3-3
- update patches style for rdopkg
* Thu Sep 29 2016 Boris Ranto <branto(a)redhat.com> - 1:10.2.3-2
diff --git a/sources b/sources
index 2920e75..7eb8c41 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-12.2.6.tar.gz) = 7512623a9bf64b5da1989f62a6c5a3761ae21773bdac185f36a4dcd470f6489724807f28e98e9861ff34f86d2b0ce7af00eac2ddbaa5bba43b49786cb65ad7ff
+SHA512 (ceph-12.2.7.tar.gz) = dde75a41c31cca54d4a680b0b93e28680a23f912a348cccca65e55e600c315bbfe3ced008d758fe8ba1e801a9d6e5f8079cc157ec5eef8916b86f487f42064ea
commit 0fb6d3aa054e1c0996b84b670cc454a841c1460b
Author: Fedora Release Engineering <releng(a)fedoraproject.org>
Date: Thu Jul 12 21:41:00 2018 +0000
- Rebuilt for https://fedoraproject.org/wiki/Fedora_29_Mass_Rebuild
Signed-off-by: Fedora Release Engineering <releng(a)fedoraproject.org>
diff --git a/ceph.spec b/ceph.spec
index bc2111c..f5c4121 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -86,7 +86,7 @@
#################################################################################
Name: ceph
Version: 12.2.6
-Release: 1%{?dist}
+Release: 2%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -1796,6 +1796,9 @@ exit 0
%changelog
+* Thu Jul 12 2018 Fedora Release Engineering <releng(a)fedoraproject.org> - 1:12.2.6-2
+- Rebuilt for https://fedoraproject.org/wiki/Fedora_29_Mass_Rebuild
+
* Wed Jul 11 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.6-1
- New release (1:12.2.6-1)
commit 294576f608fc74be499777b70d41ed4444c5cb4a
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Wed Jul 11 09:01:29 2018 -0400
New release (1:12.2.6-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index e7a968c..bc2111c 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -85,8 +85,8 @@
# main package definition
#################################################################################
Name: ceph
-Version: 12.2.5
-Release: 3%{?dist}
+Version: 12.2.6
+Release: 1%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -190,7 +190,7 @@ BuildRequires: pkgconfig(systemd)
BuildRequires: systemd-rpm-macros
BuildRequires: systemd
%{?systemd_requires}
-PreReq: %fillup_prereq
+# PreReq: %%fillup_prereq
BuildRequires: net-tools
BuildRequires: libbz2-devel
BuildRequires: btrfsprogs
@@ -1796,6 +1796,9 @@ exit 0
%changelog
+* Wed Jul 11 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.6-1
+- New release (1:12.2.6-1)
+
* Mon Jul 2 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-3
- New release (1:12.2.5-3) w/ python-3.7
diff --git a/sources b/sources
index 021832c..2920e75 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-12.2.5.tar.gz) = bd78358590eb077afcc3c1e12864fadee906c95ac8afee25bdca6155f99b75f89d923159fcc13430abba21a89fee3e566177ff5f9ffabdd3b3f220d05602ab86
+SHA512 (ceph-12.2.6.tar.gz) = 7512623a9bf64b5da1989f62a6c5a3761ae21773bdac185f36a4dcd470f6489724807f28e98e9861ff34f86d2b0ce7af00eac2ddbaa5bba43b49786cb65ad7ff
commit 9aeaf53af545215050e285a73f59f7368dceb896
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Mon Jul 2 09:34:50 2018 -0400
New release (1:12.2.5-3) w/ python-3.7
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/ceph.spec b/ceph.spec
index 4e62552..e7a968c 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -86,7 +86,7 @@
#################################################################################
Name: ceph
Version: 12.2.5
-Release: 2%{?dist}
+Release: 3%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -1796,6 +1796,9 @@ exit 0
%changelog
+* Mon Jul 2 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-3
+- New release (1:12.2.5-3) w/ python-3.7
+
* Fri Jun 29 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-2
- New release (1:12.2.5-2)
commit 9b4d3225ea28de8647d7dae8a659f3ddfb793911
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Sun Jul 1 06:35:53 2018 -0400
New release (1:12.2.5-2)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/0001-src-rocksdb-util-murmurhash.patch b/0001-src-rocksdb-util-murmurhash.patch
new file mode 100644
index 0000000..f67a0ff
--- /dev/null
+++ b/0001-src-rocksdb-util-murmurhash.patch
@@ -0,0 +1,13 @@
+--- ceph-12.1.1.orig/src/rocksdb/util/murmurhash.cc 2017-04-27 01:13:46.000000000 +0100
++++ ceph-12.1.1.orig/src/rocksdb/util/murmurhash.cc 2017-07-25 11:37:28.910266684 +0100
+@@ -113,8 +113,8 @@ unsigned int MurmurHash2 ( const void *
+
+ switch(len)
+ {
+- case 3: h ^= data[2] << 16;
+- case 2: h ^= data[1] << 8;
++ case 3: h ^= data[2] << 16; // fallthrough
++ case 2: h ^= data[1] << 8; // fallthrough
+ case 1: h ^= data[0];
+ h *= m;
+ };
diff --git a/0002-cmake-Support-ppc64.patch b/0002-cmake-Support-ppc64.patch
new file mode 100644
index 0000000..6b09549
--- /dev/null
+++ b/0002-cmake-Support-ppc64.patch
@@ -0,0 +1,127 @@
+From 2f0a7153460acc3f21462236f470ec3471fa2ee1 Mon Sep 17 00:00:00 2001
+From: Boris Ranto <branto(a)redhat.com>
+Date: Mon, 31 Jul 2017 19:50:23 +0200
+Subject: [PATCH] cmake: Support ppc64
+
+The ppc64 support requires a couple of changes:
+- adding the ppc64 support to cmake
+- changing optimized crc32 code to compile on ppc64le only
+- moving ifdef condition before crc32_align to avoid defined but not
+used warning
+
+Signed-off-by: Boris Ranto <branto(a)redhat.com>
+---
+ cmake/modules/SIMDExt.cmake | 15 ++++++++++++++-
+ src/CMakeLists.txt | 4 +++-
+ src/arch/ppc.c | 8 ++++----
+ src/common/crc32c_ppc.c | 6 +++---
+ 4 files changed, 24 insertions(+), 9 deletions(-)
+
+diff --git a/cmake/modules/SIMDExt.cmake b/cmake/modules/SIMDExt.cmake
+index 5330835..c47667d 100644
+--- a/cmake/modules/SIMDExt.cmake
++++ b/cmake/modules/SIMDExt.cmake
+@@ -109,7 +109,20 @@ elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "i386|i686|amd64|x86_64|AMD64")
+ endif(CMAKE_SYSTEM_PROCESSOR MATCHES "i686|amd64|x86_64|AMD64")
+ elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "(powerpc|ppc)64le")
+ set(HAVE_PPC64LE 1)
+- message(STATUS " we are ppc64le")
++ message(STATUS " we are ppc64")
++ CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
++ if(HAS_ALTIVEC)
++ message(STATUS " HAS_ALTIVEC yes")
++ set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maltivec")
++ set(CMAKE_CXX_FLAGS "${CMAKE_C_FLAGS} -maltivec")
++ endif()
++ CHECK_C_COMPILER_FLAG("-mcpu=power8" HAVE_POWER8)
++ if(HAVE_POWER8)
++ message(STATUS " HAVE_POWER8 yes")
++ endif()
++elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "(power|ppc)64")
++ set(HAVE_PPC64 1)
++ message(STATUS " we are ppc64")
+ CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
+ if(HAS_ALTIVEC)
+ message(STATUS " HAS_ALTIVEC yes")
+diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
+index 66f0c14..38d1913 100644
+--- a/src/CMakeLists.txt
++++ b/src/CMakeLists.txt
+@@ -568,7 +568,9 @@ if(HAVE_INTEL)
+ endif(HAVE_GOOD_YASM_ELF64)
+ elseif(HAVE_POWER8)
+ list(APPEND libcommon_files
+- common/crc32c_ppc.c
++ common/crc32c_ppc.c)
++elseif(HAVE_PPC64LE)
++ list(APPEND libcommon_files
+ common/crc32c_ppc_asm.S
+ common/crc32c_ppc_fast_zero_asm.S)
+ endif(HAVE_INTEL)
+diff --git a/src/arch/ppc.c b/src/arch/ppc.c
+index f21e2fe..11d3a49 100644
+--- a/src/arch/ppc.c
++++ b/src/arch/ppc.c
+@@ -14,10 +14,10 @@ int ceph_arch_ppc_crc32 = 0;
+
+ #include <stdio.h>
+
+-#if __linux__ && __powerpc64__
++#ifdef HAVE_PPC64LE
+ #include <sys/auxv.h>
+ #include <asm/cputable.h>
+-#endif /* __linux__ && __powerpc64__ */
++#endif /* HAVE_PPC64LE */
+
+ #ifndef PPC_FEATURE2_VEC_CRYPTO
+ #define PPC_FEATURE2_VEC_CRYPTO 0x02000000
+@@ -31,9 +31,9 @@ int ceph_arch_ppc_probe(void)
+ {
+ ceph_arch_ppc_crc32 = 0;
+
+-#if __linux__ && __powerpc64__
++#ifdef HAVE_PPC64LE
+ if (getauxval(AT_HWCAP2) & PPC_FEATURE2_VEC_CRYPTO) ceph_arch_ppc_crc32 = 1;
+-#endif /* __linux__ && __powerpc64__ */
++#endif /* HAVE_PPC64LE */
+
+ return 0;
+ }
+diff --git a/src/common/crc32c_ppc.c b/src/common/crc32c_ppc.c
+index 43756e2..52fd1c4 100644
+--- a/src/common/crc32c_ppc.c
++++ b/src/common/crc32c_ppc.c
+@@ -20,6 +20,7 @@
+ #define VMX_ALIGN 16
+ #define VMX_ALIGN_MASK (VMX_ALIGN-1)
+
++#ifdef HAVE_PPC64LE
+ #ifdef REFLECT
+ static unsigned int crc32_align(unsigned int crc, unsigned char const *p,
+ unsigned long len)
+@@ -38,7 +39,6 @@ static unsigned int crc32_align(unsigned int crc, unsigned char const *p,
+ }
+ #endif
+
+-#ifdef HAVE_POWER8
+ static inline unsigned long polynomial_multiply(unsigned int a, unsigned int b) {
+ vector unsigned int va = {a, 0, 0, 0};
+ vector unsigned int vb = {b, 0, 0, 0};
+@@ -134,7 +134,7 @@ uint32_t ceph_crc32c_ppc(uint32_t crc, unsigned char const *data, unsigned len)
+ return crc;
+ }
+
+-#else /* HAVE_POWER8 */
++#else /* HAVE_PPC64LE */
+
+ /* This symbol has to exist on non-ppc architectures (and on legacy
+ * ppc systems using power7 or below) in order to compile properly
+@@ -145,4 +145,4 @@ uint32_t ceph_crc32c_ppc(uint32_t crc, unsigned char const *data, unsigned len)
+ return 0;
+ }
+
+-#endif /* HAVE_POWER8 */
++#endif /* HAVE_PPC64LE */
+--
+2.9.4
+
diff --git a/0003-librbd-Conditionally-import-TrimRequest.cc.patch b/0003-librbd-Conditionally-import-TrimRequest.cc.patch
new file mode 100644
index 0000000..2f02efb
--- /dev/null
+++ b/0003-librbd-Conditionally-import-TrimRequest.cc.patch
@@ -0,0 +1,31 @@
+From 74a754690736f6608b0d4d9c807df0bd777a129d Mon Sep 17 00:00:00 2001
+From: Boris Ranto <branto(a)redhat.com>
+Date: Fri, 8 Dec 2017 00:21:38 +0100
+Subject: [PATCH] librbd: Conditionally import TrimRequest.cc
+
+We include TrimRequest.cc in librbd tests at two places:
+ - operation/test_mock_TrimRequest.cc
+ - operation/test_mock_ResizeRequest.cc
+
+That causes linking errors when doing the builds because some of the
+structures are defined twice.
+
+Signed-off-by: Boris Ranto <branto(a)redhat.com>
+---
+ src/librbd/operation/TrimRequest.cc | 2 ++
+ 1 file changed, 2 insertions(+)
+
+diff --git a/src/librbd/operation/TrimRequest.cc b/src/librbd/operation/TrimRequest.cc
+index 28f2deb..929ca51 100644
+--- a/src/librbd/operation/TrimRequest.cc
++++ b/src/librbd/operation/TrimRequest.cc
+@@ -362,4 +362,6 @@ void TrimRequest<I>::send_finish(int r) {
+ } // namespace operation
+ } // namespace librbd
+
++#ifndef TEST_F
+ template class librbd::operation::TrimRequest<librbd::ImageCtx>;
++#endif
+--
+2.9.5
+
diff --git a/0005-src-rocksdb-table-block.h.patch b/0005-src-rocksdb-table-block.h.patch
new file mode 100644
index 0000000..8acb401
--- /dev/null
+++ b/0005-src-rocksdb-table-block.h.patch
@@ -0,0 +1,155 @@
+--- ceph-12.2.3/src/rocksdb/table/block.h.orig 2018-02-22 07:49:38.044899631 -0500
++++ ceph-12.2.3/src/rocksdb/table/block.h 2018-02-22 07:58:55.855899631 -0500
+@@ -65,7 +65,8 @@
+
+ // Create bitmap and set all the bits to 0
+ bitmap_ = new std::atomic<uint32_t>[bitmap_size];
+- memset(bitmap_, 0, bitmap_size * kBytesPersEntry);
++ // memset(bitmap_, 0, bitmap_size * kBytesPersEntry);
++ { unsigned i = 0; for (; i < bitmap_size;) bitmap_[i++] = 0; }
+
+ RecordTick(GetStatistics(), READ_AMP_TOTAL_READ_BYTES,
+ num_bits_needed << bytes_per_bit_pow_);
+--- ceph-12.2.3/src/rocksdb/db/c.cc.orig 2018-02-22 08:14:56.033899631 -0500
++++ ceph-12.2.3/src/rocksdb/db/c.cc 2018-02-22 10:06:39.759899631 -0500
+@@ -1322,11 +1322,6 @@
+ b->rep.PutLogData(Slice(blob, len));
+ }
+
+-void rocksdb_writebatch_iterate(
+- rocksdb_writebatch_t* b,
+- void* state,
+- void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
+- void (*deleted)(void*, const char* k, size_t klen)) {
+ class H : public WriteBatch::Handler {
+ public:
+ void* state_;
+@@ -1339,6 +1334,12 @@
+ (*deleted_)(state_, key.data(), key.size());
+ }
+ };
++
++void rocksdb_writebatch_iterate(
++ rocksdb_writebatch_t* b,
++ void* state,
++ void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
++ void (*deleted)(void*, const char* k, size_t klen)) {
+ H handler;
+ handler.state_ = state;
+ handler.put_ = put;
+@@ -1579,18 +1580,6 @@
+ void* state,
+ void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
+ void (*deleted)(void*, const char* k, size_t klen)) {
+- class H : public WriteBatch::Handler {
+- public:
+- void* state_;
+- void (*put_)(void*, const char* k, size_t klen, const char* v, size_t vlen);
+- void (*deleted_)(void*, const char* k, size_t klen);
+- virtual void Put(const Slice& key, const Slice& value) override {
+- (*put_)(state_, key.data(), key.size(), value.data(), value.size());
+- }
+- virtual void Delete(const Slice& key) override {
+- (*deleted_)(state_, key.data(), key.size());
+- }
+- };
+ H handler;
+ handler.state_ = state;
+ handler.put_ = put;
+@@ -2532,13 +2521,9 @@
+ delete filter;
+ }
+
+-rocksdb_filterpolicy_t* rocksdb_filterpolicy_create_bloom_format(int bits_per_key, bool original_format) {
+- // Make a rocksdb_filterpolicy_t, but override all of its methods so
+- // they delegate to a NewBloomFilterPolicy() instead of user
+- // supplied C functions.
+- struct Wrapper : public rocksdb_filterpolicy_t {
++ struct WrapperFP : public rocksdb_filterpolicy_t {
+ const FilterPolicy* rep_;
+- ~Wrapper() { delete rep_; }
++ ~WrapperFP() { delete rep_; }
+ const char* Name() const override { return rep_->Name(); }
+ void CreateFilter(const Slice* keys, int n,
+ std::string* dst) const override {
+@@ -2549,11 +2534,16 @@
+ }
+ static void DoNothing(void*) { }
+ };
+- Wrapper* wrapper = new Wrapper;
++
++rocksdb_filterpolicy_t* rocksdb_filterpolicy_create_bloom_format(int bits_per_key, bool original_format) {
++ // Make a rocksdb_filterpolicy_t, but override all of its methods so
++ // they delegate to a NewBloomFilterPolicy() instead of user
++ // supplied C functions.
++ WrapperFP* wrapper = new WrapperFP;
+ wrapper->rep_ = NewBloomFilterPolicy(bits_per_key, original_format);
+ wrapper->state_ = nullptr;
+ wrapper->delete_filter_ = nullptr;
+- wrapper->destructor_ = &Wrapper::DoNothing;
++ wrapper->destructor_ = &WrapperFP::DoNothing;
+ return wrapper;
+ }
+
+@@ -2889,10 +2879,9 @@
+ delete st;
+ }
+
+-rocksdb_slicetransform_t* rocksdb_slicetransform_create_fixed_prefix(size_t prefixLen) {
+- struct Wrapper : public rocksdb_slicetransform_t {
++ struct WrapperST : public rocksdb_slicetransform_t {
+ const SliceTransform* rep_;
+- ~Wrapper() { delete rep_; }
++ ~WrapperST() { delete rep_; }
+ const char* Name() const override { return rep_->Name(); }
+ Slice Transform(const Slice& src) const override {
+ return rep_->Transform(src);
+@@ -2903,31 +2892,20 @@
+ bool InRange(const Slice& src) const override { return rep_->InRange(src); }
+ static void DoNothing(void*) { }
+ };
+- Wrapper* wrapper = new Wrapper;
++
++rocksdb_slicetransform_t* rocksdb_slicetransform_create_fixed_prefix(size_t prefixLen) {
++ WrapperST* wrapper = new WrapperST;
+ wrapper->rep_ = rocksdb::NewFixedPrefixTransform(prefixLen);
+ wrapper->state_ = nullptr;
+- wrapper->destructor_ = &Wrapper::DoNothing;
++ wrapper->destructor_ = &WrapperST::DoNothing;
+ return wrapper;
+ }
+
+ rocksdb_slicetransform_t* rocksdb_slicetransform_create_noop() {
+- struct Wrapper : public rocksdb_slicetransform_t {
+- const SliceTransform* rep_;
+- ~Wrapper() { delete rep_; }
+- const char* Name() const override { return rep_->Name(); }
+- Slice Transform(const Slice& src) const override {
+- return rep_->Transform(src);
+- }
+- bool InDomain(const Slice& src) const override {
+- return rep_->InDomain(src);
+- }
+- bool InRange(const Slice& src) const override { return rep_->InRange(src); }
+- static void DoNothing(void*) { }
+- };
+- Wrapper* wrapper = new Wrapper;
++ WrapperST* wrapper = new WrapperST;
+ wrapper->rep_ = rocksdb::NewNoopTransform();
+ wrapper->state_ = nullptr;
+- wrapper->destructor_ = &Wrapper::DoNothing;
++ wrapper->destructor_ = &WrapperST::DoNothing;
+ return wrapper;
+ }
+
+--- ceph-12.2.3/src/rocksdb/memtable/inlineskiplist.h.orig 2018-02-22 10:34:06.918899631 -0500
++++ ceph-12.2.3/src/rocksdb/memtable/inlineskiplist.h 2018-02-22 10:34:44.145899631 -0500
+@@ -279,7 +279,7 @@
+ // next_[0]. This is used for passing data from AllocateKey to Insert.
+ void StashHeight(const int height) {
+ assert(sizeof(int) <= sizeof(next_[0]));
+- memcpy(&next_[0], &height, sizeof(int));
++ memcpy(static_cast<void*>(&next_[0]), &height, sizeof(int));
+ }
+
+ // Retrieves the value passed to StashHeight. Undefined after a call
diff --git a/ceph.spec b/ceph.spec
index 96ba41f..4e62552 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -16,59 +16,58 @@
#
%global _hardened_build 1
-%bcond_with python3
%bcond_without ocf
+%ifnarch armv7hl
+%bcond_without cephfs_java
+%else
+%bcond_with cephfs_java
+%endif
+%if 0%{?suse_version}
+%bcond_with ceph_test_package
+%else
+%bcond_without ceph_test_package
+%endif
%bcond_with make_check
%ifarch s390 s390x
%bcond_with tcmalloc
%else
%bcond_without tcmalloc
%endif
+%ifnarch armv7hl
+%bcond_with lowmem_builder
+%else
+%if 0%{?rhel}
+%ifnarch ppc64le
+%bcond_with lowmem_builder
+%else
+%bcond_without lowmem_builder
+%endif
+%else
+%bcond_without lowmem_builder
+%endif
+%endif
%if 0%{?fedora} || 0%{?rhel}
%bcond_without selinux
-%bcond_without ceph_test_package
-%bcond_without cephfs_java
-%bcond_without lttng
-%global _remote_tarball_prefix https://download.ceph.com/tarballs/
%endif
%if 0%{?suse_version}
%bcond_with selinux
-%bcond_with ceph_test_package
-%bcond_with cephfs_java
-#Compat macro for new _fillupdir macro introduced in Nov 2017
-%if ! %{defined _fillupdir}
-%global _fillupdir /var/adm/fillup-templates
%endif
-%if 0%{?is_opensuse}
-%bcond_without lttng
-%else
-%ifarch x86_64 aarch64
+
+# LTTng-UST enabled on Fedora, RHEL 6+, and SLE (not openSUSE)
+%if 0%{?fedora} || 0%{?rhel} >= 6 || 0%{?suse_version}
+%if ! 0%{?is_opensuse}
%bcond_without lttng
-%else
-%bcond_with lttng
%endif
%endif
-%endif
-%if 0%{?suse_version} >= 1500
-%bcond_with python2
-%else
-%bcond_without python2
-%endif
-%if 0%{without python2}
-%global _defined_if_python2_absent 1
-%endif
%if %{with selinux}
# get selinux policy version
-%{!?_selinux_policy_version: %global _selinux_policy_version 0.0.0}
+%{!?_selinux_policy_version: %global _selinux_policy_version %(sed -e 's,.*selinux-policy-\\([^/]*\\)/.*,\\1,' /usr/share/selinux/devel/policyhelp 2>/dev/null || echo 0.0.0)}
%endif
%{!?_udevrulesdir: %global _udevrulesdir /lib/udev/rules.d}
%{!?tmpfiles_create: %global tmpfiles_create systemd-tmpfiles --create}
%{!?python3_pkgversion: %global python3_pkgversion 3}
-# define _python_buildid macro which will expand to the empty string when
-# building with python2
-%global _python_buildid %{?_defined_if_python2_absent:%{python3_pkgversion}}
# unify libexec for all targets
%global _libexecdir %{_exec_prefix}/lib
@@ -77,23 +76,22 @@
%global _find_debuginfo_dwz_opts %{nil}
%if ( 0%{?rhel} && 0%{?rhel} < 7 )
-%global _rundir %{_localstatedir}/run/
+%global _rundir %{_localstatedir}/run
%else
-%global _rundir /run/
+%global _rundir /run
%endif
#################################################################################
# main package definition
#################################################################################
Name: ceph
-Version: 13.2.0
-Release: 3%{?dist}
+Version: 12.2.5
+Release: 2%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
-# define _epoch_prefix macro which will expand to the empty string if epoch is
-# undefined
+# define %%_epoch_prefix macro which will expand to the empty string if %%epoch is undefined
%global _epoch_prefix %{?epoch:%{epoch}:}
Summary: User space components of the Ceph file system
@@ -102,11 +100,14 @@ License: LGPL-2.1 and CC-BY-SA-3.0 and GPL-2.0 and BSL-1.0 and BSD-3-Clause and
Group: System/Filesystems
%endif
URL: http://ceph.com/
-Source0: %{?_remote_tarball_prefix}%{name}-%{version}.tar.gz
+Source0: http://download.ceph.com/tarballs/%{name}-%{version}.tar.gz
# https://bugzilla.redhat.com/show_bug.cgi?id=1474773
-Patch001: 0001-blobstore.patch
+Patch001: 0001-src-rocksdb-util-murmurhash.patch
+# https://bugzilla.redhat.com/show_bug.cgi?id=1474774
+Patch002: 0002-cmake-Support-ppc64.patch
+Patch003: 0003-librbd-Conditionally-import-TrimRequest.cc.patch
+Patch005: 0005-src-rocksdb-table-block.h.patch
%if 0%{?suse_version}
-# _insert_obs_source_lines_here
%if 0%{?is_opensuse}
ExclusiveArch: x86_64 aarch64 ppc64 ppc64le
%else
@@ -116,10 +117,10 @@ ExclusiveArch: x86_64 aarch64 ppc64le s390x
#################################################################################
# dependencies that apply across all distro families
#################################################################################
-Requires: ceph-osd = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mds = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mgr = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mon = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-osd = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mds = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mgr = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mon = %{_epoch_prefix}%{version}-%{release}
Requires(post): binutils
%if 0%{with cephfs_java}
BuildRequires: java-devel
@@ -128,48 +129,50 @@ BuildRequires: sharutils
%if 0%{with selinux}
BuildRequires: checkpolicy
BuildRequires: selinux-policy-devel
+BuildRequires: selinux-policy-doc
+%endif
+%if 0%{with make_check}
+%if 0%{?fedora} || 0%{?rhel}
+BuildRequires: python-cherrypy
+BuildRequires: python-werkzeug
+%endif
+%if 0%{?suse_version}
+BuildRequires: python-CherryPy
+BuildRequires: python-Werkzeug
+BuildRequires: python-numpy-devel
+%endif
+BuildRequires: python-coverage
+BuildRequires: python-pecan
+BuildRequires: socat
%endif
BuildRequires: bc
BuildRequires: gperf
-BuildRequires: cmake
+BuildRequires: cmake
BuildRequires: cryptsetup
BuildRequires: fuse-devel
-BuildRequires: cryptopp-devel
-%if 0%{?rhel} == 7
-# devtoolset offers newer make and valgrind-devel, but the old ones are good
-# enough.
-BuildRequires: devtoolset-7-gcc-c++
-%else
BuildRequires: gcc-c++
-%endif
BuildRequires: gdbm
%if 0%{with tcmalloc}
BuildRequires: gperftools-devel >= 2.4
%endif
-BuildRequires: jq
+BuildRequires: jq
BuildRequires: leveldb-devel > 1.2
BuildRequires: libaio-devel
BuildRequires: libblkid-devel >= 2.17
BuildRequires: libcurl-devel
BuildRequires: libudev-devel
-BuildRequires: liboath-devel
BuildRequires: libtool
BuildRequires: libxml2-devel
-BuildRequires: libuuid-devel
BuildRequires: make
BuildRequires: parted
BuildRequires: perl
BuildRequires: pkgconfig
-BuildRequires: procps
-BuildRequires: python%{_python_buildid}
-BuildRequires: python%{_python_buildid}-devel
-BuildRequires: python%{_python_buildid}-nose
-BuildRequires: python%{_python_buildid}-requests
-BuildRequires: python%{_python_buildid}-virtualenv
+BuildRequires: python
+BuildRequires: python-devel
+BuildRequires: python-nose
+BuildRequires: python-requests
+BuildRequires: python-virtualenv
BuildRequires: snappy-devel
-%if 0%{with make_check}
-BuildRequires: socat
-%endif
BuildRequires: udev
BuildRequires: util-linux
BuildRequires: valgrind-devel
@@ -183,49 +186,43 @@ BuildRequires: yasm
# distro-conditional dependencies
#################################################################################
%if 0%{?suse_version}
-BuildRequires: pkgconfig(systemd)
+BuildRequires: pkgconfig(systemd)
BuildRequires: systemd-rpm-macros
BuildRequires: systemd
%{?systemd_requires}
PreReq: %fillup_prereq
BuildRequires: net-tools
BuildRequires: libbz2-devel
-BuildRequires: btrfsprogs
+BuildRequires: btrfsprogs
BuildRequires: mozilla-nss-devel
BuildRequires: keyutils-devel
-BuildRequires: libopenssl-devel
-BuildRequires: lsb-release
-BuildRequires: openldap2-devel
-BuildRequires: cunit-devel
-BuildRequires: python%{_python_buildid}-base
-BuildRequires: python%{_python_buildid}-Cython
-BuildRequires: python%{_python_buildid}-PrettyTable
-BuildRequires: python%{_python_buildid}-Sphinx
-BuildRequires: rdma-core-devel
-BuildRequires: liblz4-devel >= 1.7
+BuildRequires: libopenssl-devel
+BuildRequires: lsb-release
+BuildRequires: openldap2-devel
+BuildRequires: python-Cython
+BuildRequires: python-PrettyTable
+BuildRequires: python-Sphinx
+BuildRequires: rdma-core-devel
%endif
%if 0%{?fedora} || 0%{?rhel}
Requires: systemd
-BuildRequires: boost-random
+BuildRequires: boost-random
BuildRequires: btrfs-progs
BuildRequires: nss-devel
BuildRequires: keyutils-libs-devel
-# RDMA is no longer built on 32-bit ARM: see rhbz#1484155
+# RDMA is no longer built on 32-bit ARM: see #1484155
%ifnarch %{arm}
BuildRequires: rdma-core-devel
%endif
-BuildRequires: openldap-devel
-BuildRequires: openssl-devel
-BuildRequires: redhat-lsb-core
-BuildRequires: CUnit-devel
-BuildRequires: Cython%{_python_buildid}
-BuildRequires: python%{_python_buildid}-prettytable
-BuildRequires: python%{_python_buildid}-sphinx
-BuildRequires: lz4-devel >= 1.7
+BuildRequires: openldap-devel
+BuildRequires: openssl-devel
+BuildRequires: redhat-lsb-core
+BuildRequires: Cython
+BuildRequires: python-prettytable
+BuildRequires: python-sphinx
%endif
# python34-... for RHEL, python3-... for all other supported distros
-%if %{with python3}
-%if 0%{?rhel}
+%if ( 0%{?rhel} && 0%{?rhel} <= 7 )
BuildRequires: python34-devel
BuildRequires: python34-setuptools
BuildRequires: python34-Cython
@@ -234,34 +231,6 @@ BuildRequires: python3-devel
BuildRequires: python3-setuptools
BuildRequires: python3-Cython
%endif
-%endif
-# distro-conditional make check dependencies
-%if 0%{with make_check}
-%if 0%{?fedora} || 0%{?rhel}
-BuildRequires: python%{_python_buildid}-cherrypy
-BuildRequires: python%{_python_buildid}-routes
-BuildRequires: python%{_python_buildid}-pecan
-BuildRequires: python%{_python_buildid}-werkzeug
-BuildRequires: python%{_python_buildid}-tox
-BuildRequires: python%{_python_buildid}-coverage
-%if 0%{?fedora}
-BuildRequires: python%{_python_buildid}-bcrypt
-%endif
-%if 0%{?rhel}
-BuildRequires: py-bcrypt
-%endif
-%endif
-%if 0%{?suse_version}
-BuildRequires: python%{_python_buildid}-CherryPy
-BuildRequires: python%{_python_buildid}-Routes
-BuildRequires: python%{_python_buildid}-Werkzeug
-BuildRequires: python%{_python_buildid}-pecan
-BuildRequires: python%{_python_buildid}-numpy-devel
-BuildRequires: python%{_python_buildid}-bcrypt
-BuildRequires: python%{_python_buildid}-tox
-BuildRequires: python%{_python_buildid}-coverage
-%endif
-%endif
# lttng and babeltrace for rbd-replay-prep
%if %{with lttng}
%if 0%{?fedora} || 0%{?rhel}
@@ -270,7 +239,7 @@ BuildRequires: libbabeltrace-devel
%endif
%if 0%{?suse_version}
BuildRequires: lttng-ust-devel
-BuildRequires: babeltrace-devel
+BuildRequires: babeltrace-devel
%endif
%endif
%if 0%{?suse_version}
@@ -281,7 +250,7 @@ BuildRequires: expat-devel
%endif
#hardened-cc1
%if 0%{?fedora} || 0%{?rhel}
-BuildRequires: redhat-rpm-config
+BuildRequires: redhat-rpm-config
%endif
%description
@@ -293,39 +262,33 @@ on commodity hardware and delivers object, block and file system storage.
# subpackages
#################################################################################
%package base
-Summary: Ceph Base Package
+Summary: Ceph Base Package
%if 0%{?suse_version}
-Group: System/Filesystems
+Group: System/Filesystems
%endif
-Requires: ceph-common = %{_epoch_prefix}%{version}-%{release}
-Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
-Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
-Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
-Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-common = %{_epoch_prefix}%{version}-%{release}
+Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
+Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
+Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
+Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
%if 0%{with selinux}
-Requires: ceph-selinux = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-selinux = %{_epoch_prefix}%{version}-%{release}
%endif
Requires(post):/sbin/ldconfig
Requires(postun):/sbin/ldconfig
-Requires: cryptsetup
-Requires: e2fsprogs
-Requires: findutils
-Requires: grep
-Requires: logrotate
-Requires: parted
-Requires: psmisc
-Requires: python%{_python_buildid}-requests
-Requires: python%{_python_buildid}-setuptools
-Requires: util-linux
-Requires: xfsprogs
-Requires: which
-%if 0%{?fedora} || 0%{?rhel}
-Requires: gdisk
-%endif
-%if 0%{?suse_version}
-Recommends: ntp-daemon
-Recommends: chrony
-Requires: gptfdisk
+Requires: python
+Requires: python-requests
+Requires: python-setuptools
+Requires: grep
+Requires: xfsprogs
+Requires: logrotate
+Requires: util-linux
+Requires: cryptsetup
+Requires: findutils
+Requires: psmisc
+Requires: which
+%if 0%{?suse_version}
+Recommends: ntp-daemon
%endif
%description base
Base is the package that includes all the files shared amongst ceph servers
@@ -338,18 +301,17 @@ Group: System/Filesystems
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
-Requires: python%{_python_buildid}-rados = %{_epoch_prefix}%{version}-%{release}
-Requires: python%{_python_buildid}-rbd = %{_epoch_prefix}%{version}-%{release}
-Requires: python%{_python_buildid}-cephfs = %{_epoch_prefix}%{version}-%{release}
-Requires: python%{_python_buildid}-rgw = %{_epoch_prefix}%{version}-%{release}
+Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
+Requires: python-rbd = %{_epoch_prefix}%{version}-%{release}
+Requires: python-cephfs = %{_epoch_prefix}%{version}-%{release}
+Requires: python-rgw = %{_epoch_prefix}%{version}-%{release}
%if 0%{?fedora} || 0%{?rhel}
-Requires: python%{_python_buildid}-prettytable
-Requires: python%{_python_buildid}-requests
+Requires: python-prettytable
%endif
%if 0%{?suse_version}
-Requires: python%{_python_buildid}-PrettyTable
-Requires: python%{_python_buildid}-requests
+Requires: python-PrettyTable
%endif
+Requires: python-requests
%{?systemd_requires}
%if 0%{?suse_version}
Requires(pre): pwdutils
@@ -375,6 +337,13 @@ Summary: Ceph Monitor Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
+# For ceph-rest-api
+%if 0%{?fedora} || 0%{?rhel}
+Requires: python-flask
+%endif
+%if 0%{?suse_version}
+Requires: python-Flask
+%endif
%description mon
ceph-mon is the cluster monitor daemon for the Ceph distributed file
system. One or more instances of ceph-mon form a Paxos part-time
@@ -382,35 +351,24 @@ parliament cluster that provides extremely reliable and durable storage
of cluster membership, configuration, and state.
%package mgr
-Summary: Ceph Manager Daemon
+Summary: Ceph Manager Daemon
%if 0%{?suse_version}
-Group: System/Filesystems
+Group: System/Filesystems
%endif
-Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
%if 0%{?fedora} || 0%{?rhel}
-Requires: python%{_python_buildid}-cherrypy
-Requires: python%{_python_buildid}-routes
-Requires: python%{_python_buildid}-jinja2
-Requires: python%{_python_buildid}-pecan
-Requires: python%{_python_buildid}-werkzeug
-Requires: pyOpenSSL%{_python_buildid}
-%if 0%{?fedora}
-Requires: python%{_python_buildid}-bcrypt
-%endif
-%if 0%{?rhel}
-Requires: py-bcrypt
-%endif
+Requires: python-cherrypy
+Requires: python-jinja2
+Requires: python-werkzeug
+Requires: pyOpenSSL
%endif
%if 0%{?suse_version}
-Requires: python%{_python_buildid}-CherryPy
-Requires: python%{_python_buildid}-Routes
-Requires: python%{_python_buildid}-Jinja2
-Requires: python%{_python_buildid}-Werkzeug
-Requires: python%{_python_buildid}-pecan
-Requires: python%{_python_buildid}-pyOpenSSL
-Requires: python%{_python_buildid}-bcrypt
-Recommends: python%{_python_buildid}-influxdb
+Requires: python-CherryPy
+Requires: python-jinja2
+Requires: python-Werkzeug
+Requires: python-pyOpenSSL
%endif
+Requires: python-pecan
%description mgr
ceph-mgr enables python modules that provide services (such as the REST
module derived from Calamari) and expose CLI hooks. ceph-mgr gathers
@@ -497,7 +455,15 @@ Summary: Ceph Object Storage Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
-Requires: lvm2
+# for sgdisk, used by ceph-disk
+%if 0%{?fedora} || 0%{?rhel}
+Requires: gdisk
+%endif
+%if 0%{?suse_version}
+Requires: gptfdisk
+%endif
+Requires: parted
+Requires: lvm2
%description osd
ceph-osd is the object storage daemon for the Ceph distributed file
system. It is responsible for storing objects on a local file system
@@ -554,7 +520,6 @@ Obsoletes: librgw2-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use RADOS gateway client library.
-%if 0%{with python2}
%package -n python-rgw
Summary: Python 2 libraries for the RADOS gateway
%if 0%{?suse_version}
@@ -566,22 +531,18 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rgw
This package contains Python 2 libraries for interacting with Cephs RADOS
gateway.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-rgw
Summary: Python 3 libraries for the RADOS gateway
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-rgw
This package contains Python 3 libraries for interacting with Cephs RADOS
gateway.
-%endif
-%if 0%{with python2}
%package -n python-rados
Summary: Python 2 libraries for the RADOS object store
%if 0%{?suse_version}
@@ -592,20 +553,17 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rados
This package contains Python 2 libraries for interacting with Cephs RADOS
object store.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-rados
Summary: Python 3 libraries for the RADOS object store
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: python%{python3_pkgversion}
Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-rados
This package contains Python 3 libraries for interacting with Cephs RADOS
object store.
-%endif
%package -n libradosstriper1
Summary: RADOS striping interface
@@ -665,11 +623,10 @@ Obsoletes: librbd1-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use RADOS block device.
-%if 0%{with python2}
%package -n python-rbd
Summary: Python 2 libraries for the RADOS block device
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
@@ -677,20 +634,17 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rbd
This package contains Python 2 libraries for interacting with Cephs RADOS
block device.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-rbd
Summary: Python 3 libraries for the RADOS block device
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-rbd
This package contains Python 3 libraries for interacting with Cephs RADOS
block device.
-%endif
%package -n libcephfs2
Summary: Ceph distributed file system client library
@@ -723,47 +677,41 @@ Obsoletes: libcephfs2-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use Cephs distributed file system.
-%if 0%{with python2}
%package -n python-cephfs
Summary: Python 2 libraries for Ceph distributed file system
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
%if 0%{?suse_version}
-Recommends: python-rados = %{_epoch_prefix}%{version}-%{release}
+Recommends: python-rados = %{_epoch_prefix}%{version}-%{release}
%endif
Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-cephfs
This package contains Python 2 libraries for interacting with Cephs distributed
file system.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-cephfs
Summary: Python 3 libraries for Ceph distributed file system
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-cephfs
This package contains Python 3 libraries for interacting with Cephs distributed
file system.
-%endif
-%if 0%{with python3}
%package -n python%{python3_pkgversion}-ceph-argparse
Summary: Python 3 utility libraries for Ceph CLI
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
%description -n python%{python3_pkgversion}-ceph-argparse
This package contains types and routines for Python 3 used by the Ceph CLI as
well as the RESTful interface. These have to do with querying the daemons for
command-description information, validating user command input against those
descriptions, and submitting the command to the appropriate daemon.
-%endif
%if 0%{with ceph_test_package}
%package -n ceph-test
@@ -813,18 +761,17 @@ Group: System/Libraries
%endif
Requires: java
Requires: libcephfs_jni1 = %{_epoch_prefix}%{version}-%{release}
-BuildRequires: junit
+Requires: junit
+BuildRequires: junit
%description -n cephfs-java
This package contains the Java libraries for the Ceph File System.
%endif
%package -n rados-objclass-devel
-Summary: RADOS object class development kit
-%if 0%{?suse_version}
-Group: Development/Libraries/C and C++
-%endif
-Requires: librados2-devel = %{_epoch_prefix}%{version}-%{release}
+Summary: RADOS object class development kit
+Group: Development/Libraries
+Requires: librados2-devel = %{_epoch_prefix}%{version}-%{release}
%description -n rados-objclass-devel
This package contains libraries and headers needed to develop RADOS object
class plugins.
@@ -848,11 +795,10 @@ populated file-systems.
%endif
-%if 0%{with python2}
%package -n python-ceph-compat
Summary: Compatibility package for Cephs python libraries
%if 0%{?suse_version}
-Group: Development/Libraries/Python
+Group: Development/Languages/Python
%endif
Obsoletes: python-ceph
Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
@@ -865,7 +811,6 @@ This is a compatibility package to accommodate python-ceph split into
python-rados, python-rbd, python-rgw and python-cephfs. Packages still
depending on python-ceph should be fixed to depend on python-rados,
python-rbd, python-rgw or python-cephfs instead.
-%endif
#################################################################################
# common
@@ -874,11 +819,6 @@ python-rbd, python-rgw or python-cephfs instead.
%autosetup -p1 -n %{name}-%{version}
%build
-
-%if 0%{?rhel} == 7
-. /opt/rh/devtoolset-7/enable
-%endif
-
%if 0%{with cephfs_java}
# Find jni.h
for i in /usr/{lib64,lib}/jvm/java/include{,/linux}; do
@@ -886,40 +826,37 @@ for i in /usr/{lib64,lib}/jvm/java/include{,/linux}; do
done
%endif
-%if 0%{?suse_version}
-# the following setting fixed an OOM condition we once encountered in the OBS
+%if %{with lowmem_builder}
RPM_OPT_FLAGS="$RPM_OPT_FLAGS --param ggc-min-expand=20 --param ggc-min-heapsize=32768"
%endif
+%ifnarch armv7hl
+export RPM_OPT_FLAGS=`echo $RPM_OPT_FLAGS | sed -e 's/i386/i486/'`
+%else
+export RPM_OPT_FLAGS=`echo $RPM_OPT_FLAGS | sed -e 's/i386/i486/' -e 's/-pipe//g'`
+%endif
export CPPFLAGS="$java_inc"
export CFLAGS="$RPM_OPT_FLAGS"
export CXXFLAGS="$RPM_OPT_FLAGS"
export LDFLAGS="$RPM_LD_FLAGS"
-# Parallel build settings ...
-CEPH_MFLAGS_JOBS="%{?_smp_mflags}"
-CEPH_SMP_NCPUS=$(echo "$CEPH_MFLAGS_JOBS" | sed 's/-j//')
-%if 0%{?__isa_bits} == 32
-# 32-bit builds can use 3G memory max, which is not enough even for -j2
-CEPH_SMP_NCPUS="1"
-%endif
-# do not eat all memory
-echo "Available memory:"
-free -h
-echo "System limits:"
-ulimit -a
-if test -n "$CEPH_SMP_NCPUS" -a "$CEPH_SMP_NCPUS" -gt 1 ; then
- mem_per_process=1800
- max_mem=$(LANG=C free -m | sed -n "s|^Mem: *\([0-9]*\).*$|\1|p")
- max_jobs="$(($max_mem / $mem_per_process))"
- test "$CEPH_SMP_NCPUS" -gt "$max_jobs" && CEPH_SMP_NCPUS="$max_jobs" && echo "Warning: Reducing build parallelism to -j$max_jobs because of memory limits"
- test "$CEPH_SMP_NCPUS" -le 0 && CEPH_SMP_NCPUS="1" && echo "Warning: Not using parallel build at all because of memory limits"
-fi
-export CEPH_SMP_NCPUS
-export CEPH_MFLAGS_JOBS="-j$CEPH_SMP_NCPUS"
-
env | sort
+%if %{with lowmem_builder}
+%ifnarch armv7hl
+%if 0%{?jobs} > 8
+%define _smp_mflags -j8
+%endif
+%else
+%define _smp_mflags -j1
+%endif
+%endif
+
+# unlimit _smp_mflags in system macro if not set above
+%define _smp_ncpus_max 0
+# extract the number of processors for use with cmake
+%define _smp_ncpus %(echo %{_smp_mflags} | sed 's/-j//')
+
mkdir build
cd build
cmake .. \
@@ -933,19 +870,9 @@ cmake .. \
-DCMAKE_INSTALL_INCLUDEDIR=%{_includedir} \
-DWITH_EMBEDDED=OFF \
-DWITH_MANPAGE=ON \
-%if %{with python3}
-DWITH_PYTHON3=ON \
-%else
- -DWITH_PYTHON3=OFF \
-%endif
- -DWITH_MGR_DASHBOARD_FRONTEND=OFF \
-%if %{with python2}
- -DWITH_PYTHON2=ON \
-%else
- -DWITH_PYTHON2=OFF \
- -DMGR_PYTHON_VERSION=3 \
-%endif
-%if ( ( 0%{?rhel} && 0%{?rhel} < 8) && ! 0%{?centos} )
+ -DWITH_SYSTEMD=ON \
+%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
-DWITH_SUBMAN=ON \
%endif
%if 0%{without ceph_test_package}
@@ -973,23 +900,24 @@ cmake .. \
%else
-DWITH_BOOST_CONTEXT=OFF \
%endif
-%ifarch %{arm}
+%ifnarch %{arm}
-DWITH_RDMA=OFF \
%endif
- -DWITH_DPDK=ON \
- -DBOOST_J=$CEPH_SMP_NCPUS
+ -DBOOST_J=%{_smp_ncpus}
-make "$CEPH_MFLAGS_JOBS"
+make %{?_smp_mflags}
%if 0%{with make_check}
%check
# run in-tree unittests
cd build
-ctest "$CEPH_MFLAGS_JOBS"
+ctest %{?_smp_mflags}
+
%endif
+
%install
pushd build
make DESTDIR=%{buildroot} install
@@ -1001,9 +929,10 @@ install -m 0644 -D src/etc-rbdmap %{buildroot}%{_sysconfdir}/ceph/rbdmap
install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_sysconfdir}/sysconfig/ceph
%endif
%if 0%{?suse_version}
-install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_fillupdir}/sysconfig.%{name}
+install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_localstatedir}/adm/fillup-templates/sysconfig.%{name}
%endif
install -m 0644 -D systemd/ceph.tmpfiles.d %{buildroot}%{_tmpfilesdir}/ceph-common.conf
+install -m 0755 -D systemd/ceph %{buildroot}%{_sbindir}/rcceph
install -m 0644 -D systemd/50-ceph.preset %{buildroot}%{_libexecdir}/systemd/system-preset/50-ceph.preset
mkdir -p %{buildroot}%{_sbindir}
install -m 0644 -D src/logrotate.conf %{buildroot}%{_sysconfdir}/logrotate.d/ceph
@@ -1026,7 +955,7 @@ install -m 0644 -D udev/95-ceph-osd.rules %{buildroot}%{_udevrulesdir}/95-ceph-o
#set up placeholder directories
mkdir -p %{buildroot}%{_sysconfdir}/ceph
-mkdir -p %{buildroot}%{_rundir}ceph
+mkdir -p %{buildroot}%{_rundir}/ceph
mkdir -p %{buildroot}%{_localstatedir}/log/ceph
mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/tmp
mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/mon
@@ -1045,9 +974,6 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%py3_compile %{buildroot}%{python3_sitelib}
%endif
-%clean
-rm -rf %{buildroot}
-
#################################################################################
# files and systemd scriptlets
#################################################################################
@@ -1063,6 +989,7 @@ rm -rf %{buildroot}
%{_libexecdir}/systemd/system-preset/50-ceph.preset
%{_sbindir}/ceph-create-keys
%{_sbindir}/ceph-disk
+%{_sbindir}/rcceph
%dir %{_libexecdir}/ceph
%{_libexecdir}/ceph/ceph_common.sh
%dir %{_libdir}/rados-classes
@@ -1085,30 +1012,17 @@ rm -rf %{buildroot}
%config(noreplace) %{_sysconfdir}/sysconfig/ceph
%endif
%if 0%{?suse_version}
-%{_fillupdir}/sysconfig.*
+%{_localstatedir}/adm/fillup-templates/sysconfig.*
%config %{_sysconfdir}/sysconfig/SuSEfirewall2.d/services/ceph-mon
%config %{_sysconfdir}/sysconfig/SuSEfirewall2.d/services/ceph-osd-mds
%endif
%{_unitdir}/ceph-disk@.service
%{_unitdir}/ceph.target
-%if 0%{with python2}
%{python_sitelib}/ceph_detect_init*
%{python_sitelib}/ceph_disk*
-%endif
-%if 0%{with python3}
-%{python3_sitelib}/ceph_detect_init*
-%{python3_sitelib}/ceph_disk*
-%endif
-%if 0%{with python2}
%dir %{python_sitelib}/ceph_volume
%{python_sitelib}/ceph_volume/*
%{python_sitelib}/ceph_volume-*
-%endif
-%if 0%{with python3}
-%dir %{python3_sitelib}/ceph_volume
-%{python3_sitelib}/ceph_volume/*
-%{python3_sitelib}/ceph_volume-*
-%endif
%{_mandir}/man8/ceph-deploy.8*
%{_mandir}/man8/ceph-detect-init.8*
%{_mandir}/man8/ceph-create-keys.8*
@@ -1127,7 +1041,6 @@ rm -rf %{buildroot}
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/bootstrap-rbd
%post base
-/sbin/ldconfig
%if 0%{?suse_version}
%fillup_only
if [ $1 -eq 1 ] ; then
@@ -1150,7 +1063,6 @@ fi
%endif
%postun base
-/sbin/ldconfig
test -n "$FIRST_ARG" || FIRST_ARG=$1
%if 0%{?suse_version}
DISABLE_RESTART_ON_UPDATE="yes"
@@ -1181,6 +1093,7 @@ fi
%{_bindir}/ceph-dencoder
%{_bindir}/ceph-rbdnamer
%{_bindir}/ceph-syn
+%{_bindir}/ceph-crush-location
%{_bindir}/cephfs-data-scan
%{_bindir}/cephfs-journal-tool
%{_bindir}/cephfs-table-tool
@@ -1198,6 +1111,7 @@ fi
%{_bindir}/rbd-replay-prep
%endif
%{_bindir}/ceph-post-file
+%{_bindir}/ceph-brag
%{_tmpfilesdir}/ceph-common.conf
%{_mandir}/man8/ceph-authtool.8*
%{_mandir}/man8/ceph-conf.8*
@@ -1225,16 +1139,8 @@ fi
%config %{_sysconfdir}/bash_completion.d/radosgw-admin
%config(noreplace) %{_sysconfdir}/ceph/rbdmap
%{_unitdir}/rbdmap.service
-%if 0%{with python2}
%{python_sitelib}/ceph_argparse.py*
%{python_sitelib}/ceph_daemon.py*
-%endif
-%if 0%{with python3}
-%{python3_sitelib}/ceph_argparse.py
-%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
-%{python3_sitelib}/ceph_daemon.py
-%{python3_sitelib}/__pycache__/ceph_daemon.cpython*.py*
-%endif
%dir %{_udevrulesdir}
%{_udevrulesdir}/50-rbd.rules
%attr(3770,ceph,ceph) %dir %{_localstatedir}/log/ceph/
@@ -1376,8 +1282,11 @@ fi
%files mon
%{_bindir}/ceph-mon
+%{_bindir}/ceph-rest-api
%{_bindir}/ceph-monstore-tool
%{_mandir}/man8/ceph-mon.8*
+%{_mandir}/man8/ceph-rest-api.8*
+%{python_sitelib}/ceph_rest_api.py*
%{_unitdir}/ceph-mon@.service
%{_unitdir}/ceph-mon.target
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/mon
@@ -1556,7 +1465,7 @@ fi
%{_mandir}/man8/ceph-bluestore-tool.8*
%{_mandir}/man8/ceph-volume.8*
%{_mandir}/man8/ceph-volume-systemd.8*
-%if ( ( 0%{?rhel} && 0%{?rhel} < 8) && ! 0%{?centos} )
+%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
%attr(0755,-,-) %{_sysconfdir}/cron.hourly/subman
%endif
%{_unitdir}/ceph-osd@.service
@@ -1570,6 +1479,11 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl preset ceph-osd(a)\*.service ceph-volume(a)\*.service ceph-osd.target >/dev/null 2>&1 || :
fi
+%if 0%{?sysctl_apply}
+ %sysctl_apply 90-ceph-osd.conf
+%else
+ /usr/lib/systemd/systemd-sysctl %{_sysctldir}/90-ceph-osd.conf > /dev/null 2>&1 || :
+%endif
%endif
%if 0%{?fedora} || 0%{?rhel}
%systemd_post ceph-osd(a)\*.service ceph-volume(a)\*.service ceph-osd.target
@@ -1577,11 +1491,6 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl start ceph-osd.target >/dev/null 2>&1 || :
fi
-%if 0%{?sysctl_apply}
- %sysctl_apply 90-ceph-osd.conf
-%else
- /usr/lib/systemd/systemd-sysctl %{_sysctldir}/90-ceph-osd.conf > /dev/null 2>&1 || :
-%endif
%preun osd
%if 0%{?suse_version}
@@ -1650,17 +1559,13 @@ fi
%{_bindir}/librados-config
%{_mandir}/man8/librados-config.8*
-%if 0%{with python2}
%files -n python-rados
%{python_sitearch}/rados.so
%{python_sitearch}/rados-*.egg-info
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-rados
%{python3_sitearch}/rados.cpython*.so
%{python3_sitearch}/rados-*.egg-info
-%endif
%ldconfig_scriptlets -n libradosstriper1
%files -n libradosstriper1
@@ -1692,10 +1597,6 @@ fi
%ldconfig_scriptlets -n librgw2
%files -n librgw2
%{_libdir}/librgw.so.*
-%if %{with lttng}
-%{_libdir}/librgw_op_tp.so*
-%{_libdir}/librgw_rados_tp.so*
-%endif
%files -n librgw-devel
%dir %{_includedir}/rados
@@ -1703,29 +1604,21 @@ fi
%{_includedir}/rados/rgw_file.h
%{_libdir}/librgw.so
-%if 0%{with python2}
%files -n python-rgw
%{python_sitearch}/rgw.so
%{python_sitearch}/rgw-*.egg-info
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-rgw
%{python3_sitearch}/rgw.cpython*.so
%{python3_sitearch}/rgw-*.egg-info
-%endif
-%if 0%{with python2}
%files -n python-rbd
%{python_sitearch}/rbd.so
%{python_sitearch}/rbd-*.egg-info
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-rbd
%{python3_sitearch}/rbd.cpython*.so
%{python3_sitearch}/rbd-*.egg-info
-%endif
%ldconfig_scriptlets -n libcephfs2
%files -n libcephfs2
@@ -1737,28 +1630,22 @@ fi
%{_includedir}/cephfs/ceph_statx.h
%{_libdir}/libcephfs.so
-%if 0%{with python2}
%files -n python-cephfs
%{python_sitearch}/cephfs.so
%{python_sitearch}/cephfs-*.egg-info
%{python_sitelib}/ceph_volume_client.py*
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-cephfs
%{python3_sitearch}/cephfs.cpython*.so
%{python3_sitearch}/cephfs-*.egg-info
%{python3_sitelib}/ceph_volume_client.py
%{python3_sitelib}/__pycache__/ceph_volume_client.cpython*.py*
-%endif
-%if 0%{with python3}
%files -n python%{python3_pkgversion}-ceph-argparse
%{python3_sitelib}/ceph_argparse.py
%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
%{python3_sitelib}/ceph_daemon.py
%{python3_sitelib}/__pycache__/ceph_daemon.cpython*.py*
-%endif
%if 0%{with ceph_test_package}
%files -n ceph-test
@@ -1780,7 +1667,13 @@ fi
%{_bindir}/ceph_rgw_multiparser
%{_bindir}/ceph_scratchtool
%{_bindir}/ceph_scratchtoolpp
+%{_bindir}/ceph_smalliobench
+%{_bindir}/ceph_smalliobenchdumb
+%{_bindir}/ceph_smalliobenchfs
+%{_bindir}/ceph_smalliobenchrbd
%{_bindir}/ceph_test_*
+%{_bindir}/ceph_tpbench
+%{_bindir}/ceph_xattr_bench
%{_bindir}/ceph-coverage
%{_bindir}/ceph-debugpack
%{_mandir}/man8/ceph-debugpack.8*
@@ -1850,7 +1743,7 @@ fi
rm -f ${FILE_CONTEXT}.pre
# The fixfiles command won't fix label for /var/run/ceph
-/usr/sbin/restorecon -R %{_rundir}ceph > /dev/null 2>&1
+/usr/sbin/restorecon -R %{_rundir}/ceph > /dev/null 2>&1
# Start the daemons iff they were running before
if test $STATUS -eq 0; then
@@ -1886,7 +1779,7 @@ if [ $1 -eq 0 ]; then
/usr/sbin/fixfiles -C ${FILE_CONTEXT}.pre restore 2> /dev/null
rm -f ${FILE_CONTEXT}.pre
# The fixfiles command won't fix label for /var/run/ceph
- /usr/sbin/restorecon -R %{_rundir}ceph > /dev/null 2>&1
+ /usr/sbin/restorecon -R %{_rundir}/ceph > /dev/null 2>&1
# Start the daemons if they were running before
if test $STATUS -eq 0; then
@@ -1897,28 +1790,14 @@ exit 0
%endif # with selinux
-%if 0%{with python2}
%files -n python-ceph-compat
# We need an empty %%files list for python-ceph-compat, to tell rpmbuild to
# actually build this meta package.
-%endif
%changelog
-* Thu Jun 28 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.2.0-3
-- New release (1:13.2.0-3)
-
-* Tue Jun 19 2018 Miro Hronok <mhroncok(a)redhat.com> - 1:13.1.0-2
-- Rebuilt for Python 3.7
-
-* Thu May 31 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.2.0-1
-- New release (1:13.2.0-1)
-
-* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-2
-- New release (1:13.1.0-2) +crypto_plugins
-
-* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-1
-- New release (1:13.1.0-1)
+* Fri Jun 29 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-2
+- New release (1:12.2.5-2)
* Fri Apr 27 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-1
- New release (1:12.2.5-1)
diff --git a/sources b/sources
index ae5b656..021832c 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-13.2.0.tar.gz) = 2524ad5a868024463e1a1edc5bd9c3f7bc1f58e47bf8d331902ed53206120abc6a427dd9c1d657131e67efd955d6ed8052b6c3fed8658aa905d58ac58ee62973
+SHA512 (ceph-12.2.5.tar.gz) = bd78358590eb077afcc3c1e12864fadee906c95ac8afee25bdca6155f99b75f89d923159fcc13430abba21a89fee3e566177ff5f9ffabdd3b3f220d05602ab86
commit 9094835645403e63f2e8a34a33e8fc9672d9f64d
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Thu Jun 28 02:04:51 2018 -0400
New release (1:13.2.0-3)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch b/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
deleted file mode 100644
index ffc38da..0000000
--- a/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
+++ /dev/null
@@ -1,165 +0,0 @@
---- ceph-13.1.0/src/mon/LogMonitor.cc.orig 2018-05-08 08:12:50.101933925 -0400
-+++ ceph-13.1.0/src/mon/LogMonitor.cc 2018-05-08 08:12:59.470933925 -0400
-@@ -62,7 +62,6 @@
- {
- dout(10) << "create_initial -- creating initial map" << dendl;
- LogEntry e;
-- memset(&e.who, 0, sizeof(e.who));
- e.name = g_conf->name;
- e.stamp = ceph_clock_now();
- e.prio = CLOG_INFO;
---- ceph-13.1.0/src/mon/MonMap.h.orig 2018-05-08 08:13:08.173933925 -0400
-+++ ceph-13.1.0/src/mon/MonMap.h 2018-05-08 08:13:17.391933925 -0400
-@@ -118,7 +118,6 @@
-
- MonMap()
- : epoch(0) {
-- memset(&fsid, 0, sizeof(fsid));
- }
-
- uuid_d& get_fsid() { return fsid; }
---- ceph-13.1.0/src/os/filestore/HashIndex.h.orig 2018-05-08 08:13:35.377933925 -0400
-+++ ceph-13.1.0/src/os/filestore/HashIndex.h 2018-05-08 08:13:59.853933925 -0400
-@@ -395,7 +395,7 @@
-
- struct CmpPairBitwise {
- bool operator()(const pair<string, ghobject_t>& l,
-- const pair<string, ghobject_t>& r)
-+ const pair<string, ghobject_t>& r) const
- {
- if (l.first < r.first)
- return true;
-@@ -408,7 +408,7 @@
- };
-
- struct CmpHexdigitStringBitwise {
-- bool operator()(const string& l, const string& r) {
-+ bool operator()(const string& l, const string& r) const {
- return reverse_hexdigit_bits_string(l) < reverse_hexdigit_bits_string(r);
- }
- };
---- ceph-13.1.0/src/os/filestore/LFNIndex.h.orig 2018-05-08 08:14:06.721933925 -0400
-+++ ceph-13.1.0/src/os/filestore/LFNIndex.h 2018-05-08 08:14:34.671933925 -0400
-@@ -63,7 +63,7 @@
- out: \
- complete_inject_failure(); \
- return r; \
-- } catch (RetryException) { \
-+ } catch (RetryException&) { \
- failed = true; \
- } catch (...) { \
- ceph_abort(); \
---- ceph-13.1.0/src/client/Inode.h.orig 2018-05-08 08:08:26.305933925 -0400
-+++ ceph-13.1.0/src/client/Inode.h 2018-05-08 08:08:38.029933925 -0400
-@@ -279,7 +279,6 @@
- _ref(0), ll_ref(0)
- {
- memset(&dir_layout, 0, sizeof(dir_layout));
-- memset("a, 0, sizeof(quota));
- }
- ~Inode();
-
---- ceph-13.1.0/src/osd/OSDMap.h.orig 2018-05-08 08:14:51.066933925 -0400
-+++ ceph-13.1.0/src/osd/OSDMap.h 2018-05-08 08:15:14.148933925 -0400
-@@ -427,7 +427,6 @@
- encode_features(0),
- epoch(e), new_pool_max(-1), new_flags(-1), new_max_osd(-1),
- have_crc(false), full_crc(0), inc_crc(0) {
-- memset(&fsid, 0, sizeof(fsid));
- }
- explicit Incremental(bufferlist &bl) {
- bufferlist::iterator p = bl.begin();
-@@ -607,7 +606,6 @@
- cached_up_osd_features(0),
- crc_defined(false), crc(0),
- crush(std::make_shared<CrushWrapper>()) {
-- memset(&fsid, 0, sizeof(fsid));
- }
-
- private:
---- ceph-13.1.0/src/common/cmdparse.h.orig 2018-05-08 08:09:17.772933925 -0400
-+++ ceph-13.1.0/src/common/cmdparse.h 2018-05-08 08:09:36.500933925 -0400
-@@ -54,7 +54,7 @@
- try {
- val = boost::get<T>(cmdmap.find(k)->second);
- return true;
-- } catch (boost::bad_get) {
-+ } catch (boost::bad_get&) {
- handle_bad_get(cct, k, typeid(T).name());
- }
- }
---- ceph-13.1.0/src/messages/MClientReply.h.orig 2018-05-08 08:10:22.281933925 -0400
-+++ ceph-13.1.0/src/messages/MClientReply.h 2018-05-08 08:10:46.660933925 -0400
-@@ -187,7 +187,7 @@
- if (features & CEPH_FEATURE_MDS_QUOTA)
- decode(quota, p);
- else
-- memset("a, 0, sizeof(quota));
-+ quota = quota_info_t{};
-
- if ((features & CEPH_FEATURE_FS_FILE_LAYOUT_V2))
- decode(layout.pool_ns, p);
---- ceph-13.1.0/src/messages/MMonSubscribeAck.h.orig 2018-05-08 08:10:58.478933925 -0400
-+++ ceph-13.1.0/src/messages/MMonSubscribeAck.h 2018-05-08 08:11:17.484933925 -0400
-@@ -23,7 +23,6 @@
-
- MMonSubscribeAck() : Message(CEPH_MSG_MON_SUBSCRIBE_ACK),
- interval(0) {
-- memset(&fsid, 0, sizeof(fsid));
- }
- MMonSubscribeAck(uuid_d& f, int i) : Message(CEPH_MSG_MON_SUBSCRIBE_ACK),
- interval(i), fsid(f) { }
---- ceph-13.1.0/src/mgr/DaemonState.h.orig 2018-05-08 08:12:16.521933925 -0400
-+++ ceph-13.1.0/src/mgr/DaemonState.h 2018-05-08 08:12:40.995933925 -0400
-@@ -130,7 +130,7 @@
- auto p = config_defaults_bl.begin();
- try {
- decode(config_defaults, p);
-- } catch (buffer::error e) {
-+ } catch (buffer::error& e) {
- }
- }
- return config_defaults;
---- ceph-13.1.0/cmake/modules/BuildDPDK.cmake.orig 2018-05-08 08:41:02.168933925 -0400
-+++ ceph-13.1.0/cmake/modules/BuildDPDK.cmake 2018-05-08 08:41:47.411933925 -0400
-@@ -71,7 +71,7 @@
- BUILD_IN_SOURCE 1
- INSTALL_COMMAND "true")
- ExternalProject_Add_Step(dpdk-ext patch-config
-- COMMAND ${CMAKE_MODULE_PATH}/patch-dpdk-conf.sh ${dpdk_dir} ${machine}
-+ COMMAND ${CMAKE_MODULE_PATH}/patch-dpdk-conf.sh ${dpdk_dir} ${machine} ${arch}
- DEPENDEES configure
- DEPENDERS build)
- # easier to adjust the config
-@@ -86,7 +86,7 @@
- # target
- file(MAKE_DIRECTORY ${DPDK_INCLUDE_DIR})
- foreach(c
-- pci bus_pci
-+ bus_pci pci
- eal
- mempool mempool_ring mempool_stack ring)
- add_library(dpdk::${c} STATIC IMPORTED)
---- ceph-13.1.0/cmake/modules/patch-dpdk-conf.sh.orig 2018-05-08 08:42:01.089933925 -0400
-+++ ceph-13.1.0/cmake/modules/patch-dpdk-conf.sh 2018-05-08 08:43:11.781933925 -0400
-@@ -15,8 +15,12 @@
- shift
- machine=$1
- shift
-+arch=$1
-+shift
-
- setconf CONFIG_RTE_MACHINE "${machine}"
-+setconf CONFIG_RTE_ARCH "${arch}"
-+
- # Disable experimental features
- setconf CONFIG_RTE_NEXT_ABI n
- setconf CONFIG_RTE_LIBRTE_MBUF_OFFLOAD n
-@@ -38,6 +42,7 @@
- setconf CONFIG_RTE_LIBRTE_VMXNET3_PMD n
- setconf CONFIG_RTE_LIBRTE_PMD_VHOST n
- setconf CONFIG_RTE_APP_EVENTDEV n
-+setconf CONFIG_RTE_MAX_VFIO_GROUPS 64
-
- # no test
- setconf CONFIG_RTE_APP_TEST n
diff --git a/0001-blobstore.patch b/0001-blobstore.patch
new file mode 100644
index 0000000..97462f0
--- /dev/null
+++ b/0001-blobstore.patch
@@ -0,0 +1,11 @@
+--- ceph-13.1.0/src/spdk/lib/blob/blobstore.c.orig 2018-05-08 16:04:30.329933925 -0400
++++ ceph-13.1.0/src/spdk/lib/blob/blobstore.c 2018-05-08 16:04:55.240933925 -0400
+@@ -2947,7 +2947,7 @@
+
+ /* START spdk_blob_resize */
+ int
+-spdk_blob_resize(struct spdk_blob *_blob, uint64_t sz)
++spdk_blob_resize(struct spdk_blob *_blob, size_t sz)
+ {
+ struct spdk_blob_data *blob = __blob_to_data(_blob);
+ int rc;
diff --git a/ceph.spec b/ceph.spec
index 8b0de11..96ba41f 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -16,6 +16,7 @@
#
%global _hardened_build 1
+%bcond_with python3
%bcond_without ocf
%bcond_with make_check
%ifarch s390 s390x
@@ -28,6 +29,7 @@
%bcond_without ceph_test_package
%bcond_without cephfs_java
%bcond_without lttng
+%global _remote_tarball_prefix https://download.ceph.com/tarballs/
%endif
%if 0%{?suse_version}
%bcond_with selinux
@@ -75,22 +77,23 @@
%global _find_debuginfo_dwz_opts %{nil}
%if ( 0%{?rhel} && 0%{?rhel} < 7 )
-%global _rundir %{_localstatedir}/run
+%global _rundir %{_localstatedir}/run/
%else
-%global _rundir /run
+%global _rundir /run/
%endif
#################################################################################
# main package definition
#################################################################################
Name: ceph
-Version: 13.1.0
-Release: 2%{?dist}
+Version: 13.2.0
+Release: 3%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
-# define %%_epoch_prefix macro which will expand to the empty string if %%epoch is undefined
+# define _epoch_prefix macro which will expand to the empty string if epoch is
+# undefined
%global _epoch_prefix %{?epoch:%{epoch}:}
Summary: User space components of the Ceph file system
@@ -99,14 +102,9 @@ License: LGPL-2.1 and CC-BY-SA-3.0 and GPL-2.0 and BSL-1.0 and BSD-3-Clause and
Group: System/Filesystems
%endif
URL: http://ceph.com/
-Source0: http://download.ceph.com/tarballs/%{name}-%{version}.tar.gz
+Source0: %{?_remote_tarball_prefix}%{name}-%{version}.tar.gz
# https://bugzilla.redhat.com/show_bug.cgi?id=1474773
-Patch001: 0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
-#Patch001: 0001-src-rocksdb-util-murmurhash.patch
-# https://bugzilla.redhat.com/show_bug.cgi?id=1474774
-#Patch002: 0002-cmake-Support-ppc64.patch
-#Patch003: 0003-librbd-Conditionally-import-TrimRequest.cc.patch
-#Patch005: 0005-src-rocksdb-table-block.h.patch
+Patch001: 0001-blobstore.patch
%if 0%{?suse_version}
# _insert_obs_source_lines_here
%if 0%{?is_opensuse}
@@ -136,6 +134,7 @@ BuildRequires: gperf
BuildRequires: cmake
BuildRequires: cryptsetup
BuildRequires: fuse-devel
+BuildRequires: cryptopp-devel
%if 0%{?rhel} == 7
# devtoolset offers newer make and valgrind-devel, but the old ones are good
# enough.
@@ -204,7 +203,6 @@ BuildRequires: python%{_python_buildid}-PrettyTable
BuildRequires: python%{_python_buildid}-Sphinx
BuildRequires: rdma-core-devel
BuildRequires: liblz4-devel >= 1.7
-BuildRequires: rdma-core-devel
%endif
%if 0%{?fedora} || 0%{?rhel}
Requires: systemd
@@ -212,7 +210,7 @@ BuildRequires: boost-random
BuildRequires: btrfs-progs
BuildRequires: nss-devel
BuildRequires: keyutils-libs-devel
-# RDMA is no longer built on 32-bit ARM: see #1484155
+# RDMA is no longer built on 32-bit ARM: see rhbz#1484155
%ifnarch %{arm}
BuildRequires: rdma-core-devel
%endif
@@ -226,7 +224,8 @@ BuildRequires: python%{_python_buildid}-sphinx
BuildRequires: lz4-devel >= 1.7
%endif
# python34-... for RHEL, python3-... for all other supported distros
-%if ( 0%{?rhel} && 0%{?rhel} <= 7 )
+%if %{with python3}
+%if 0%{?rhel}
BuildRequires: python34-devel
BuildRequires: python34-setuptools
BuildRequires: python34-Cython
@@ -235,6 +234,7 @@ BuildRequires: python3-devel
BuildRequires: python3-setuptools
BuildRequires: python3-Cython
%endif
+%endif
# distro-conditional make check dependencies
%if 0%{with make_check}
%if 0%{?fedora} || 0%{?rhel}
@@ -394,6 +394,12 @@ Requires: python%{_python_buildid}-jinja2
Requires: python%{_python_buildid}-pecan
Requires: python%{_python_buildid}-werkzeug
Requires: pyOpenSSL%{_python_buildid}
+%if 0%{?fedora}
+Requires: python%{_python_buildid}-bcrypt
+%endif
+%if 0%{?rhel}
+Requires: py-bcrypt
+%endif
%endif
%if 0%{?suse_version}
Requires: python%{_python_buildid}-CherryPy
@@ -562,6 +568,7 @@ This package contains Python 2 libraries for interacting with Cephs RADOS
gateway.
%endif
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-rgw
Summary: Python 3 libraries for the RADOS gateway
%if 0%{?suse_version}
@@ -572,6 +579,7 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
%description -n python%{python3_pkgversion}-rgw
This package contains Python 3 libraries for interacting with Cephs RADOS
gateway.
+%endif
%if 0%{with python2}
%package -n python-rados
@@ -586,6 +594,7 @@ This package contains Python 2 libraries for interacting with Cephs RADOS
object store.
%endif
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-rados
Summary: Python 3 libraries for the RADOS object store
%if 0%{?suse_version}
@@ -596,6 +605,7 @@ Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
%description -n python%{python3_pkgversion}-rados
This package contains Python 3 libraries for interacting with Cephs RADOS
object store.
+%endif
%package -n libradosstriper1
Summary: RADOS striping interface
@@ -669,6 +679,7 @@ This package contains Python 2 libraries for interacting with Cephs RADOS
block device.
%endif
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-rbd
Summary: Python 3 libraries for the RADOS block device
%if 0%{?suse_version}
@@ -679,6 +690,7 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
%description -n python%{python3_pkgversion}-rbd
This package contains Python 3 libraries for interacting with Cephs RADOS
block device.
+%endif
%package -n libcephfs2
Summary: Ceph distributed file system client library
@@ -727,6 +739,7 @@ This package contains Python 2 libraries for interacting with Cephs distributed
file system.
%endif
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-cephfs
Summary: Python 3 libraries for Ceph distributed file system
%if 0%{?suse_version}
@@ -737,8 +750,9 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
%description -n python%{python3_pkgversion}-cephfs
This package contains Python 3 libraries for interacting with Cephs distributed
file system.
+%endif
-%if 0%{with python2}
+%if 0%{with python3}
%package -n python%{python3_pkgversion}-ceph-argparse
Summary: Python 3 utility libraries for Ceph CLI
%if 0%{?suse_version}
@@ -919,7 +933,11 @@ cmake .. \
-DCMAKE_INSTALL_INCLUDEDIR=%{_includedir} \
-DWITH_EMBEDDED=OFF \
-DWITH_MANPAGE=ON \
+%if %{with python3}
-DWITH_PYTHON3=ON \
+%else
+ -DWITH_PYTHON3=OFF \
+%endif
-DWITH_MGR_DASHBOARD_FRONTEND=OFF \
%if %{with python2}
-DWITH_PYTHON2=ON \
@@ -927,7 +945,7 @@ cmake .. \
-DWITH_PYTHON2=OFF \
-DMGR_PYTHON_VERSION=3 \
%endif
-%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
+%if ( ( 0%{?rhel} && 0%{?rhel} < 8) && ! 0%{?centos} )
-DWITH_SUBMAN=ON \
%endif
%if 0%{without ceph_test_package}
@@ -955,13 +973,15 @@ cmake .. \
%else
-DWITH_BOOST_CONTEXT=OFF \
%endif
-%ifnarch %{arm}
+%ifarch %{arm}
-DWITH_RDMA=OFF \
%endif
+ -DWITH_DPDK=ON \
-DBOOST_J=$CEPH_SMP_NCPUS
make "$CEPH_MFLAGS_JOBS"
+
%if 0%{with make_check}
%check
# run in-tree unittests
@@ -1006,7 +1026,7 @@ install -m 0644 -D udev/95-ceph-osd.rules %{buildroot}%{_udevrulesdir}/95-ceph-o
#set up placeholder directories
mkdir -p %{buildroot}%{_sysconfdir}/ceph
-mkdir -p %{buildroot}%{_rundir}/ceph
+mkdir -p %{buildroot}%{_rundir}ceph
mkdir -p %{buildroot}%{_localstatedir}/log/ceph
mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/tmp
mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/mon
@@ -1025,6 +1045,9 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%py3_compile %{buildroot}%{python3_sitelib}
%endif
+%clean
+rm -rf %{buildroot}
+
#################################################################################
# files and systemd scriptlets
#################################################################################
@@ -1071,7 +1094,8 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%if 0%{with python2}
%{python_sitelib}/ceph_detect_init*
%{python_sitelib}/ceph_disk*
-%else
+%endif
+%if 0%{with python3}
%{python3_sitelib}/ceph_detect_init*
%{python3_sitelib}/ceph_disk*
%endif
@@ -1079,7 +1103,8 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%dir %{python_sitelib}/ceph_volume
%{python_sitelib}/ceph_volume/*
%{python_sitelib}/ceph_volume-*
-%else
+%endif
+%if 0%{with python3}
%dir %{python3_sitelib}/ceph_volume
%{python3_sitelib}/ceph_volume/*
%{python3_sitelib}/ceph_volume-*
@@ -1102,6 +1127,7 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/bootstrap-rbd
%post base
+/sbin/ldconfig
%if 0%{?suse_version}
%fillup_only
if [ $1 -eq 1 ] ; then
@@ -1124,6 +1150,7 @@ fi
%endif
%postun base
+/sbin/ldconfig
test -n "$FIRST_ARG" || FIRST_ARG=$1
%if 0%{?suse_version}
DISABLE_RESTART_ON_UPDATE="yes"
@@ -1171,7 +1198,6 @@ fi
%{_bindir}/rbd-replay-prep
%endif
%{_bindir}/ceph-post-file
-%{_bindir}/ceph-brag
%{_tmpfilesdir}/ceph-common.conf
%{_mandir}/man8/ceph-authtool.8*
%{_mandir}/man8/ceph-conf.8*
@@ -1202,7 +1228,8 @@ fi
%if 0%{with python2}
%{python_sitelib}/ceph_argparse.py*
%{python_sitelib}/ceph_daemon.py*
-%else
+%endif
+%if 0%{with python3}
%{python3_sitelib}/ceph_argparse.py
%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
%{python3_sitelib}/ceph_daemon.py
@@ -1529,7 +1556,7 @@ fi
%{_mandir}/man8/ceph-bluestore-tool.8*
%{_mandir}/man8/ceph-volume.8*
%{_mandir}/man8/ceph-volume-systemd.8*
-%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
+%if ( ( 0%{?rhel} && 0%{?rhel} < 8) && ! 0%{?centos} )
%attr(0755,-,-) %{_sysconfdir}/cron.hourly/subman
%endif
%{_unitdir}/ceph-osd@.service
@@ -1543,11 +1570,6 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl preset ceph-osd(a)\*.service ceph-volume(a)\*.service ceph-osd.target >/dev/null 2>&1 || :
fi
-%if 0%{?sysctl_apply}
- %sysctl_apply 90-ceph-osd.conf
-%else
- /usr/lib/systemd/systemd-sysctl %{_sysctldir}/90-ceph-osd.conf > /dev/null 2>&1 || :
-%endif
%endif
%if 0%{?fedora} || 0%{?rhel}
%systemd_post ceph-osd(a)\*.service ceph-volume(a)\*.service ceph-osd.target
@@ -1555,6 +1577,11 @@ fi
if [ $1 -eq 1 ] ; then
/usr/bin/systemctl start ceph-osd.target >/dev/null 2>&1 || :
fi
+%if 0%{?sysctl_apply}
+ %sysctl_apply 90-ceph-osd.conf
+%else
+ /usr/lib/systemd/systemd-sysctl %{_sysctldir}/90-ceph-osd.conf > /dev/null 2>&1 || :
+%endif
%preun osd
%if 0%{?suse_version}
@@ -1629,9 +1656,11 @@ fi
%{python_sitearch}/rados-*.egg-info
%endif
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-rados
%{python3_sitearch}/rados.cpython*.so
%{python3_sitearch}/rados-*.egg-info
+%endif
%ldconfig_scriptlets -n libradosstriper1
%files -n libradosstriper1
@@ -1680,9 +1709,11 @@ fi
%{python_sitearch}/rgw-*.egg-info
%endif
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-rgw
%{python3_sitearch}/rgw.cpython*.so
%{python3_sitearch}/rgw-*.egg-info
+%endif
%if 0%{with python2}
%files -n python-rbd
@@ -1690,9 +1721,11 @@ fi
%{python_sitearch}/rbd-*.egg-info
%endif
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-rbd
%{python3_sitearch}/rbd.cpython*.so
%{python3_sitearch}/rbd-*.egg-info
+%endif
%ldconfig_scriptlets -n libcephfs2
%files -n libcephfs2
@@ -1711,13 +1744,15 @@ fi
%{python_sitelib}/ceph_volume_client.py*
%endif
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-cephfs
%{python3_sitearch}/cephfs.cpython*.so
%{python3_sitearch}/cephfs-*.egg-info
%{python3_sitelib}/ceph_volume_client.py
%{python3_sitelib}/__pycache__/ceph_volume_client.cpython*.py*
+%endif
-%if 0%{with python2}
+%if 0%{with python3}
%files -n python%{python3_pkgversion}-ceph-argparse
%{python3_sitelib}/ceph_argparse.py
%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
@@ -1815,7 +1850,7 @@ fi
rm -f ${FILE_CONTEXT}.pre
# The fixfiles command won't fix label for /var/run/ceph
-/usr/sbin/restorecon -R %{_rundir}/ceph > /dev/null 2>&1
+/usr/sbin/restorecon -R %{_rundir}ceph > /dev/null 2>&1
# Start the daemons iff they were running before
if test $STATUS -eq 0; then
@@ -1851,7 +1886,7 @@ if [ $1 -eq 0 ]; then
/usr/sbin/fixfiles -C ${FILE_CONTEXT}.pre restore 2> /dev/null
rm -f ${FILE_CONTEXT}.pre
# The fixfiles command won't fix label for /var/run/ceph
- /usr/sbin/restorecon -R %{_rundir}/ceph > /dev/null 2>&1
+ /usr/sbin/restorecon -R %{_rundir}ceph > /dev/null 2>&1
# Start the daemons if they were running before
if test $STATUS -eq 0; then
@@ -1870,9 +1905,18 @@ exit 0
%changelog
+* Thu Jun 28 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.2.0-3
+- New release (1:13.2.0-3)
+
* Tue Jun 19 2018 Miro Hronok <mhroncok(a)redhat.com> - 1:13.1.0-2
- Rebuilt for Python 3.7
+* Thu May 31 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.2.0-1
+- New release (1:13.2.0-1)
+
+* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-2
+- New release (1:13.1.0-2) +crypto_plugins
+
* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-1
- New release (1:13.1.0-1)
diff --git a/sources b/sources
index e0ac550..ae5b656 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-13.1.0.tar.gz) = d4feb0b0bac5e30232eddcc3f69c1d939f0b5268491acecba9a58dcadbbdaab6367f415ede924ca84de9a89a686410613cdc631f8aff15aec0eb4531138eec7a
+SHA512 (ceph-13.2.0.tar.gz) = 2524ad5a868024463e1a1edc5bd9c3f7bc1f58e47bf8d331902ed53206120abc6a427dd9c1d657131e67efd955d6ed8052b6c3fed8658aa905d58ac58ee62973
commit 200f2e06d2a2cdc3ac4483467e9aff7ded9599c6
Author: Miro Hronok <miro(a)hroncok.cz>
Date: Tue Jun 19 10:41:00 2018 +0200
Rebuilt for Python 3.7
diff --git a/ceph.spec b/ceph.spec
index 1fa45d7..8b0de11 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -85,7 +85,7 @@
#################################################################################
Name: ceph
Version: 13.1.0
-Release: 1%{?dist}
+Release: 2%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
%endif
@@ -1870,6 +1870,9 @@ exit 0
%changelog
+* Tue Jun 19 2018 Miro Hronok <mhroncok(a)redhat.com> - 1:13.1.0-2
+- Rebuilt for Python 3.7
+
* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-1
- New release (1:13.1.0-1)
commit b79737d3dfe8949eee6764ee6298ee6a715a3042
Author: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
Date: Tue May 8 11:57:37 2018 -0400
New release (1:13.1.0-1)
Signed-off-by: Kaleb S. KEITHLEY <kkeithle(a)redhat.com>
diff --git a/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch b/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
new file mode 100644
index 0000000..ffc38da
--- /dev/null
+++ b/0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
@@ -0,0 +1,165 @@
+--- ceph-13.1.0/src/mon/LogMonitor.cc.orig 2018-05-08 08:12:50.101933925 -0400
++++ ceph-13.1.0/src/mon/LogMonitor.cc 2018-05-08 08:12:59.470933925 -0400
+@@ -62,7 +62,6 @@
+ {
+ dout(10) << "create_initial -- creating initial map" << dendl;
+ LogEntry e;
+- memset(&e.who, 0, sizeof(e.who));
+ e.name = g_conf->name;
+ e.stamp = ceph_clock_now();
+ e.prio = CLOG_INFO;
+--- ceph-13.1.0/src/mon/MonMap.h.orig 2018-05-08 08:13:08.173933925 -0400
++++ ceph-13.1.0/src/mon/MonMap.h 2018-05-08 08:13:17.391933925 -0400
+@@ -118,7 +118,6 @@
+
+ MonMap()
+ : epoch(0) {
+- memset(&fsid, 0, sizeof(fsid));
+ }
+
+ uuid_d& get_fsid() { return fsid; }
+--- ceph-13.1.0/src/os/filestore/HashIndex.h.orig 2018-05-08 08:13:35.377933925 -0400
++++ ceph-13.1.0/src/os/filestore/HashIndex.h 2018-05-08 08:13:59.853933925 -0400
+@@ -395,7 +395,7 @@
+
+ struct CmpPairBitwise {
+ bool operator()(const pair<string, ghobject_t>& l,
+- const pair<string, ghobject_t>& r)
++ const pair<string, ghobject_t>& r) const
+ {
+ if (l.first < r.first)
+ return true;
+@@ -408,7 +408,7 @@
+ };
+
+ struct CmpHexdigitStringBitwise {
+- bool operator()(const string& l, const string& r) {
++ bool operator()(const string& l, const string& r) const {
+ return reverse_hexdigit_bits_string(l) < reverse_hexdigit_bits_string(r);
+ }
+ };
+--- ceph-13.1.0/src/os/filestore/LFNIndex.h.orig 2018-05-08 08:14:06.721933925 -0400
++++ ceph-13.1.0/src/os/filestore/LFNIndex.h 2018-05-08 08:14:34.671933925 -0400
+@@ -63,7 +63,7 @@
+ out: \
+ complete_inject_failure(); \
+ return r; \
+- } catch (RetryException) { \
++ } catch (RetryException&) { \
+ failed = true; \
+ } catch (...) { \
+ ceph_abort(); \
+--- ceph-13.1.0/src/client/Inode.h.orig 2018-05-08 08:08:26.305933925 -0400
++++ ceph-13.1.0/src/client/Inode.h 2018-05-08 08:08:38.029933925 -0400
+@@ -279,7 +279,6 @@
+ _ref(0), ll_ref(0)
+ {
+ memset(&dir_layout, 0, sizeof(dir_layout));
+- memset("a, 0, sizeof(quota));
+ }
+ ~Inode();
+
+--- ceph-13.1.0/src/osd/OSDMap.h.orig 2018-05-08 08:14:51.066933925 -0400
++++ ceph-13.1.0/src/osd/OSDMap.h 2018-05-08 08:15:14.148933925 -0400
+@@ -427,7 +427,6 @@
+ encode_features(0),
+ epoch(e), new_pool_max(-1), new_flags(-1), new_max_osd(-1),
+ have_crc(false), full_crc(0), inc_crc(0) {
+- memset(&fsid, 0, sizeof(fsid));
+ }
+ explicit Incremental(bufferlist &bl) {
+ bufferlist::iterator p = bl.begin();
+@@ -607,7 +606,6 @@
+ cached_up_osd_features(0),
+ crc_defined(false), crc(0),
+ crush(std::make_shared<CrushWrapper>()) {
+- memset(&fsid, 0, sizeof(fsid));
+ }
+
+ private:
+--- ceph-13.1.0/src/common/cmdparse.h.orig 2018-05-08 08:09:17.772933925 -0400
++++ ceph-13.1.0/src/common/cmdparse.h 2018-05-08 08:09:36.500933925 -0400
+@@ -54,7 +54,7 @@
+ try {
+ val = boost::get<T>(cmdmap.find(k)->second);
+ return true;
+- } catch (boost::bad_get) {
++ } catch (boost::bad_get&) {
+ handle_bad_get(cct, k, typeid(T).name());
+ }
+ }
+--- ceph-13.1.0/src/messages/MClientReply.h.orig 2018-05-08 08:10:22.281933925 -0400
++++ ceph-13.1.0/src/messages/MClientReply.h 2018-05-08 08:10:46.660933925 -0400
+@@ -187,7 +187,7 @@
+ if (features & CEPH_FEATURE_MDS_QUOTA)
+ decode(quota, p);
+ else
+- memset("a, 0, sizeof(quota));
++ quota = quota_info_t{};
+
+ if ((features & CEPH_FEATURE_FS_FILE_LAYOUT_V2))
+ decode(layout.pool_ns, p);
+--- ceph-13.1.0/src/messages/MMonSubscribeAck.h.orig 2018-05-08 08:10:58.478933925 -0400
++++ ceph-13.1.0/src/messages/MMonSubscribeAck.h 2018-05-08 08:11:17.484933925 -0400
+@@ -23,7 +23,6 @@
+
+ MMonSubscribeAck() : Message(CEPH_MSG_MON_SUBSCRIBE_ACK),
+ interval(0) {
+- memset(&fsid, 0, sizeof(fsid));
+ }
+ MMonSubscribeAck(uuid_d& f, int i) : Message(CEPH_MSG_MON_SUBSCRIBE_ACK),
+ interval(i), fsid(f) { }
+--- ceph-13.1.0/src/mgr/DaemonState.h.orig 2018-05-08 08:12:16.521933925 -0400
++++ ceph-13.1.0/src/mgr/DaemonState.h 2018-05-08 08:12:40.995933925 -0400
+@@ -130,7 +130,7 @@
+ auto p = config_defaults_bl.begin();
+ try {
+ decode(config_defaults, p);
+- } catch (buffer::error e) {
++ } catch (buffer::error& e) {
+ }
+ }
+ return config_defaults;
+--- ceph-13.1.0/cmake/modules/BuildDPDK.cmake.orig 2018-05-08 08:41:02.168933925 -0400
++++ ceph-13.1.0/cmake/modules/BuildDPDK.cmake 2018-05-08 08:41:47.411933925 -0400
+@@ -71,7 +71,7 @@
+ BUILD_IN_SOURCE 1
+ INSTALL_COMMAND "true")
+ ExternalProject_Add_Step(dpdk-ext patch-config
+- COMMAND ${CMAKE_MODULE_PATH}/patch-dpdk-conf.sh ${dpdk_dir} ${machine}
++ COMMAND ${CMAKE_MODULE_PATH}/patch-dpdk-conf.sh ${dpdk_dir} ${machine} ${arch}
+ DEPENDEES configure
+ DEPENDERS build)
+ # easier to adjust the config
+@@ -86,7 +86,7 @@
+ # target
+ file(MAKE_DIRECTORY ${DPDK_INCLUDE_DIR})
+ foreach(c
+- pci bus_pci
++ bus_pci pci
+ eal
+ mempool mempool_ring mempool_stack ring)
+ add_library(dpdk::${c} STATIC IMPORTED)
+--- ceph-13.1.0/cmake/modules/patch-dpdk-conf.sh.orig 2018-05-08 08:42:01.089933925 -0400
++++ ceph-13.1.0/cmake/modules/patch-dpdk-conf.sh 2018-05-08 08:43:11.781933925 -0400
+@@ -15,8 +15,12 @@
+ shift
+ machine=$1
+ shift
++arch=$1
++shift
+
+ setconf CONFIG_RTE_MACHINE "${machine}"
++setconf CONFIG_RTE_ARCH "${arch}"
++
+ # Disable experimental features
+ setconf CONFIG_RTE_NEXT_ABI n
+ setconf CONFIG_RTE_LIBRTE_MBUF_OFFLOAD n
+@@ -38,6 +42,7 @@
+ setconf CONFIG_RTE_LIBRTE_VMXNET3_PMD n
+ setconf CONFIG_RTE_LIBRTE_PMD_VHOST n
+ setconf CONFIG_RTE_APP_EVENTDEV n
++setconf CONFIG_RTE_MAX_VFIO_GROUPS 64
+
+ # no test
+ setconf CONFIG_RTE_APP_TEST n
diff --git a/0001-src-rocksdb-util-murmurhash.patch b/0001-src-rocksdb-util-murmurhash.patch
deleted file mode 100644
index f67a0ff..0000000
--- a/0001-src-rocksdb-util-murmurhash.patch
+++ /dev/null
@@ -1,13 +0,0 @@
---- ceph-12.1.1.orig/src/rocksdb/util/murmurhash.cc 2017-04-27 01:13:46.000000000 +0100
-+++ ceph-12.1.1.orig/src/rocksdb/util/murmurhash.cc 2017-07-25 11:37:28.910266684 +0100
-@@ -113,8 +113,8 @@ unsigned int MurmurHash2 ( const void *
-
- switch(len)
- {
-- case 3: h ^= data[2] << 16;
-- case 2: h ^= data[1] << 8;
-+ case 3: h ^= data[2] << 16; // fallthrough
-+ case 2: h ^= data[1] << 8; // fallthrough
- case 1: h ^= data[0];
- h *= m;
- };
diff --git a/0002-cmake-Support-ppc64.patch b/0002-cmake-Support-ppc64.patch
deleted file mode 100644
index 6b09549..0000000
--- a/0002-cmake-Support-ppc64.patch
+++ /dev/null
@@ -1,127 +0,0 @@
-From 2f0a7153460acc3f21462236f470ec3471fa2ee1 Mon Sep 17 00:00:00 2001
-From: Boris Ranto <branto(a)redhat.com>
-Date: Mon, 31 Jul 2017 19:50:23 +0200
-Subject: [PATCH] cmake: Support ppc64
-
-The ppc64 support requires a couple of changes:
-- adding the ppc64 support to cmake
-- changing optimized crc32 code to compile on ppc64le only
-- moving ifdef condition before crc32_align to avoid defined but not
-used warning
-
-Signed-off-by: Boris Ranto <branto(a)redhat.com>
----
- cmake/modules/SIMDExt.cmake | 15 ++++++++++++++-
- src/CMakeLists.txt | 4 +++-
- src/arch/ppc.c | 8 ++++----
- src/common/crc32c_ppc.c | 6 +++---
- 4 files changed, 24 insertions(+), 9 deletions(-)
-
-diff --git a/cmake/modules/SIMDExt.cmake b/cmake/modules/SIMDExt.cmake
-index 5330835..c47667d 100644
---- a/cmake/modules/SIMDExt.cmake
-+++ b/cmake/modules/SIMDExt.cmake
-@@ -109,7 +109,20 @@ elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "i386|i686|amd64|x86_64|AMD64")
- endif(CMAKE_SYSTEM_PROCESSOR MATCHES "i686|amd64|x86_64|AMD64")
- elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "(powerpc|ppc)64le")
- set(HAVE_PPC64LE 1)
-- message(STATUS " we are ppc64le")
-+ message(STATUS " we are ppc64")
-+ CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
-+ if(HAS_ALTIVEC)
-+ message(STATUS " HAS_ALTIVEC yes")
-+ set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} -maltivec")
-+ set(CMAKE_CXX_FLAGS "${CMAKE_C_FLAGS} -maltivec")
-+ endif()
-+ CHECK_C_COMPILER_FLAG("-mcpu=power8" HAVE_POWER8)
-+ if(HAVE_POWER8)
-+ message(STATUS " HAVE_POWER8 yes")
-+ endif()
-+elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "(power|ppc)64")
-+ set(HAVE_PPC64 1)
-+ message(STATUS " we are ppc64")
- CHECK_C_COMPILER_FLAG("-maltivec" HAS_ALTIVEC)
- if(HAS_ALTIVEC)
- message(STATUS " HAS_ALTIVEC yes")
-diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt
-index 66f0c14..38d1913 100644
---- a/src/CMakeLists.txt
-+++ b/src/CMakeLists.txt
-@@ -568,7 +568,9 @@ if(HAVE_INTEL)
- endif(HAVE_GOOD_YASM_ELF64)
- elseif(HAVE_POWER8)
- list(APPEND libcommon_files
-- common/crc32c_ppc.c
-+ common/crc32c_ppc.c)
-+elseif(HAVE_PPC64LE)
-+ list(APPEND libcommon_files
- common/crc32c_ppc_asm.S
- common/crc32c_ppc_fast_zero_asm.S)
- endif(HAVE_INTEL)
-diff --git a/src/arch/ppc.c b/src/arch/ppc.c
-index f21e2fe..11d3a49 100644
---- a/src/arch/ppc.c
-+++ b/src/arch/ppc.c
-@@ -14,10 +14,10 @@ int ceph_arch_ppc_crc32 = 0;
-
- #include <stdio.h>
-
--#if __linux__ && __powerpc64__
-+#ifdef HAVE_PPC64LE
- #include <sys/auxv.h>
- #include <asm/cputable.h>
--#endif /* __linux__ && __powerpc64__ */
-+#endif /* HAVE_PPC64LE */
-
- #ifndef PPC_FEATURE2_VEC_CRYPTO
- #define PPC_FEATURE2_VEC_CRYPTO 0x02000000
-@@ -31,9 +31,9 @@ int ceph_arch_ppc_probe(void)
- {
- ceph_arch_ppc_crc32 = 0;
-
--#if __linux__ && __powerpc64__
-+#ifdef HAVE_PPC64LE
- if (getauxval(AT_HWCAP2) & PPC_FEATURE2_VEC_CRYPTO) ceph_arch_ppc_crc32 = 1;
--#endif /* __linux__ && __powerpc64__ */
-+#endif /* HAVE_PPC64LE */
-
- return 0;
- }
-diff --git a/src/common/crc32c_ppc.c b/src/common/crc32c_ppc.c
-index 43756e2..52fd1c4 100644
---- a/src/common/crc32c_ppc.c
-+++ b/src/common/crc32c_ppc.c
-@@ -20,6 +20,7 @@
- #define VMX_ALIGN 16
- #define VMX_ALIGN_MASK (VMX_ALIGN-1)
-
-+#ifdef HAVE_PPC64LE
- #ifdef REFLECT
- static unsigned int crc32_align(unsigned int crc, unsigned char const *p,
- unsigned long len)
-@@ -38,7 +39,6 @@ static unsigned int crc32_align(unsigned int crc, unsigned char const *p,
- }
- #endif
-
--#ifdef HAVE_POWER8
- static inline unsigned long polynomial_multiply(unsigned int a, unsigned int b) {
- vector unsigned int va = {a, 0, 0, 0};
- vector unsigned int vb = {b, 0, 0, 0};
-@@ -134,7 +134,7 @@ uint32_t ceph_crc32c_ppc(uint32_t crc, unsigned char const *data, unsigned len)
- return crc;
- }
-
--#else /* HAVE_POWER8 */
-+#else /* HAVE_PPC64LE */
-
- /* This symbol has to exist on non-ppc architectures (and on legacy
- * ppc systems using power7 or below) in order to compile properly
-@@ -145,4 +145,4 @@ uint32_t ceph_crc32c_ppc(uint32_t crc, unsigned char const *data, unsigned len)
- return 0;
- }
-
--#endif /* HAVE_POWER8 */
-+#endif /* HAVE_PPC64LE */
---
-2.9.4
-
diff --git a/0003-librbd-Conditionally-import-TrimRequest.cc.patch b/0003-librbd-Conditionally-import-TrimRequest.cc.patch
deleted file mode 100644
index 2f02efb..0000000
--- a/0003-librbd-Conditionally-import-TrimRequest.cc.patch
+++ /dev/null
@@ -1,31 +0,0 @@
-From 74a754690736f6608b0d4d9c807df0bd777a129d Mon Sep 17 00:00:00 2001
-From: Boris Ranto <branto(a)redhat.com>
-Date: Fri, 8 Dec 2017 00:21:38 +0100
-Subject: [PATCH] librbd: Conditionally import TrimRequest.cc
-
-We include TrimRequest.cc in librbd tests at two places:
- - operation/test_mock_TrimRequest.cc
- - operation/test_mock_ResizeRequest.cc
-
-That causes linking errors when doing the builds because some of the
-structures are defined twice.
-
-Signed-off-by: Boris Ranto <branto(a)redhat.com>
----
- src/librbd/operation/TrimRequest.cc | 2 ++
- 1 file changed, 2 insertions(+)
-
-diff --git a/src/librbd/operation/TrimRequest.cc b/src/librbd/operation/TrimRequest.cc
-index 28f2deb..929ca51 100644
---- a/src/librbd/operation/TrimRequest.cc
-+++ b/src/librbd/operation/TrimRequest.cc
-@@ -362,4 +362,6 @@ void TrimRequest<I>::send_finish(int r) {
- } // namespace operation
- } // namespace librbd
-
-+#ifndef TEST_F
- template class librbd::operation::TrimRequest<librbd::ImageCtx>;
-+#endif
---
-2.9.5
-
diff --git a/0004-cmake-modules-BuildBoost.cmake.patch b/0004-cmake-modules-BuildBoost.cmake.patch
deleted file mode 100644
index 644a6c4..0000000
--- a/0004-cmake-modules-BuildBoost.cmake.patch
+++ /dev/null
@@ -1,11 +0,0 @@
---- ceph-12.2.3/cmake/modules/BuildBoost.cmake.orig 2018-02-21 14:13:32.527899631 -0500
-+++ ceph-12.2.3/cmake/modules/BuildBoost.cmake 2018-02-21 14:14:04.796899631 -0500
-@@ -75,7 +75,7 @@
- set(install_command
- ${b2} install)
- set(boost_root_dir "${CMAKE_BINARY_DIR}/boost")
-- if(EXISTS "${PROJECT_SOURCE_DIR}/src/boost/libs/config/include/boost/config.hpp")
-+ if(EXISTS "${PROJECT_SOURCE_DIR}/src/boost/boost/config.hpp")
- message(STATUS "boost already in src")
- set(source_dir
- SOURCE_DIR "${PROJECT_SOURCE_DIR}/src/boost")
diff --git a/0005-src-rocksdb-table-block.h.patch b/0005-src-rocksdb-table-block.h.patch
deleted file mode 100644
index 8acb401..0000000
--- a/0005-src-rocksdb-table-block.h.patch
+++ /dev/null
@@ -1,155 +0,0 @@
---- ceph-12.2.3/src/rocksdb/table/block.h.orig 2018-02-22 07:49:38.044899631 -0500
-+++ ceph-12.2.3/src/rocksdb/table/block.h 2018-02-22 07:58:55.855899631 -0500
-@@ -65,7 +65,8 @@
-
- // Create bitmap and set all the bits to 0
- bitmap_ = new std::atomic<uint32_t>[bitmap_size];
-- memset(bitmap_, 0, bitmap_size * kBytesPersEntry);
-+ // memset(bitmap_, 0, bitmap_size * kBytesPersEntry);
-+ { unsigned i = 0; for (; i < bitmap_size;) bitmap_[i++] = 0; }
-
- RecordTick(GetStatistics(), READ_AMP_TOTAL_READ_BYTES,
- num_bits_needed << bytes_per_bit_pow_);
---- ceph-12.2.3/src/rocksdb/db/c.cc.orig 2018-02-22 08:14:56.033899631 -0500
-+++ ceph-12.2.3/src/rocksdb/db/c.cc 2018-02-22 10:06:39.759899631 -0500
-@@ -1322,11 +1322,6 @@
- b->rep.PutLogData(Slice(blob, len));
- }
-
--void rocksdb_writebatch_iterate(
-- rocksdb_writebatch_t* b,
-- void* state,
-- void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
-- void (*deleted)(void*, const char* k, size_t klen)) {
- class H : public WriteBatch::Handler {
- public:
- void* state_;
-@@ -1339,6 +1334,12 @@
- (*deleted_)(state_, key.data(), key.size());
- }
- };
-+
-+void rocksdb_writebatch_iterate(
-+ rocksdb_writebatch_t* b,
-+ void* state,
-+ void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
-+ void (*deleted)(void*, const char* k, size_t klen)) {
- H handler;
- handler.state_ = state;
- handler.put_ = put;
-@@ -1579,18 +1580,6 @@
- void* state,
- void (*put)(void*, const char* k, size_t klen, const char* v, size_t vlen),
- void (*deleted)(void*, const char* k, size_t klen)) {
-- class H : public WriteBatch::Handler {
-- public:
-- void* state_;
-- void (*put_)(void*, const char* k, size_t klen, const char* v, size_t vlen);
-- void (*deleted_)(void*, const char* k, size_t klen);
-- virtual void Put(const Slice& key, const Slice& value) override {
-- (*put_)(state_, key.data(), key.size(), value.data(), value.size());
-- }
-- virtual void Delete(const Slice& key) override {
-- (*deleted_)(state_, key.data(), key.size());
-- }
-- };
- H handler;
- handler.state_ = state;
- handler.put_ = put;
-@@ -2532,13 +2521,9 @@
- delete filter;
- }
-
--rocksdb_filterpolicy_t* rocksdb_filterpolicy_create_bloom_format(int bits_per_key, bool original_format) {
-- // Make a rocksdb_filterpolicy_t, but override all of its methods so
-- // they delegate to a NewBloomFilterPolicy() instead of user
-- // supplied C functions.
-- struct Wrapper : public rocksdb_filterpolicy_t {
-+ struct WrapperFP : public rocksdb_filterpolicy_t {
- const FilterPolicy* rep_;
-- ~Wrapper() { delete rep_; }
-+ ~WrapperFP() { delete rep_; }
- const char* Name() const override { return rep_->Name(); }
- void CreateFilter(const Slice* keys, int n,
- std::string* dst) const override {
-@@ -2549,11 +2534,16 @@
- }
- static void DoNothing(void*) { }
- };
-- Wrapper* wrapper = new Wrapper;
-+
-+rocksdb_filterpolicy_t* rocksdb_filterpolicy_create_bloom_format(int bits_per_key, bool original_format) {
-+ // Make a rocksdb_filterpolicy_t, but override all of its methods so
-+ // they delegate to a NewBloomFilterPolicy() instead of user
-+ // supplied C functions.
-+ WrapperFP* wrapper = new WrapperFP;
- wrapper->rep_ = NewBloomFilterPolicy(bits_per_key, original_format);
- wrapper->state_ = nullptr;
- wrapper->delete_filter_ = nullptr;
-- wrapper->destructor_ = &Wrapper::DoNothing;
-+ wrapper->destructor_ = &WrapperFP::DoNothing;
- return wrapper;
- }
-
-@@ -2889,10 +2879,9 @@
- delete st;
- }
-
--rocksdb_slicetransform_t* rocksdb_slicetransform_create_fixed_prefix(size_t prefixLen) {
-- struct Wrapper : public rocksdb_slicetransform_t {
-+ struct WrapperST : public rocksdb_slicetransform_t {
- const SliceTransform* rep_;
-- ~Wrapper() { delete rep_; }
-+ ~WrapperST() { delete rep_; }
- const char* Name() const override { return rep_->Name(); }
- Slice Transform(const Slice& src) const override {
- return rep_->Transform(src);
-@@ -2903,31 +2892,20 @@
- bool InRange(const Slice& src) const override { return rep_->InRange(src); }
- static void DoNothing(void*) { }
- };
-- Wrapper* wrapper = new Wrapper;
-+
-+rocksdb_slicetransform_t* rocksdb_slicetransform_create_fixed_prefix(size_t prefixLen) {
-+ WrapperST* wrapper = new WrapperST;
- wrapper->rep_ = rocksdb::NewFixedPrefixTransform(prefixLen);
- wrapper->state_ = nullptr;
-- wrapper->destructor_ = &Wrapper::DoNothing;
-+ wrapper->destructor_ = &WrapperST::DoNothing;
- return wrapper;
- }
-
- rocksdb_slicetransform_t* rocksdb_slicetransform_create_noop() {
-- struct Wrapper : public rocksdb_slicetransform_t {
-- const SliceTransform* rep_;
-- ~Wrapper() { delete rep_; }
-- const char* Name() const override { return rep_->Name(); }
-- Slice Transform(const Slice& src) const override {
-- return rep_->Transform(src);
-- }
-- bool InDomain(const Slice& src) const override {
-- return rep_->InDomain(src);
-- }
-- bool InRange(const Slice& src) const override { return rep_->InRange(src); }
-- static void DoNothing(void*) { }
-- };
-- Wrapper* wrapper = new Wrapper;
-+ WrapperST* wrapper = new WrapperST;
- wrapper->rep_ = rocksdb::NewNoopTransform();
- wrapper->state_ = nullptr;
-- wrapper->destructor_ = &Wrapper::DoNothing;
-+ wrapper->destructor_ = &WrapperST::DoNothing;
- return wrapper;
- }
-
---- ceph-12.2.3/src/rocksdb/memtable/inlineskiplist.h.orig 2018-02-22 10:34:06.918899631 -0500
-+++ ceph-12.2.3/src/rocksdb/memtable/inlineskiplist.h 2018-02-22 10:34:44.145899631 -0500
-@@ -279,7 +279,7 @@
- // next_[0]. This is used for passing data from AllocateKey to Insert.
- void StashHeight(const int height) {
- assert(sizeof(int) <= sizeof(next_[0]));
-- memcpy(&next_[0], &height, sizeof(int));
-+ memcpy(static_cast<void*>(&next_[0]), &height, sizeof(int));
- }
-
- // Retrieves the value passed to StashHeight. Undefined after a call
diff --git a/ceph.spec b/ceph.spec
index 9017056..1fa45d7 100644
--- a/ceph.spec
+++ b/ceph.spec
@@ -17,57 +17,56 @@
%global _hardened_build 1
%bcond_without ocf
-%ifnarch armv7hl
-%bcond_without cephfs_java
-%else
-%bcond_with cephfs_java
-%endif
-%if 0%{?suse_version}
-%bcond_with ceph_test_package
-%else
-%bcond_without ceph_test_package
-%endif
%bcond_with make_check
%ifarch s390 s390x
%bcond_with tcmalloc
%else
%bcond_without tcmalloc
%endif
-%ifnarch armv7hl
-%bcond_with lowmem_builder
-%else
-%if 0%{?rhel}
-%ifnarch ppc64le
-%bcond_with lowmem_builder
-%else
-%bcond_without lowmem_builder
-%endif
-%else
-%bcond_without lowmem_builder
-%endif
-%endif
%if 0%{?fedora} || 0%{?rhel}
%bcond_without selinux
+%bcond_without ceph_test_package
+%bcond_without cephfs_java
+%bcond_without lttng
%endif
%if 0%{?suse_version}
%bcond_with selinux
+%bcond_with ceph_test_package
+%bcond_with cephfs_java
+#Compat macro for new _fillupdir macro introduced in Nov 2017
+%if ! %{defined _fillupdir}
+%global _fillupdir /var/adm/fillup-templates
%endif
-
-# LTTng-UST enabled on Fedora, RHEL 6+, and SLE (not openSUSE)
-%if 0%{?fedora} || 0%{?rhel} >= 6 || 0%{?suse_version}
-%if ! 0%{?is_opensuse}
+%if 0%{?is_opensuse}
+%bcond_without lttng
+%else
+%ifarch x86_64 aarch64
%bcond_without lttng
+%else
+%bcond_with lttng
%endif
%endif
+%endif
+%if 0%{?suse_version} >= 1500
+%bcond_with python2
+%else
+%bcond_without python2
+%endif
+%if 0%{without python2}
+%global _defined_if_python2_absent 1
+%endif
%if %{with selinux}
# get selinux policy version
-%{!?_selinux_policy_version: %global _selinux_policy_version %(sed -e 's,.*selinux-policy-\\([^/]*\\)/.*,\\1,' /usr/share/selinux/devel/policyhelp 2>/dev/null || echo 0.0.0)}
+%{!?_selinux_policy_version: %global _selinux_policy_version 0.0.0}
%endif
%{!?_udevrulesdir: %global _udevrulesdir /lib/udev/rules.d}
%{!?tmpfiles_create: %global tmpfiles_create systemd-tmpfiles --create}
%{!?python3_pkgversion: %global python3_pkgversion 3}
+# define _python_buildid macro which will expand to the empty string when
+# building with python2
+%global _python_buildid %{?_defined_if_python2_absent:%{python3_pkgversion}}
# unify libexec for all targets
%global _libexecdir %{_exec_prefix}/lib
@@ -85,7 +84,7 @@
# main package definition
#################################################################################
Name: ceph
-Version: 12.2.5
+Version: 13.1.0
Release: 1%{?dist}
%if 0%{?fedora} || 0%{?rhel}
Epoch: 1
@@ -102,12 +101,14 @@ Group: System/Filesystems
URL: http://ceph.com/
Source0: http://download.ceph.com/tarballs/%{name}-%{version}.tar.gz
# https://bugzilla.redhat.com/show_bug.cgi?id=1474773
-Patch001: 0001-src-rocksdb-util-murmurhash.patch
+Patch001: 0001-9df56dc509a6cbb7c1da8073e82fc1d2a284418d.patch
+#Patch001: 0001-src-rocksdb-util-murmurhash.patch
# https://bugzilla.redhat.com/show_bug.cgi?id=1474774
-Patch002: 0002-cmake-Support-ppc64.patch
-Patch003: 0003-librbd-Conditionally-import-TrimRequest.cc.patch
-Patch005: 0005-src-rocksdb-table-block.h.patch
+#Patch002: 0002-cmake-Support-ppc64.patch
+#Patch003: 0003-librbd-Conditionally-import-TrimRequest.cc.patch
+#Patch005: 0005-src-rocksdb-table-block.h.patch
%if 0%{?suse_version}
+# _insert_obs_source_lines_here
%if 0%{?is_opensuse}
ExclusiveArch: x86_64 aarch64 ppc64 ppc64le
%else
@@ -117,10 +118,10 @@ ExclusiveArch: x86_64 aarch64 ppc64le s390x
#################################################################################
# dependencies that apply across all distro families
#################################################################################
-Requires: ceph-osd = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mds = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mgr = %{_epoch_prefix}%{version}-%{release}
-Requires: ceph-mon = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-osd = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mds = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mgr = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-mon = %{_epoch_prefix}%{version}-%{release}
Requires(post): binutils
%if 0%{with cephfs_java}
BuildRequires: java-devel
@@ -129,50 +130,47 @@ BuildRequires: sharutils
%if 0%{with selinux}
BuildRequires: checkpolicy
BuildRequires: selinux-policy-devel
-BuildRequires: selinux-policy-doc
-%endif
-%if 0%{with make_check}
-%if 0%{?fedora} || 0%{?rhel}
-BuildRequires: python-cherrypy
-BuildRequires: python-werkzeug
-%endif
-%if 0%{?suse_version}
-BuildRequires: python-CherryPy
-BuildRequires: python-Werkzeug
-BuildRequires: python-numpy-devel
-%endif
-BuildRequires: python-coverage
-BuildRequires: python-pecan
-BuildRequires: socat
%endif
BuildRequires: bc
BuildRequires: gperf
-BuildRequires: cmake
+BuildRequires: cmake
BuildRequires: cryptsetup
BuildRequires: fuse-devel
+%if 0%{?rhel} == 7
+# devtoolset offers newer make and valgrind-devel, but the old ones are good
+# enough.
+BuildRequires: devtoolset-7-gcc-c++
+%else
BuildRequires: gcc-c++
+%endif
BuildRequires: gdbm
%if 0%{with tcmalloc}
BuildRequires: gperftools-devel >= 2.4
%endif
-BuildRequires: jq
+BuildRequires: jq
BuildRequires: leveldb-devel > 1.2
BuildRequires: libaio-devel
BuildRequires: libblkid-devel >= 2.17
BuildRequires: libcurl-devel
BuildRequires: libudev-devel
+BuildRequires: liboath-devel
BuildRequires: libtool
BuildRequires: libxml2-devel
+BuildRequires: libuuid-devel
BuildRequires: make
BuildRequires: parted
BuildRequires: perl
BuildRequires: pkgconfig
-BuildRequires: python
-BuildRequires: python-devel
-BuildRequires: python-nose
-BuildRequires: python-requests
-BuildRequires: python-virtualenv
+BuildRequires: procps
+BuildRequires: python%{_python_buildid}
+BuildRequires: python%{_python_buildid}-devel
+BuildRequires: python%{_python_buildid}-nose
+BuildRequires: python%{_python_buildid}-requests
+BuildRequires: python%{_python_buildid}-virtualenv
BuildRequires: snappy-devel
+%if 0%{with make_check}
+BuildRequires: socat
+%endif
BuildRequires: udev
BuildRequires: util-linux
BuildRequires: valgrind-devel
@@ -186,27 +184,31 @@ BuildRequires: yasm
# distro-conditional dependencies
#################################################################################
%if 0%{?suse_version}
-BuildRequires: pkgconfig(systemd)
+BuildRequires: pkgconfig(systemd)
BuildRequires: systemd-rpm-macros
BuildRequires: systemd
%{?systemd_requires}
PreReq: %fillup_prereq
BuildRequires: net-tools
BuildRequires: libbz2-devel
-BuildRequires: btrfsprogs
+BuildRequires: btrfsprogs
BuildRequires: mozilla-nss-devel
BuildRequires: keyutils-devel
-BuildRequires: libopenssl-devel
-BuildRequires: lsb-release
-BuildRequires: openldap2-devel
-BuildRequires: python-Cython
-BuildRequires: python-PrettyTable
-BuildRequires: python-Sphinx
-BuildRequires: rdma-core-devel
+BuildRequires: libopenssl-devel
+BuildRequires: lsb-release
+BuildRequires: openldap2-devel
+BuildRequires: cunit-devel
+BuildRequires: python%{_python_buildid}-base
+BuildRequires: python%{_python_buildid}-Cython
+BuildRequires: python%{_python_buildid}-PrettyTable
+BuildRequires: python%{_python_buildid}-Sphinx
+BuildRequires: rdma-core-devel
+BuildRequires: liblz4-devel >= 1.7
+BuildRequires: rdma-core-devel
%endif
%if 0%{?fedora} || 0%{?rhel}
Requires: systemd
-BuildRequires: boost-random
+BuildRequires: boost-random
BuildRequires: btrfs-progs
BuildRequires: nss-devel
BuildRequires: keyutils-libs-devel
@@ -214,12 +216,14 @@ BuildRequires: keyutils-libs-devel
%ifnarch %{arm}
BuildRequires: rdma-core-devel
%endif
-BuildRequires: openldap-devel
-BuildRequires: openssl-devel
-BuildRequires: redhat-lsb-core
-BuildRequires: Cython
-BuildRequires: python-prettytable
-BuildRequires: python-sphinx
+BuildRequires: openldap-devel
+BuildRequires: openssl-devel
+BuildRequires: redhat-lsb-core
+BuildRequires: CUnit-devel
+BuildRequires: Cython%{_python_buildid}
+BuildRequires: python%{_python_buildid}-prettytable
+BuildRequires: python%{_python_buildid}-sphinx
+BuildRequires: lz4-devel >= 1.7
%endif
# python34-... for RHEL, python3-... for all other supported distros
%if ( 0%{?rhel} && 0%{?rhel} <= 7 )
@@ -231,6 +235,33 @@ BuildRequires: python3-devel
BuildRequires: python3-setuptools
BuildRequires: python3-Cython
%endif
+# distro-conditional make check dependencies
+%if 0%{with make_check}
+%if 0%{?fedora} || 0%{?rhel}
+BuildRequires: python%{_python_buildid}-cherrypy
+BuildRequires: python%{_python_buildid}-routes
+BuildRequires: python%{_python_buildid}-pecan
+BuildRequires: python%{_python_buildid}-werkzeug
+BuildRequires: python%{_python_buildid}-tox
+BuildRequires: python%{_python_buildid}-coverage
+%if 0%{?fedora}
+BuildRequires: python%{_python_buildid}-bcrypt
+%endif
+%if 0%{?rhel}
+BuildRequires: py-bcrypt
+%endif
+%endif
+%if 0%{?suse_version}
+BuildRequires: python%{_python_buildid}-CherryPy
+BuildRequires: python%{_python_buildid}-Routes
+BuildRequires: python%{_python_buildid}-Werkzeug
+BuildRequires: python%{_python_buildid}-pecan
+BuildRequires: python%{_python_buildid}-numpy-devel
+BuildRequires: python%{_python_buildid}-bcrypt
+BuildRequires: python%{_python_buildid}-tox
+BuildRequires: python%{_python_buildid}-coverage
+%endif
+%endif
# lttng and babeltrace for rbd-replay-prep
%if %{with lttng}
%if 0%{?fedora} || 0%{?rhel}
@@ -239,7 +270,7 @@ BuildRequires: libbabeltrace-devel
%endif
%if 0%{?suse_version}
BuildRequires: lttng-ust-devel
-BuildRequires: babeltrace-devel
+BuildRequires: babeltrace-devel
%endif
%endif
%if 0%{?suse_version}
@@ -250,7 +281,7 @@ BuildRequires: expat-devel
%endif
#hardened-cc1
%if 0%{?fedora} || 0%{?rhel}
-BuildRequires: redhat-rpm-config
+BuildRequires: redhat-rpm-config
%endif
%description
@@ -262,33 +293,39 @@ on commodity hardware and delivers object, block and file system storage.
# subpackages
#################################################################################
%package base
-Summary: Ceph Base Package
+Summary: Ceph Base Package
%if 0%{?suse_version}
-Group: System/Filesystems
+Group: System/Filesystems
%endif
-Requires: ceph-common = %{_epoch_prefix}%{version}-%{release}
-Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
-Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
-Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
-Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-common = %{_epoch_prefix}%{version}-%{release}
+Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
+Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
+Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
+Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
%if 0%{with selinux}
-Requires: ceph-selinux = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-selinux = %{_epoch_prefix}%{version}-%{release}
%endif
Requires(post):/sbin/ldconfig
Requires(postun):/sbin/ldconfig
-Requires: python
-Requires: python-requests
-Requires: python-setuptools
-Requires: grep
-Requires: xfsprogs
-Requires: logrotate
-Requires: util-linux
-Requires: cryptsetup
-Requires: findutils
-Requires: psmisc
-Requires: which
-%if 0%{?suse_version}
-Recommends: ntp-daemon
+Requires: cryptsetup
+Requires: e2fsprogs
+Requires: findutils
+Requires: grep
+Requires: logrotate
+Requires: parted
+Requires: psmisc
+Requires: python%{_python_buildid}-requests
+Requires: python%{_python_buildid}-setuptools
+Requires: util-linux
+Requires: xfsprogs
+Requires: which
+%if 0%{?fedora} || 0%{?rhel}
+Requires: gdisk
+%endif
+%if 0%{?suse_version}
+Recommends: ntp-daemon
+Recommends: chrony
+Requires: gptfdisk
%endif
%description base
Base is the package that includes all the files shared amongst ceph servers
@@ -301,17 +338,18 @@ Group: System/Filesystems
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
-Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
-Requires: python-rbd = %{_epoch_prefix}%{version}-%{release}
-Requires: python-cephfs = %{_epoch_prefix}%{version}-%{release}
-Requires: python-rgw = %{_epoch_prefix}%{version}-%{release}
+Requires: python%{_python_buildid}-rados = %{_epoch_prefix}%{version}-%{release}
+Requires: python%{_python_buildid}-rbd = %{_epoch_prefix}%{version}-%{release}
+Requires: python%{_python_buildid}-cephfs = %{_epoch_prefix}%{version}-%{release}
+Requires: python%{_python_buildid}-rgw = %{_epoch_prefix}%{version}-%{release}
%if 0%{?fedora} || 0%{?rhel}
-Requires: python-prettytable
+Requires: python%{_python_buildid}-prettytable
+Requires: python%{_python_buildid}-requests
%endif
%if 0%{?suse_version}
-Requires: python-PrettyTable
+Requires: python%{_python_buildid}-PrettyTable
+Requires: python%{_python_buildid}-requests
%endif
-Requires: python-requests
%{?systemd_requires}
%if 0%{?suse_version}
Requires(pre): pwdutils
@@ -337,13 +375,6 @@ Summary: Ceph Monitor Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
-# For ceph-rest-api
-%if 0%{?fedora} || 0%{?rhel}
-Requires: python-flask
-%endif
-%if 0%{?suse_version}
-Requires: python-Flask
-%endif
%description mon
ceph-mon is the cluster monitor daemon for the Ceph distributed file
system. One or more instances of ceph-mon form a Paxos part-time
@@ -351,24 +382,29 @@ parliament cluster that provides extremely reliable and durable storage
of cluster membership, configuration, and state.
%package mgr
-Summary: Ceph Manager Daemon
+Summary: Ceph Manager Daemon
%if 0%{?suse_version}
-Group: System/Filesystems
+Group: System/Filesystems
%endif
-Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
+Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
%if 0%{?fedora} || 0%{?rhel}
-Requires: python-cherrypy
-Requires: python-jinja2
-Requires: python-werkzeug
-Requires: pyOpenSSL
+Requires: python%{_python_buildid}-cherrypy
+Requires: python%{_python_buildid}-routes
+Requires: python%{_python_buildid}-jinja2
+Requires: python%{_python_buildid}-pecan
+Requires: python%{_python_buildid}-werkzeug
+Requires: pyOpenSSL%{_python_buildid}
+%endif
+%if 0%{?suse_version}
+Requires: python%{_python_buildid}-CherryPy
+Requires: python%{_python_buildid}-Routes
+Requires: python%{_python_buildid}-Jinja2
+Requires: python%{_python_buildid}-Werkzeug
+Requires: python%{_python_buildid}-pecan
+Requires: python%{_python_buildid}-pyOpenSSL
+Requires: python%{_python_buildid}-bcrypt
+Recommends: python%{_python_buildid}-influxdb
%endif
-%if 0%{?suse_version}
-Requires: python-CherryPy
-Requires: python-jinja2
-Requires: python-Werkzeug
-Requires: python-pyOpenSSL
-%endif
-Requires: python-pecan
%description mgr
ceph-mgr enables python modules that provide services (such as the REST
module derived from Calamari) and expose CLI hooks. ceph-mgr gathers
@@ -455,15 +491,7 @@ Summary: Ceph Object Storage Daemon
Group: System/Filesystems
%endif
Requires: ceph-base = %{_epoch_prefix}%{version}-%{release}
-# for sgdisk, used by ceph-disk
-%if 0%{?fedora} || 0%{?rhel}
-Requires: gdisk
-%endif
-%if 0%{?suse_version}
-Requires: gptfdisk
-%endif
-Requires: parted
-Requires: lvm2
+Requires: lvm2
%description osd
ceph-osd is the object storage daemon for the Ceph distributed file
system. It is responsible for storing objects on a local file system
@@ -520,6 +548,7 @@ Obsoletes: librgw2-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use RADOS gateway client library.
+%if 0%{with python2}
%package -n python-rgw
Summary: Python 2 libraries for the RADOS gateway
%if 0%{?suse_version}
@@ -531,11 +560,12 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rgw
This package contains Python 2 libraries for interacting with Cephs RADOS
gateway.
+%endif
%package -n python%{python3_pkgversion}-rgw
Summary: Python 3 libraries for the RADOS gateway
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: librgw2 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
@@ -543,6 +573,7 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
This package contains Python 3 libraries for interacting with Cephs RADOS
gateway.
+%if 0%{with python2}
%package -n python-rados
Summary: Python 2 libraries for the RADOS object store
%if 0%{?suse_version}
@@ -553,11 +584,12 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rados
This package contains Python 2 libraries for interacting with Cephs RADOS
object store.
+%endif
%package -n python%{python3_pkgversion}-rados
Summary: Python 3 libraries for the RADOS object store
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: python%{python3_pkgversion}
Requires: librados2 = %{_epoch_prefix}%{version}-%{release}
@@ -623,10 +655,11 @@ Obsoletes: librbd1-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use RADOS block device.
+%if 0%{with python2}
%package -n python-rbd
Summary: Python 2 libraries for the RADOS block device
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
@@ -634,11 +667,12 @@ Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-rbd
This package contains Python 2 libraries for interacting with Cephs RADOS
block device.
+%endif
%package -n python%{python3_pkgversion}-rbd
Summary: Python 3 libraries for the RADOS block device
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: librbd1 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
@@ -677,24 +711,26 @@ Obsoletes: libcephfs2-devel < %{_epoch_prefix}%{version}-%{release}
This package contains libraries and headers needed to develop programs
that use Cephs distributed file system.
+%if 0%{with python2}
%package -n python-cephfs
Summary: Python 2 libraries for Ceph distributed file system
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
%if 0%{?suse_version}
-Recommends: python-rados = %{_epoch_prefix}%{version}-%{release}
+Recommends: python-rados = %{_epoch_prefix}%{version}-%{release}
%endif
Obsoletes: python-ceph < %{_epoch_prefix}%{version}-%{release}
%description -n python-cephfs
This package contains Python 2 libraries for interacting with Cephs distributed
file system.
+%endif
%package -n python%{python3_pkgversion}-cephfs
Summary: Python 3 libraries for Ceph distributed file system
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Requires: libcephfs2 = %{_epoch_prefix}%{version}-%{release}
Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{release}
@@ -702,16 +738,18 @@ Requires: python%{python3_pkgversion}-rados = %{_epoch_prefix}%{version}-%{relea
This package contains Python 3 libraries for interacting with Cephs distributed
file system.
+%if 0%{with python2}
%package -n python%{python3_pkgversion}-ceph-argparse
Summary: Python 3 utility libraries for Ceph CLI
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
%description -n python%{python3_pkgversion}-ceph-argparse
This package contains types and routines for Python 3 used by the Ceph CLI as
well as the RESTful interface. These have to do with querying the daemons for
command-description information, validating user command input against those
descriptions, and submitting the command to the appropriate daemon.
+%endif
%if 0%{with ceph_test_package}
%package -n ceph-test
@@ -761,17 +799,18 @@ Group: System/Libraries
%endif
Requires: java
Requires: libcephfs_jni1 = %{_epoch_prefix}%{version}-%{release}
-Requires: junit
-BuildRequires: junit
+BuildRequires: junit
%description -n cephfs-java
This package contains the Java libraries for the Ceph File System.
%endif
%package -n rados-objclass-devel
-Summary: RADOS object class development kit
-Group: Development/Libraries
-Requires: librados2-devel = %{_epoch_prefix}%{version}-%{release}
+Summary: RADOS object class development kit
+%if 0%{?suse_version}
+Group: Development/Libraries/C and C++
+%endif
+Requires: librados2-devel = %{_epoch_prefix}%{version}-%{release}
%description -n rados-objclass-devel
This package contains libraries and headers needed to develop RADOS object
class plugins.
@@ -795,10 +834,11 @@ populated file-systems.
%endif
+%if 0%{with python2}
%package -n python-ceph-compat
Summary: Compatibility package for Cephs python libraries
%if 0%{?suse_version}
-Group: Development/Languages/Python
+Group: Development/Libraries/Python
%endif
Obsoletes: python-ceph
Requires: python-rados = %{_epoch_prefix}%{version}-%{release}
@@ -811,6 +851,7 @@ This is a compatibility package to accommodate python-ceph split into
python-rados, python-rbd, python-rgw and python-cephfs. Packages still
depending on python-ceph should be fixed to depend on python-rados,
python-rbd, python-rgw or python-cephfs instead.
+%endif
#################################################################################
# common
@@ -819,6 +860,11 @@ python-rbd, python-rgw or python-cephfs instead.
%autosetup -p1 -n %{name}-%{version}
%build
+
+%if 0%{?rhel} == 7
+. /opt/rh/devtoolset-7/enable
+%endif
+
%if 0%{with cephfs_java}
# Find jni.h
for i in /usr/{lib64,lib}/jvm/java/include{,/linux}; do
@@ -826,36 +872,39 @@ for i in /usr/{lib64,lib}/jvm/java/include{,/linux}; do
done
%endif
-%if %{with lowmem_builder}
+%if 0%{?suse_version}
+# the following setting fixed an OOM condition we once encountered in the OBS
RPM_OPT_FLAGS="$RPM_OPT_FLAGS --param ggc-min-expand=20 --param ggc-min-heapsize=32768"
%endif
-%ifnarch armv7hl
-export RPM_OPT_FLAGS=`echo $RPM_OPT_FLAGS | sed -e 's/i386/i486/'`
-%else
-export RPM_OPT_FLAGS=`echo $RPM_OPT_FLAGS | sed -e 's/i386/i486/' -e 's/-pipe//g'`
-%endif
export CPPFLAGS="$java_inc"
export CFLAGS="$RPM_OPT_FLAGS"
export CXXFLAGS="$RPM_OPT_FLAGS"
export LDFLAGS="$RPM_LD_FLAGS"
-env | sort
-
-%if %{with lowmem_builder}
-%ifnarch armv7hl
-%if 0%{?jobs} > 8
-%define _smp_mflags -j8
-%endif
-%else
-%define _smp_mflags -j1
-%endif
-%endif
+# Parallel build settings ...
+CEPH_MFLAGS_JOBS="%{?_smp_mflags}"
+CEPH_SMP_NCPUS=$(echo "$CEPH_MFLAGS_JOBS" | sed 's/-j//')
+%if 0%{?__isa_bits} == 32
+# 32-bit builds can use 3G memory max, which is not enough even for -j2
+CEPH_SMP_NCPUS="1"
+%endif
+# do not eat all memory
+echo "Available memory:"
+free -h
+echo "System limits:"
+ulimit -a
+if test -n "$CEPH_SMP_NCPUS" -a "$CEPH_SMP_NCPUS" -gt 1 ; then
+ mem_per_process=1800
+ max_mem=$(LANG=C free -m | sed -n "s|^Mem: *\([0-9]*\).*$|\1|p")
+ max_jobs="$(($max_mem / $mem_per_process))"
+ test "$CEPH_SMP_NCPUS" -gt "$max_jobs" && CEPH_SMP_NCPUS="$max_jobs" && echo "Warning: Reducing build parallelism to -j$max_jobs because of memory limits"
+ test "$CEPH_SMP_NCPUS" -le 0 && CEPH_SMP_NCPUS="1" && echo "Warning: Not using parallel build at all because of memory limits"
+fi
+export CEPH_SMP_NCPUS
+export CEPH_MFLAGS_JOBS="-j$CEPH_SMP_NCPUS"
-# unlimit _smp_mflags in system macro if not set above
-%define _smp_ncpus_max 0
-# extract the number of processors for use with cmake
-%define _smp_ncpus %(echo %{_smp_mflags} | sed 's/-j//')
+env | sort
mkdir build
cd build
@@ -871,7 +920,13 @@ cmake .. \
-DWITH_EMBEDDED=OFF \
-DWITH_MANPAGE=ON \
-DWITH_PYTHON3=ON \
- -DWITH_SYSTEMD=ON \
+ -DWITH_MGR_DASHBOARD_FRONTEND=OFF \
+%if %{with python2}
+ -DWITH_PYTHON2=ON \
+%else
+ -DWITH_PYTHON2=OFF \
+ -DMGR_PYTHON_VERSION=3 \
+%endif
%if ( ( 0%{?rhel} && 0%{?rhel} <= 7) && ! 0%{?centos} )
-DWITH_SUBMAN=ON \
%endif
@@ -903,21 +958,18 @@ cmake .. \
%ifnarch %{arm}
-DWITH_RDMA=OFF \
%endif
- -DBOOST_J=%{_smp_ncpus}
-
-make %{?_smp_mflags}
+ -DBOOST_J=$CEPH_SMP_NCPUS
+make "$CEPH_MFLAGS_JOBS"
%if 0%{with make_check}
%check
# run in-tree unittests
cd build
-ctest %{?_smp_mflags}
-
+ctest "$CEPH_MFLAGS_JOBS"
%endif
-
%install
pushd build
make DESTDIR=%{buildroot} install
@@ -929,10 +981,9 @@ install -m 0644 -D src/etc-rbdmap %{buildroot}%{_sysconfdir}/ceph/rbdmap
install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_sysconfdir}/sysconfig/ceph
%endif
%if 0%{?suse_version}
-install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_localstatedir}/adm/fillup-templates/sysconfig.%{name}
+install -m 0644 -D etc/sysconfig/ceph %{buildroot}%{_fillupdir}/sysconfig.%{name}
%endif
install -m 0644 -D systemd/ceph.tmpfiles.d %{buildroot}%{_tmpfilesdir}/ceph-common.conf
-install -m 0755 -D systemd/ceph %{buildroot}%{_sbindir}/rcceph
install -m 0644 -D systemd/50-ceph.preset %{buildroot}%{_libexecdir}/systemd/system-preset/50-ceph.preset
mkdir -p %{buildroot}%{_sbindir}
install -m 0644 -D src/logrotate.conf %{buildroot}%{_sysconfdir}/logrotate.d/ceph
@@ -989,7 +1040,6 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%{_libexecdir}/systemd/system-preset/50-ceph.preset
%{_sbindir}/ceph-create-keys
%{_sbindir}/ceph-disk
-%{_sbindir}/rcceph
%dir %{_libexecdir}/ceph
%{_libexecdir}/ceph/ceph_common.sh
%dir %{_libdir}/rados-classes
@@ -1012,17 +1062,28 @@ mkdir -p %{buildroot}%{_localstatedir}/lib/ceph/bootstrap-rbd
%config(noreplace) %{_sysconfdir}/sysconfig/ceph
%endif
%if 0%{?suse_version}
-%{_localstatedir}/adm/fillup-templates/sysconfig.*
+%{_fillupdir}/sysconfig.*
%config %{_sysconfdir}/sysconfig/SuSEfirewall2.d/services/ceph-mon
%config %{_sysconfdir}/sysconfig/SuSEfirewall2.d/services/ceph-osd-mds
%endif
%{_unitdir}/ceph-disk@.service
%{_unitdir}/ceph.target
+%if 0%{with python2}
%{python_sitelib}/ceph_detect_init*
%{python_sitelib}/ceph_disk*
+%else
+%{python3_sitelib}/ceph_detect_init*
+%{python3_sitelib}/ceph_disk*
+%endif
+%if 0%{with python2}
%dir %{python_sitelib}/ceph_volume
%{python_sitelib}/ceph_volume/*
%{python_sitelib}/ceph_volume-*
+%else
+%dir %{python3_sitelib}/ceph_volume
+%{python3_sitelib}/ceph_volume/*
+%{python3_sitelib}/ceph_volume-*
+%endif
%{_mandir}/man8/ceph-deploy.8*
%{_mandir}/man8/ceph-detect-init.8*
%{_mandir}/man8/ceph-create-keys.8*
@@ -1093,7 +1154,6 @@ fi
%{_bindir}/ceph-dencoder
%{_bindir}/ceph-rbdnamer
%{_bindir}/ceph-syn
-%{_bindir}/ceph-crush-location
%{_bindir}/cephfs-data-scan
%{_bindir}/cephfs-journal-tool
%{_bindir}/cephfs-table-tool
@@ -1139,8 +1199,15 @@ fi
%config %{_sysconfdir}/bash_completion.d/radosgw-admin
%config(noreplace) %{_sysconfdir}/ceph/rbdmap
%{_unitdir}/rbdmap.service
+%if 0%{with python2}
%{python_sitelib}/ceph_argparse.py*
%{python_sitelib}/ceph_daemon.py*
+%else
+%{python3_sitelib}/ceph_argparse.py
+%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
+%{python3_sitelib}/ceph_daemon.py
+%{python3_sitelib}/__pycache__/ceph_daemon.cpython*.py*
+%endif
%dir %{_udevrulesdir}
%{_udevrulesdir}/50-rbd.rules
%attr(3770,ceph,ceph) %dir %{_localstatedir}/log/ceph/
@@ -1282,11 +1349,8 @@ fi
%files mon
%{_bindir}/ceph-mon
-%{_bindir}/ceph-rest-api
%{_bindir}/ceph-monstore-tool
%{_mandir}/man8/ceph-mon.8*
-%{_mandir}/man8/ceph-rest-api.8*
-%{python_sitelib}/ceph_rest_api.py*
%{_unitdir}/ceph-mon@.service
%{_unitdir}/ceph-mon.target
%attr(750,ceph,ceph) %dir %{_localstatedir}/lib/ceph/mon
@@ -1559,9 +1623,11 @@ fi
%{_bindir}/librados-config
%{_mandir}/man8/librados-config.8*
+%if 0%{with python2}
%files -n python-rados
%{python_sitearch}/rados.so
%{python_sitearch}/rados-*.egg-info
+%endif
%files -n python%{python3_pkgversion}-rados
%{python3_sitearch}/rados.cpython*.so
@@ -1597,6 +1663,10 @@ fi
%ldconfig_scriptlets -n librgw2
%files -n librgw2
%{_libdir}/librgw.so.*
+%if %{with lttng}
+%{_libdir}/librgw_op_tp.so*
+%{_libdir}/librgw_rados_tp.so*
+%endif
%files -n librgw-devel
%dir %{_includedir}/rados
@@ -1604,17 +1674,21 @@ fi
%{_includedir}/rados/rgw_file.h
%{_libdir}/librgw.so
+%if 0%{with python2}
%files -n python-rgw
%{python_sitearch}/rgw.so
%{python_sitearch}/rgw-*.egg-info
+%endif
%files -n python%{python3_pkgversion}-rgw
%{python3_sitearch}/rgw.cpython*.so
%{python3_sitearch}/rgw-*.egg-info
+%if 0%{with python2}
%files -n python-rbd
%{python_sitearch}/rbd.so
%{python_sitearch}/rbd-*.egg-info
+%endif
%files -n python%{python3_pkgversion}-rbd
%{python3_sitearch}/rbd.cpython*.so
@@ -1630,10 +1704,12 @@ fi
%{_includedir}/cephfs/ceph_statx.h
%{_libdir}/libcephfs.so
+%if 0%{with python2}
%files -n python-cephfs
%{python_sitearch}/cephfs.so
%{python_sitearch}/cephfs-*.egg-info
%{python_sitelib}/ceph_volume_client.py*
+%endif
%files -n python%{python3_pkgversion}-cephfs
%{python3_sitearch}/cephfs.cpython*.so
@@ -1641,11 +1717,13 @@ fi
%{python3_sitelib}/ceph_volume_client.py
%{python3_sitelib}/__pycache__/ceph_volume_client.cpython*.py*
+%if 0%{with python2}
%files -n python%{python3_pkgversion}-ceph-argparse
%{python3_sitelib}/ceph_argparse.py
%{python3_sitelib}/__pycache__/ceph_argparse.cpython*.py*
%{python3_sitelib}/ceph_daemon.py
%{python3_sitelib}/__pycache__/ceph_daemon.cpython*.py*
+%endif
%if 0%{with ceph_test_package}
%files -n ceph-test
@@ -1667,13 +1745,7 @@ fi
%{_bindir}/ceph_rgw_multiparser
%{_bindir}/ceph_scratchtool
%{_bindir}/ceph_scratchtoolpp
-%{_bindir}/ceph_smalliobench
-%{_bindir}/ceph_smalliobenchdumb
-%{_bindir}/ceph_smalliobenchfs
-%{_bindir}/ceph_smalliobenchrbd
%{_bindir}/ceph_test_*
-%{_bindir}/ceph_tpbench
-%{_bindir}/ceph_xattr_bench
%{_bindir}/ceph-coverage
%{_bindir}/ceph-debugpack
%{_mandir}/man8/ceph-debugpack.8*
@@ -1790,12 +1862,17 @@ exit 0
%endif # with selinux
+%if 0%{with python2}
%files -n python-ceph-compat
# We need an empty %%files list for python-ceph-compat, to tell rpmbuild to
# actually build this meta package.
+%endif
%changelog
+* Tue May 8 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:13.1.0-1
+- New release (1:13.1.0-1)
+
* Fri Apr 27 2018 Kaleb S. KEITHLEY <kkeithle[at]redhat.com> - 1:12.2.5-1
- New release (1:12.2.5-1)
diff --git a/sources b/sources
index 021832c..e0ac550 100644
--- a/sources
+++ b/sources
@@ -1 +1 @@
-SHA512 (ceph-12.2.5.tar.gz) = bd78358590eb077afcc3c1e12864fadee906c95ac8afee25bdca6155f99b75f89d923159fcc13430abba21a89fee3e566177ff5f9ffabdd3b3f220d05602ab86
+SHA512 (ceph-13.1.0.tar.gz) = d4feb0b0bac5e30232eddcc3f69c1d939f0b5268491acecba9a58dcadbbdaab6367f415ede924ca84de9a89a686410613cdc631f8aff15aec0eb4531138eec7a
1
0