[Bug 631763] Review Request: zif - Simple wrapper for rpm

bugzilla at redhat.com bugzilla at redhat.com
Fri Oct 1 17:20:15 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=631763

--- Comment #10 from Michal Schmidt <mschmidt at redhat.com> 2010-10-01 13:20:15 EDT ---
> $ rpmlint {SRPMS,RPMS/x86_64}/zif*-3.*rpm
> zif.src: I: enchant-dictionary-not-found en_US
> zif.src: W: non-standard-group Unspecified
> zif.src: W: no-cleaning-of-buildroot %clean
> zif.src: W: no-buildroot-tag
> zif.src: W: no-%clean-section
> zif.x86_64: W: non-standard-group Unspecified

All OK with current rpm.

> zif.x86_64: W: shared-lib-calls-exit /usr/lib64/libzif.so.1.0.1 exit at GLIBC_2.2.5

OK. exit() is called only from egg_error_real() which is never used in the
library.
BTW, a clever use of unsafe linker options might even elimitate the function.

> zif.x86_64: W: manual-page-warning /usr/share/man/man1/zif.1.gz 1: warning: macro `\"' not defined
> zif-devel.x86_64: W: non-standard-group Unspecified
> 4 packages and 0 specfiles checked; 0 errors, 8 warnings.

Can be ignored.

Formal review
-----------------------------
[OK] ... guideline matched
[--] ... irrelevant guideline
[??] ... needs discussion
[!!] ... needs fixing
-----------------------------

[OK] rpmlint posted
[OK] Naming Guidelines
[OK] spec file name matches package name
[..] Packaging Guidelines:
  [OK] naming
  [OK] version and release
  [OK] legal
  [OK] no pre-built binaries
  [OK] spec legibility
  [OK] architecture support
  [OK] fs layout
  [OK] rpmlint
  [OK] changelog
  [OK] tags
  [OK] BuildRoot tag (not needed)
  [OK] %clean (not needed)
  [??] Requires
    I wonder what is special about sqlite-devel that it is required
    explicitly by zif-devel, but libarchive-devel is left to be pulled
    automatically via rpm's pkgconfig dependency extraction.
  [OK] BuildRequires
  [OK] Summary and description
  [OK] encoding
  [OK] documentation
  [OK] compiler flags
  [OK] debuginfo
  [OK] devel package
  [OK] requiring base package
  [OK] shared libraries
  [--] static libraries
  [OK] no duplication of system libraries
  [OK] no rpaths
  [--] config files
  [--] initscripts
  [--] desktop files
  [!!] macros
     %{_mandir} should be used instead of %{_datadir}/man/
  [--] %global preferred over %define
  [--] handling of locale files
  [OK] timestamps
  [OK] parallel make
  [OK] scriptlets
  [--] conditional deps
  [OK] not relocatable
  [OK] code vs content
  [!!] file and dir ownership
    %{_datadir}/gtk-doc/html is owned neither by zif-devel nor any
    Required package. Should Require gtk-doc?
  [OK] no duplicate files
  [OK] file permissions
  [--] users and groups
  [--] web apps
  [OK] no conflicts
  [OK] no kernel modules
  [OK] nothing under /srv
  [OK] no bundling
  [--] patches should have upstream bug link or comment
  [--] use of epochs
  [OK] symlinks
  [OK] man pages
  [--] application-specific guidelines
[OK] approved license
[OK] GPLv2+ is correct
[OK] COPYING included in %doc
[OK] American English
[OK] legible spec
[OK] source matches upstream, sha256sum:
  aefac3cce4310e942bf735ea259efc45ba32fbe1c7cb6461e7f74637fd5df2d5
[OK] builds successfully
[--] ExcludeArch
[OK] BuildRequires
[--] locales
[OK] ldconfig called
[OK] no bundling
[OK] not relocatable
[!!] owning of directories, already noted above
[OK] no listing files more than once
[OK] permissions, %defattr usage
[OK] consistent macro usage
[OK] code or permissible content
[--] large docs
[OK] docs not necessary for runtime
[OK] headers in -devel
[--] static libs in -static
[OK] *.so in -devel
[OK] -devel requires base with a fully versioned dep
[OK] no *.la archives
[--] *.desktop for GUI apps
[OK] no owning of directories already owned by other packages
[OK] valid UTF-8 filenames

-- 
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