not x or not y = not (x and y)

Gbp-Dch: Ignore

Gbp-Dch: Ignore

Simon McVittie authored Mar 05, 2022 and Axel Beckert committed Jun 13, 2022

binutils reports that it is unable to find the program interpreter name
for detached debug symbols (#1000449), but it's entirely reasonable
that detached debug symbols don't have an interpreter, since they are
not a complete executable program.

This false positive is so widespread that it's making Lintian less
useful for detecting other issues, and it seems like this is not going
to be solved in binutils any time soon, so silence it to improve the
signal-to-noise ratio.

Closes: #1000977

Closes: #1005046

We add our own second update-rc.d call for that. The replacement for
the #DEBHELPER# token will add a third one, but lintian will only
argue about the first such duplicate. (Which might be a minor issue
itself, but is for now our luck here. :-)

Additionally add "> /dev/null" to the first update-rc.d call to avoid
triggering the output-of-updaterc.d-not-redirected-to-dev-null.

This is currently not needed for the second call, which actually seems
to be another potential bug, see https://bugs.debian.org/1012732

This will hopefully be the last commit which was needed to make
Lintian's testsuite to pass again. :-)

That way, that test can also be easily run standalone _and_
Perl::Critic won't argue (with "q{.}" not even about "Quotes used with
a noisy string"). Also feels clean. :-)

Revert "t/scripts/harness/tag-coverage.t: use FindBin instead of $LINTIAN_BASE to allow one to run it via prove"

Perl::Critic policy Community::DiscouragedModules is not
happy.

This reverts commit 3900afa6.

Only affects the file ".perlcritic".

t/scripts/harness/tag-coverage.t: use FindBin instead of $LINTIAN_BASE to allow one to run it via prove

Based on the output of "git grep -hE -o 'Copyright.*Lechner' | sort -u".

Remove literal unicode character U+0334 COMBINING TILDE OVERLAY which likely had been added accidentally

… slightly over 10 years ago and nobody noticed so far.

For more details see
https://lists.debian.org/msgid-search/20220611194509.ykyasxuwnss6mpih@sym.noone.org

Replace UTF-8 characters in code with \N{…} until https://github.com/Perl-Critic/Perl-Critic/issues/925 is fixed

Lintian's own test t/scripts/01-critic/libs.t fails otherwise.

Replace UTF-8 characters in mostly Copyright comments until https://github.com/Perl-Critic/Perl-Critic/issues/925 is fixed

Actually all of them are copyright comment except for a single Unicode
SECTION SIGN character somewhere else.

Lintian's own test t/scripts/01-critic/libs.t fails otherwise.

It still fails, but no more for 167 files, now only for 64 files.

Lintian's own test t/scripts/01-critic/scripts.t fails otherwise.

Remove unnecessary usage of UTF-8 from bin/lintian until https://github.com/Perl-Critic/Perl-Critic/issues/925 is fixed

Lintian's own tests t/scripts/01-critic/{bin,private}.t fail otherwise.

… before lintian starts complaining about it. ;-)

Replace all occurrences of "Copyright ©" with "Copyright (C)" again until https://github.com/Perl-Critic/Perl-Critic/issues/925 is fixed

This more or less reverts 1a3f6eee
(Replace "Copyright (C)" with the Unicode copyright symbol in our own
code for consistency.) for now.

Gbp-Dch: Ignore

Chris Lamb stated in
https://lists.debian.org/msgid-search/5e4d0e28-a3f4-4302-8364-5afd93d8ae17@www.fastmail.com:
"I am no longer a Lintian developer"

Upgrade missing-systemd-timer-for-cron-script to warning; no longer experimental. (Closes: #1007257)

Until now, the visibility was experimental and pedantic, which is the lowest
combination still shown to users. The requested change effectively bumps the
visibility by three levels.

The change is more dramatic than what we would normally entertain in a single
step.

At the same time, the tag has existed for over three years. (See Bug#922862.)
In addition, the website showed only 234 hits in 185 sources across the entire
archive.

While some users may disagree with the change, the risk of widespread
irritation is low.

Thanks to Josh Triplett for bringing the matter to our attention!

Gbp-Dch: ignore

Allows for a recursive flag (or any other flags).

Includes a test case.

Thanks to Marc Haber for bringing the matter to our attention!

Avoids that error, which appeared after a run of 'git clean -f -x -d':

===(      83;125  0/1  0/1  0/1  0/1  0/1  0/1  0/?  0/? )==============# Hints do not match

--- debian/test-out/eval/checks/control-files/control-files-weird-files/hints.specified.calibrated
+++ debian/test-out/eval/checks/control-files/control-files-weird-files/hints.actual.parsed
+control-files-weird-files (binary): control-file-has-bad-owner 1000/1000 != root/root (or 0/0) [triggers]

Unexpected tags:
  control-file-has-bad-owner

===(      84;126  0/1  0/1  1/1  0/1  0/1  0/1  0/1  0/1  0/1 )=========#   Failed test 'Lintian passes for control-files-weird-files'
  at /lcl/lechner/lintian/git/lib/Test/Lintian/Run.pm line 341.
Looks like you failed 1 test of 1.
debian/test-out/eval/checks/control-files/control-files-weird-files/generic.t ......................................................... Dubious, test returned 1 (wstat 256, 0x100)
Failed 1/1 subtests

Gbp-Dch: ignore

Gbp-Dch: ignore

Works but is slow. Perhaps not quite so much of the readelf information needs
to be retained.

The impact on average library packages has yet to be evaluated.

It may only is be issue for some installables. The one from the bug, for
example (picolibc-arm-none-eabi_1.7.4-1_all.deb) seemed to generate several
million symbols whose purpose was not clear to me. That impression is dated
because I looked into the reason for the high memory usage in January.

Without this change, the pod2man-generated output generates a verbatim
section of troff:

.PP
.Vb 4
\&  (Note, though, that the B<lintian> command has a command line option
\&B<\-i> to display the same results as B<lintian\-annotate\-hints>, so you
\&normally do not need to pipe the output of B<lintian> through this
\&command to see the extra information.)
.Ve
.SH "OPTIONS"

With the change, it produces the following roff data, including the
expected formatting:

.PP
(Note, though, that the \fBlintian\fR command has a command line option
\&\fB\-i\fR to display the same results as \fBlintian-annotate-hints\fR, so you
normally do not need to pipe the output of \fBlintian\fR through this
command to see the extra information.)
.SH "OPTIONS"

I believe this is because pod2man treats leading a indents as
indicative of a verbatim section.

This was only visible in Salsa CI.

Gbp-Dch: ignore

Avoids this error in Salsa CI:

Warning in processable /tmp/autopkgtest-lxc.53ulvpjl/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/packages/checks/debian/rules/rules-including-deprecated-makefiles/rules-including-deprecated-makefiles_1.0.dsc:
 Can't locate object method "peinted_hint" via package "Lintian::Check::Debian::Rules"
 at /usr/share/lintian/lib/Lintian/Check/Debian/Rules.pm line 255.

Gbp-Dch: ignore

This is probably the change in Debhelper that made the tag untestable:

    debian/debhelper@da49713e

The error probably went undetected for a while because Lintian's autopkgtest
pipeline has been out of order for a while. (For details, see Bug#998367.)
Avoids that error in the unstable pipeline on Salsa CI:

Failed build tasks:

>>> t/recipes/../../autopkgtest_tmp/build-and-evaluate-test-packages/packages/checks/debhelper/compat-level-not-a-number:
make --trace -f /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/Makefile DEFAULT_DH_COMPAT=13 failed at private/runtests line 787.
Building in /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/packages/checks/debhelper/compat-level-not-a-number
/tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/Makefile:63: target 'compat-level-not-a-number_1.0.orig.tar.gz' does not exist
mkdir compat-level-not-a-number-1.0
if [ -d /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/orig/. ] ; then \
cp -rp /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/orig/. compat-level-not-a-number-1.0 ; \
fi
/tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/Makefile:35: update target 'compat-level-not-a-number_1.0_amd64.changes' due to: compat-level-not-a-number_1.0.orig.tar.gz
if [ -r compat-level-not-a-number_1.0.orig.tar.gz -a -r /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/upstream/private-key.asc ] ; then \
tmpdir=/tmp/tmp.ik7mzVU0pg ; \
gpg --homedir "$tmpdir" --import /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/upstream/private-key.asc ; \
gpg --homedir "$tmpdir" --batch --yes --armor --output compat-level-not-a-number_1.0.orig.tar.gz.asc \
--detach-sign compat-level-not-a-number_1.0.orig.tar.gz ; \
gpg-connect-agent --homedir "$tmpdir" KILLAGENT /bye ; \
rm -rf "$tmpdir" ; \
fi
if [ -d /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/diff/. ] ; then \
cp -rp /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/diff/. compat-level-not-a-number-1.0 ; \
fi
if [ -d /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/debian/. ] ; then \
cp -rp /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/debian/. compat-level-not-a-number-1.0/debian ; \
fi
if [ -x /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/pre-build ] ; then \
/tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/pre-build /tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/packages/checks/debhelper/compat-level-not-a-number/compat-level-not-a-number-1.0 ; \
fi
cd compat-level-not-a-number-1.0 && dpkg-buildpackage -rfakeroot  -us -uc -d \
-iNEVER_MATCH_ANYTHING -INEVER_MATCH_ANYTHING --source-option=--auto-commit
dpkg-buildpackage: info: source package compat-level-not-a-number
dpkg-buildpackage: info: source version 1.0
dpkg-buildpackage: info: source distribution unstable
dpkg-buildpackage: info: source changed by Debian Lintian Maintainers <lintian-maint@debian.org>
dpkg-buildpackage: warning: debian/rules is not executable; fixing that
 dpkg-source -iNEVER_MATCH_ANYTHING -INEVER_MATCH_ANYTHING --auto-commit --before-build .
dpkg-buildpackage: info: host architecture amd64
dpkg-source: warning: --auto-commit is not a valid option for Dpkg::Source::Package::V3::Native
 debian/rules clean
make[1]: Entering directory '/tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/packages/checks/debhelper/compat-level-not-a-number/compat-level-not-a-number-1.0'
debian/rules:7: target 'clean' does not exist
dh clean
dh: error: The environment variable DH_COMPAT must be a positive integer
make[1]: *** [debian/rules:7: clean] Error 25
make[1]: Leaving directory '/tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/packages/checks/debhelper/compat-level-not-a-number/compat-level-not-a-number-1.0'
dpkg-buildpackage: error: debian/rules clean subprocess returned exit status 2
make: *** [/tmp/autopkgtest-lxc.6gdrvu2z/downtmp/autopkgtest_tmp/build-and-evaluate-test-packages/package-sources/checks/debhelper/compat-level-not-a-number/Makefile:39: compat-level-not-a-number_1.0_amd64.changes] Error 2

Gbp-Dch: ignore

Thanks to Paul Wise for the valuable suggestion!

Gbp-Dch: ignore

Gbp-Dch: ignore

Gbp-Dch: ignore

Thanks to Mark Hindley from Devuan for bringing the matter to our attention!

Gbp-Dch: ignore

Gbp-Dch: ignore