Bug#1053892: lintian: Stop emitting masked tags

2023-10-13 Thread Raphaël Hertzog
Package: lintian Version: 2.116.3 Severity: wishlist X-Debbugs-Cc: raph...@freexian.com I just discovered the existence of masked tags and of the "Screen" mechanism to exclude some tags. I also notice that they are displayed together with overriden tags when you use --show-overrides. Yet they are

Bug#1053891: lintian: Document the existence of classification tags in the manual page

2023-10-13 Thread Raphaël Hertzog
Package: lintian Version: 2.116.3 Severity: wishlist X-Debbugs-Cc: raph...@freexian.com Hello, the conversation in https://salsa.debian.org/freexian-team/debusine/-/merge_requests/300#note_434002 led me to realize that as a simple lintian user reading the manual page, you can't discover the

Bug#976083: lintian: reports spelling-error-in-changelog for duplicate word even when punctuation and empty line separate two words

2020-11-29 Thread Raphaël Hertzog
Package: lintian Version: 2.103.0 Severity: normal I have this warning: W: ccrypt: spelling-error-in-changelog Debian Debian (duplicate word) Debian And I believe it matches this part of the changelog entry: * Configure git-buildpackage for Debian [ Debian Janitor ] For me, it should not

Bug#968108: lintian: False positive for no-dh-sequencer, maybe due to unexpected target dependency

2020-08-08 Thread Raphaël Hertzog
Package: lintian Version: 2.87.0 Severity: normal In the zim package I have this in debian/rules: # Zim build system requires those environment variables... export USER=fake export HOME=$(CURDIR)/debian/fake-home $(CURDIR)/debian/fake-home: mkdir $(CURDIR)/debian/fake-home %:

Bug#968041: lintian: fails with internal error

2020-08-07 Thread Raphaël Hertzog
Package: lintian Version: 2.87.0 Followup-For: Bug #968041 User: de...@kali.org Usertags: origin-kali I get the same error for lzop as well as for unzip. You need to add a dependency on lzop too. Running lintian... No such file or directory at /usr/share/perl5/IPC/Run3.pm line 417. eval

Bug#968003: lintian: bad name for duplicate-globbing-patterns and redundant-globbing-patterns

2020-08-06 Thread Raphaël Hertzog
Package: lintian Version: 2.87.0 Severity: normal User: de...@kali.org Usertags: origin-kali I was hit with those errors (reproducible with commit cd6b90a5295aa5a171a1ac9c3c51590f568ca075 in the openvas-scanner git repository): E: openvas-scanner source: duplicate-globbing-patterns

Bug#966368: lintian gets stuck when run by sbuild within rebuildd

2020-07-27 Thread Raphaël Hertzog
Package: lintian Version: 2.85.0 Severity: important User: de...@kali.org Usertags: origin-kali In Kali, our build daemons run "rebuildd" with a build script that calls sbuild --run-lintian. Since lintian 2.85 (I believe 2.84.0 is not affected), the build process get stuck at the point when

Bug#966024: lintian: send-patch tag is not properly detecting "Forwarded: not-needed"

2020-07-22 Thread Raphaël Hertzog
Package: lintian Version: 2.85.0 Severity: normal User: de...@kali.org Usertags: origin-kali I have the following tags: I: i3-gaps source: send-patch debian/patches/Disable-a-failing-test-due-to-source-package-not-being-gi.patch I: i3-gaps source: send-patch

Bug#966022: lintian: False positive on missing-depends-on-sensible-utils with commands like i3-sensible-pager

2020-07-22 Thread Raphaël Hertzog
Package: lintian Version: 2.83.0 Severity: normal User: de...@kali.org Usertags: origin-kali In this package https://gitlab.com/kalilinux/packages/i3-gaps we have the following lintian errors: E: i3-gaps-wm: missing-depends-on-sensible-utils usr/bin/i3 E: i3-gaps-wm:

Bug#954860: lintian: absolute-symbolic-link-target-in-source duplicates source-contains-unsafe-symlink

2020-03-24 Thread Raphaël Hertzog
Package: lintian Version: 2.55.0 Severity: normal User: de...@kali.org Usertags: origin-kali I don't see the point of "absolute-symbolic-link-target-in-source" since "source-contains-unsafe-symlink" is already triggered on such files. Please remove that newly added tag. -- System Information:

Bug#942493: lintian: Complain of too long header fields

2019-10-17 Thread Raphaël Hertzog
Package: lintian Version: 2.27.0 Severity: wishlist Based on the problem discovered in #942487 where a Provides line of more than 256K slipped in the archive, I believe it would be nice if lintian could: 1/ emit a warning when a field is larger than say 16K (somehow to force the maintainer to

Bug#941656: lintian: changelog-file-missing-explicit-entry false positive for 2 consecutive NMUs (-X.1, -X.2)

2019-10-03 Thread Raphaël Hertzog
Package: lintian Version: 2.24.0 Severity: normal User: de...@kali.org Usertags: origin-kali I just uploaded ddd_3.3.12-5.2.dsc and I get this warning: W: ddd source: changelog-file-missing-explicit-entry 1:3.3.12-5.1 -> 1:3.3.12-5 (missing) -> 1:3.3.12-5.2 Yet the changelog file has all the

Bug#920691: lintian gets stuck collecting info after failed objdump-info

2019-01-28 Thread Raphaël Hertzog
Package: lintian Version: 2.5.123 Severity: important User: de...@kali.org Usertags: origin-kali $ dget http://http.kali.org/pool/main/m/metasploit-framework/metasploit-framework_5.0.1-0kali1_amd64.changes $ lintian --debug metasploit-framework_5.0.1-0kali1_amd64.changes N: Lintian v2.5.124 N:

Bug#891301: lintian: privacy-breach-generic should not complain about and others

2018-02-24 Thread Raphaël Hertzog
Package: lintian Version: 2.5.75 Severity: normal live-manual-html has many privacy-breach-generic warnings all due to the presence of http://www.sisudoc.org/; /> in the of the files. This field is merely documentation of what has been used to generate the file and will not cause any request by

Bug#889816: lintian: Complain when epoch has been bumped but upstream version did not go backwards

2018-02-07 Thread Raphaël Hertzog
Package: lintian Version: 2.5.73 Severity: wishlist When the last changelog entry implements an epoch bump, we want to make sure that the epoch is justified by the fact that the upstream version is lower than the former entry in the changelog. e.g when you switch from E¹:U¹-D¹ to E²:U²-D²

Bug#889814: lintian: Improve long description of epoch-change-without-comment

2018-02-07 Thread Raphaël Hertzog
Package: lintian Version: 2.5.73 Severity: wishlist It would be nice if the long description of epoch-change-without-comment could document when it's appropriate to bump the epoch and ways to avoid the epoch bump entirely. I'm thinking of: - upstream changed version numbering scheme in a way

Bug#871575: lintian: maintainer-address-causes-mail-loops-or-bounces should be fixed to allow @packages.debian.org emails

2017-08-09 Thread Raphaël Hertzog
Package: lintian Version: 2.5.51 Severity: normal I believe that the check maintainer-address-causes-mail-loops-or-bounces should no longer refuse "*@packages.debian.org" in the Maintainer field. I did patch the code generating the aliases on packages.debian.org to avoid mail loops:

Bug#865531: lintian: testsuite-autopkgtest-missing checks the wrong thing

2017-06-22 Thread Raphaël Hertzog
Package: lintian Version: 2.5.51 Severity: normal lintian complains with testsuite-autopkgtest-missing when debian/control is missing the "Testsuite" field but that field is usually not present in the unpacked source package because it is automatically added by dpkg-source to the .dsc when it

Bug#847318: lintian: should be aware of kali's differences

2016-12-07 Thread Raphaël Hertzog
Package: lintian Version: 2.5.49 Severity: wishlist User: de...@kali.org Usertags: origin-kali I would love if lintian could stop emitting unwanted errors/warnings in Kali Linux. In particular now that debuild fails when lintian fails (see #847168). At the very least, I want to get rid of the

Bug#809440: lintian: runtime-test-file-is-not-a-regular-file complains about a symlink to a regular test file

2015-12-30 Thread Raphaël Hertzog
Package: lintian Version: 2.5.39.1 Severity: normal With python-django 1.9-2 I get those tags: I: python-django source: runtime-test-file-is-not-a-regular-file debian/tests/django-admin-py3 I: python-django source: runtime-test-file-is-not-a-regular-file debian/tests/test-suite-py3 Both files

Bug#806237: lintian: false positive for obsolete-url-in-packaging in debian/watch

2015-11-25 Thread Raphaël Hertzog
Package: lintian Version: 2.5.38.1 Severity: normal User: de...@kali.org Usertags: origin-kali I just got this warning: W: bully source: obsolete-url-in-packaging debian/watch http://code.google.com/p/wps-bully/ But my watch file is like this: $ cat debian/watch version=3 # The original

Bug#679132: lintian: false positive on package-uses-local-diversion when --local and --package are not given

2012-06-26 Thread Raphaël Hertzog
Package: lintian Version: 2.5.9 Severity: normal When you use a dpkg-divert invocation like this one: dpkg-divert --rename --divert $file.original --add $file Lintian triggers an error package-uses-local-diversion. But this error is wrong since those calls are in maintainer scripts and thus

Bug#647028: lintian: warn when changelog contains UNRELEASED even if .changes contains something valid

2011-10-29 Thread Raphaël Hertzog
Package: lintian Version: 2.5.3 Severity: wishlist I build my package with sbuild and it writes the Distribution field in the .changes according to the parameter given on the command line. I recently had the case that I have left UNRELEASED in debian/changelog and built my package with sbuild

Bug#637473: lintian: false positive on embedded-php-library for streams.php

2011-08-11 Thread Raphaël Hertzog
Package: lintian Version: 2.5.1 Severity: normal I get this with wordpress: W: wordpress: embedded-php-library usr/share/wordpress/wp-includes/pomo/streams.php This streams.php is not the one of php-gettext and thus the tag is a false positive. I would suggest to update the regex in

Bug#631786: lintian: Verify that packages do not use/set DEB_*_SET / DEB_*_APPEND to influence dpkg-buildflags

2011-06-27 Thread Raphaël Hertzog
Package: lintian Version: 2.5.1 Severity: wishlist Hello, dpkg-buildflags can use some environment variables DEB_(CFLAGS|CXXFLAGS|FFLAGS|CPPFLAGS|LDFLAGS)_(SET|APPEND) to influence the flags that it outputs. But those variables are meant for users would are recompiling the package not for

Bug#618001: lintian: maintainer-script-lacks-debhelper-token is sometimes not reported when it should

2011-03-13 Thread Raphaël Hertzog
Package: lintian Version: 2.5.0~rc1 Severity: normal I'm building libapache2-mod-qos_9.54-1.dsc. When I do a full build I get this: $ debuild [...] Now running lintian... N: 1 tag overridden (1 error) Finished running lintian. When I do a source-only build I get this: $ debuild -S [...] Now

Bug#616493: lintian: Should know about the Multi-Arch field

2011-03-04 Thread Raphaël Hertzog
Package: lintian Version: 2.5.0~rc1 Severity: normal User: debian-d...@lists.debian.org Usertags: multiarch While building my multi-arch dpkg I noted this: I: dpkg: unknown-field-in-control multi-arch I: dselect: unknown-field-in-control multi-arch Lintian should learn about this field and not

Bug#571616: lintian: Don't report timewarp-standards-version when UNRELEASED

2010-02-26 Thread Raphaël Hertzog
Package: lintian Version: 2.3.3 Severity: normal I just got this warning: W: dpkg source: timewarp-standards-version (2009-11-17 2010-01-27) The changelog entry is UNRELEASED and still has the date of when it got created. But we fixed stuff and in particular we upgraded Standards-Version to

Bug#566820: lintian: Warn about missing debian/source/format, advise switch to new 3.0 source formats

2010-01-25 Thread Raphaël Hertzog
Package: lintian Version: 2.3.1 Severity: wishlist As part of the plan to have the new source formats as the default formats in Debian I would like lintian to give a warning when debian/source/format doesn't exist, it could be named missing-debian-source-format. I suggest this description: ---

Bug#560149: lintian: disable latest-debian-changelog-entry-without-new-date when UNRELEASED

2009-12-09 Thread Raphaël Hertzog
Package: lintian Version: 2.2.18 Severity: normal I regularly get this warning on my packages during development because the latest changelog entry has been created before some other bug-fix only release of the former stable release that I merged into the current development release. Simple case

Bug#554025: lintian: *patch-missing-description should recommend following DEP-3

2009-11-02 Thread Raphaël Hertzog
Package: lintian Version: 2.2.17 Severity: wishlist It would be nice if the extended description of quilt-patch-missing-description and dpatch-missing-description would mention DEP-3, and recommend its usage. It should point to http://dep.debian.net/deps/dep3/ for further information. Cheers,

Bug#553207: lintian: wrong patch-system-but-no-source-readme when using quilt with 3.0 (quilt) source packages

2009-10-29 Thread Raphaël Hertzog
Package: lintian Version: 2.2.17 Severity: normal I just built and uploaded my first 3.0 (quilt) source package and aside from unsupported-source-format (#552707) I also get: W: ftplib source: patch-system-but-no-source-readme This warning is wrong since with this source format the quilt patch

Bug#534276: lintian: read-in-maintainer-script false positive

2009-06-23 Thread Raphaël Hertzog
Package: lintian Version: 2.2.12 Severity: normal read-in-maintainer-script is tagged severity certain but it can have false positives... example: { read field1 [...] read field2 } /var/lib/dpkg/alternatives/vi or with a function: parse_something() { read foo } echo bla |