[Bug 551911] Review Request: monodevelop-boo - A boo plugin for monodevelop

bugzilla at redhat.com bugzilla at redhat.com
Wed Feb 17 21:38:54 UTC 2010


Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.


https://bugzilla.redhat.com/show_bug.cgi?id=551911

--- Comment #9 from Christian Krause <chkr at plauener.de> 2010-02-17 16:38:49 EST ---
I've looked at the new package:

1. the lang problem is still there:
using rpmlint on the binary rpm reveals lots of the following messages:

monodevelop-boo.i686: W: file-not-in-%lang
/usr/lib/monodevelop/AddIns/BooBinding/locale/ca/LC_MESSAGES/monodevelop-boo.mo
monodevelop-boo.i686: W: file-not-in-%lang
/usr/lib/monodevelop/AddIns/BooBinding/locale/cs/LC_MESSAGES/monodevelop-boo.mo

The generated %{name}.lang file is empty.

The root cause is, that the sed command does not fully work. Actually the
following changes are necessary:

--- monodevelop-boo.spec.2      2010-02-13 15:32:32.000000000 +0100
+++ SPECS/monodevelop-boo.spec  2010-02-17 22:11:07.000000000 +0100
@@ -50,8 +50,7 @@

 find %{buildroot} -type f -o -type l|sed '
 s:'"%{buildroot}"'::
-s:\(.*/"%{_lib}"/monodevelop/AddIns/Monodevelop.Boo/locale/\)\([^/_]\+\)\(.*\.mo$\):%lang(\2)
\1\2\3:
-s:\(.*/"%{_lib}"/monodevelop/AddIns/Monodevelop.Boo/locale/\)\([^/_]\+\)\(.*\.mo$\):%lang(\2)
\1\2\3:
+s:\(.*/%{_lib}/monodevelop/AddIns/BooBinding/locale/\)\([^/]\+\)\(.*\.mo$\):%lang(\2)
\1\2\3:
 s:^\([^%].*\)::
 s:%lang(C) ::
 /^$/d' > %{name}.lang
@@ -61,7 +60,12 @@

 %files -f %{name}.lang
 %defattr(-,root,root,-)
-%{_libdir}/monodevelop/AddIns/BooBinding
+%dir %{_libdir}/monodevelop/AddIns/BooBinding
+%{_libdir}/monodevelop/AddIns/BooBinding/*.dll
+%{_libdir}/monodevelop/AddIns/BooBinding/*.mdb
+%dir %{_libdir}/monodevelop/AddIns/BooBinding/locale
+%dir %{_libdir}/monodevelop/AddIns/BooBinding/locale/*
+%dir %{_libdir}/monodevelop/AddIns/BooBinding/locale/*/LC_MESSAGES

- my example used two lines with the substitution rules because there were two
base directories with language files (that's not the case here)
- the double-quotes around %{_lib} were still in the sed rule (since the whole
block is surrounded by '')
- the \([^/_]\+\) did not work since the boo binding uses language IDs like
"de_DE" (which anki didn't)
- to ensure that all directories are correctly packaged and no files are
packaged twice there ware some changes in the %files section needed as well

- I have positively tested that this fixes the language problem I have
described in comment #3.

2. even after these changes there are the following rpmlint warnings:
rpmlint RPMS/i686/monodevelop-boo-*-4.fc* SPECS/monodevelop-boo.spec
SRPMS/monodevelop-boo-2.2-4.fc13.src.rpm 
monodevelop-boo.i686: E: no-binary
monodevelop-boo.i686: W: only-non-binary-in-usr-lib
monodevelop-boo.i686: W: no-documentation
monodevelop-boo-devel.i686: W: spelling-error Summary(en_US) Addin -> Addie,
Adding, Admin
monodevelop-boo-devel.i686: W: spelling-error %description -l en_US Addin ->
Addie, Adding, Admin
monodevelop-boo-devel.i686: W: no-documentation
SPECS/monodevelop-boo.spec:42: W: configure-without-libdir-spec
SPECS/monodevelop-boo.spec:12: W: mixed-use-of-spaces-and-tabs (spaces: line
12, tab: line 4)
monodevelop-boo.src:42: W: configure-without-libdir-spec
monodevelop-boo.src:12: W: mixed-use-of-spaces-and-tabs (spaces: line 12, tab:
line 4)
3 packages and 1 specfiles checked; 1 errors, 9 warnings.

Sure, "no-binary", "only-non-binary-in-usr-lib", "no-documentation" and
"cnofigure-without-libdir-spec" are false positives, but the others should be
fixed.

I know, this may sound like nit-picking, but that's the way how I was taught
doing reviews (and accepting when my packages were reviewed). Since it is
usually quite hard to distinguish between "minor" and "major" issues revealed
by rpmlint if there are lots of warnings, I think we should eliminate as many
as possible (IMHO is this similiar to compiler warnings) in the first place...
;-)

-- 
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.



More information about the package-review mailing list