Bug#622888: marked as done (lintian: multi-arch libraries with arch-dependent overrides)
Your message dated Tue, 10 May 2011 22:25:31 +0000
with message-id <E1QJvMx-0002oX-Rh@franck.debian.org>
and subject line Bug#622888: fixed in lintian 2.5.0
has caused the Debian Bug report #622888,
regarding lintian: multi-arch libraries with arch-dependent overrides
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)
--
622888: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=622888
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message ---
- To: Debian Bug Tracking System <submit@bugs.debian.org>
- Subject: lintian: multi-arch libraries with arch-dependent overrides
- From: Andreas Beckmann <debian@abeckmann.de>
- Date: Fri, 15 Apr 2011 17:34:13 +0200
- Message-id: <20110415153413.25307.54482.reportbug@calzone.localnet>
Package: lintian
Version: 2.5.0~rc2
Severity: normal
Hi,
I just came across a possible problem with multi-arch support of
libraries that have arch-dependent overrides. I do not mean overrides
that simply differ in the triplet, that is covered in bug #617991:
"lintian overrides should support more precise pattern matching".
The more interesting case is a package that has actually different
overrides per arch and therefore can't be covered by better wildcard
matching. For example the libgl1-nvidia-glx package has the following
(and some more) overrides on i386:
# The NVIDIA license does not allow any form of modification.
libgl1-nvidia-glx: binary-has-unneeded-section
libgl1-nvidia-glx: shared-lib-without-dependency-information
libgl1-nvidia-glx: shlib-with-non-pic-code
libgl1-nvidia-glx: shlib-without-PT_GNU_STACK-section
libgl1-nvidia-glx: spelling-error-in-binary
and on amd64:
# The NVIDIA license does not allow any form of modification.
libgl1-nvidia-glx: binary-has-unneeded-section
libgl1-nvidia-glx: shared-lib-without-dependency-information
libgl1-nvidia-glx: shlib-with-executable-stack
libgl1-nvidia-glx: spelling-error-in-binary
Installing different overrides files in different architectures of a
Multi-Arch: same package will cause a file conflict on
/usr/share/lintian/overrides/<package> as described in
http://wiki.debian.org/Multiarch/Implementation
"Note that any files in /usr/share or /etc must be byte-for-byte
identical across architectures, otherwise file conflicts will
result! ..."
Installing a superset of the overrides on all architectures will cause
unused-override warnings (which could be overridden ...).
I see the follwing possible solutions, but there are probably more and
better ones:
1. allow installation of overrides in
/usr/share/lintian/overrides/<package>:<arch>
(using the multi-arch selection syntax for dpkg and friends) or
/usr/share/lintian/overrides/<package>.<arch>
(but that allows for file conflicts between packages "foobar" and
"foobar.i386") or whatever suffix dpkg will use to distinguish files
from same package, different arch in /var/lib/dpkg/info.
This could be done automatically by dh_lintian when encountering a
debian/[<package>.]lintian-overrides.<arch> file for a Multi-Arch:
same package in debian/compat level 9.
Lintian needs to be changed to prefer overrides/<package>:<arch> over
overrides/<package>.
A package providing more than one file in
/usr/share/lintian/overrides/ is malformed.
2. add :<arch> suffix to the overrides as in
libgl1-nvidia-glx:i386: shlib-with-non-pic-code
libgl1-nvidia-glx:amd64: shlib-with-executable-stack
libgl1-nvidia-glx: spelling-error-in-binary
and ignore override entries for foreign architectures.
The :<arch> suffix may not be used for Arch: all packages.
A disadvantage I see here is that it is not easy to exclude an
override for just one architecture, one has to enumerate all other
archs and add the override for them. This could be solved by 3.:
3. add architecture selection similar to dpkg as in
libgl1-nvidia-glx [i386]: shlib-with-non-pic-code
libgl1-nvidia-glx [!i386]: shlib-with-executable-stack
libgl1-nvidia-glx [amd64, i386]: spelling-error-in-binary
Andreas
--- End Message ---
--- Begin Message ---
Source: lintian
Source-Version: 2.5.0
We believe that the bug you reported is fixed in the latest version of
lintian, which is due to be installed in the Debian FTP archive:
lintian_2.5.0.dsc
to main/l/lintian/lintian_2.5.0.dsc
lintian_2.5.0.tar.gz
to main/l/lintian/lintian_2.5.0.tar.gz
lintian_2.5.0_all.deb
to main/l/lintian/lintian_2.5.0_all.deb
A summary of the changes between this version and the previous one is
attached.
Thank you for reporting the bug, which will now be closed. If you
have further comments please address them to 622888@bugs.debian.org,
and the maintainer will reopen the bug report if appropriate.
Debian distribution maintenance software
pp.
Niels Thykier <niels@thykier.net> (supplier of updated lintian package)
(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing ftpmaster@debian.org)
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256
Format: 1.8
Date: Tue, 10 May 2011 20:44:32 +0200
Source: lintian
Binary: lintian
Architecture: source all
Version: 2.5.0
Distribution: unstable
Urgency: low
Maintainer: Debian Lintian Maintainers <lintian-maint@debian.org>
Changed-By: Niels Thykier <niels@thykier.net>
Description:
lintian - Debian package checker
Closes: 588873 607694 614715 617991 622888 623390 623780 623878 624149 626038 626210
Changes:
lintian (2.5.0) unstable; urgency=low
.
* Summary of tag changes:
+ Added:
- duplicate-files
- format-3.0-but-debian-changes-patch
- extended-description-contains-empty-paragraph
- git-patches-not-exported
.
* checks/binaries:
+ [NT] Fixed false-positive embedded-library. Lintian incorrectly
assumed that none of these binary packages would contain a source
version in their source field. (Closes: #626038)
* checks/copyright-file.desc:
+ [ADB] Correct a spelling mistake ("coypright" -> "copyright") in the
description of copyright-refers-to-deprecated-bsd-license-file;
thanks, Kees Cook. (Closes: #623780)
* checks/description{,.desc}:
+ [NT] Check for empty paragraphs in extended description. Thanks to
Gerfried Fuchs for the patch.
* checks/duplicate-files{,.desc}:
+ [NT] Added to check for duplicate files in usr/share/doc. Thanks to
Axel Beckert for the suggestion. (Closes: #614715)
* check/fields:
+ [NT] Accepted pach from Antonio Terceiro to add Ruby-Version as a
accepted field in binary packages. Thanks to Vincent Carmona for
the report. (Closes: #623390)
* check/files:
+ [NT] Accepted multiple patches (merged into one) from Marcelo Jorge
Vieira to detect a number of new javascript libraries.
(Closes: #623878)
* checks/patch-systems{,.desc}:
+ [NT] Detect when 3.0 (quilt) packages have a debian-changes-VERSION
patch without a patch header. Thanks to Iain Lane for suggestion
and patch. (Closes: #588873)
* checks/debian-source-dir{,.desc}:
+ [NT] Applied patch from David Bremner to enable
git-patches-not-exported. (Closes: #607694)
* checks/java:
+ [NT] Tightened the regex for finding java libraries and restricted
missing-classpath tag to only be emitted for packages that actually
contain at least one jar file we check. This fixes false-positives
for packages that depended on libX-java-doc packages or packages
that only contain jar files we ignore.
* check/scripts{,.desc}:
+ [NT] Accepted patch from Antonio Terceiro to allow ruby-interpreter
as alternative to ruby for /usr/bin/ruby scripts.
(Closes: #626210)
.
* data/common/source-fields:
+ [NT] Accepted pach from Antonio Terceiro to add Ruby-Version as a
accepted field in binary packages. Thanks to Vincent Carmona for
the report.
.
* lib/Lintian/ProcessableGroup.pm:
+ [NT] Properly handle if the changes file is located directly in the
file system root. Thanks to Trent W. Buck for the bug report.
(Closes: #624149)
* lib/Lintian/Tags.pm:
+ [NT] Allow overrides to use wildcards anywhere instead of only in
the beginning or the end. Thanks to Steve R. Langasek for the
report. (Closes: #617991)
+ [NT] Allow simple architecture overrides such as [i386] or [!i386].
Architecture wildcards are not supported yet. The actual syntax
for the architecture the same as the one used in the Build-Depends.
Thanks to Andreas Beckmann for the bug report and the solution
proposal. (Closes: #622888)
Checksums-Sha1:
e5ff5d46d2b702f03b72849c4bbf1100f93b3915 2402 lintian_2.5.0.dsc
fa7a65c2fdf28b4b70aae5ce866a08f49b571fe9 924659 lintian_2.5.0.tar.gz
5c5cc472d276835bc8e28f11bfb548ba26803d2c 585644 lintian_2.5.0_all.deb
Checksums-Sha256:
e343e2d8bebb2c5b5315c3c0e70948a988919364bfe4be8918bdc62f0dcd7af5 2402 lintian_2.5.0.dsc
22b2ae111fd45809f7c8933b25069079cbbbfa0352b1212c12f4fd8235f97dea 924659 lintian_2.5.0.tar.gz
e70932268f2ac11f2be032ad2dc07a761205940dbd71959cbf578ca75302082c 585644 lintian_2.5.0_all.deb
Files:
087aa590f74b65781e24571c3c31da07 2402 devel optional lintian_2.5.0.dsc
bccd22190c9f171116eadb45322c6a65 924659 devel optional lintian_2.5.0.tar.gz
29b06fe0632a663d9c096f4a2cb644ad 585644 devel optional lintian_2.5.0_all.deb
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=sKoJ
-----END PGP SIGNATURE-----
--- End Message ---
Reply to: