Age | Commit message (Collapse) | Author | Files | Lines |
|
This makes it possible to use the pkexec tool to execute u-a from GUI
programs.
We need to bump the gettext build-dependency to at least 0.19.7 which is
the first version introducing ITS support and GETTEXTDATADIR variable
support (GETTEXTDATADIRS support was added in 0.19.8).
Proposed-by: Boyuan Yang <073plan@gmail.com>
|
|
The code here was handling pre-triggers and pre-multiarch downgrades.
|
|
This should help some downstreams and derivatives which package
update-alternatives on its own package, for example.
|
|
The default source compression has been xz for a while.
|
|
[guillem@debian.org:
- Mark the spec as an experimental draft.
- Add new Background section.
- Rename Package-Creation-Requires-Root to Rules-Requires-Root.
- Rename dpkg/rules-requires-root to dpkg/target-subcommand.
- Add new dpkg/target/<target-name> keyword.
- Use dpkg-deb --build instead of dpkg --build.
- Fix DPKG_GAIN_ROOT_CMD invocation, and denote it as the one to use
instead of the alternative wrapper command.
- Rearrange sections and reflow text. ]
Signed-off-by: Guillem Jover <guillem@debian.org>
|
|
Fixes: commit c991f1bf5c799f72caecb6d62bcb339171796aa4
|
|
This will check that the module $VERSION matches the newest entry
in the CHANGES section if there is any POD present.
|
|
Fix changelog for dpkg 1.18.11 to mention PIE got enabled by default for
all architectures, not just the ones where gcc does that itself.
When emitting PIE flags on architectures where gcc does not inject those
itself, do it via a specs file too, so that maintainers can use them
unconditionally regardless of the object being compiled or linked.
When injecting -no-pie for linking via gcc specs also inject -fno-PIE.
Update the documentation to make the current situation more clear.
|
|
Add support for compiler built-in features, so that we do not set
them when enabled and set negated flags when disabled.
We use gcc spec files to set these flags so that we avoid any conflict
with other incompatible flags that would make the build fail.
Closes: #835149
Based-on-patch-by: Bálint Réczey <balint@balintreczey.hu>
|
|
This allows to detangle the libc used from the calling conventions.
|
|
Move various easy tests from the dpkg-tests repository, so that we can
test whether the infrastructure is viable.
|
|
These were upgrade recipes from 1.9.x, 1.10.x and 1.15.x.
|
|
There are very few packages now using the hardcoded paths, with bugs and
patches. It's about time to clean this up.
|
|
|
|
|
|
|
|
Make sure all SYNOPSIS sections have compilable code.
|
|
Verify that the perl version we want to restrict ourselves in the coding
style is enforced through the test suite.
|
|
|
|
Some of the options that can be configured globally can leak local
repository information when generating the ChangeLog, such as the
--decorate=long option for example.
|
|
With «use warnings» some inermodule issues such as redefinitions of
symbols are not catched, because that is too late.
Define PERL in the unit test environment so that we can invoke the
detected interpreter.
And disable the now redundant syntax_ok() check from the strict.t unit
test.
|
|
|
|
Regression introduced in commit 34b26c7db232b300991df5fb6f52707b381df28f.
Some functional tests are passing through makefile and exec barriers,
where the lib parameter for TAP::Harness is lost. Reintroduce the
explicit global setting of PERL5LIB so that it gets inherited for all
subprocesses.
Closes: #801329
Reported-by: Jérémy Bobbio <lunar@debian.org>
Stable-Candidate: 1.17.x
|
|
|
|
|
|
This allows to easily replace the shebang at installation time to use
the system perl interpreter and will make it easier to analyze all
perl scripts in the source tree.
|
|
There are still packages using those paths, but the relevant lintian
check did not list any, so these got removed prematurely.
|
|
We do not need TEST_PREFIX anymore, as the prefixes are implicit for
each typo of test case. Use the build directory for compiled programs
and the source directory for scripts.
|
|
|
|
Newer lcov versions already produce UTF-8 encoded output, so there's no
need to use our local html prolog.
Ref: #590562
|
|
Add correct summary values, create a percentage bar, and remove the
lcov-epilog template, which was being inserted in every and each
generated lcov html file, not just the indices.
The injection should be considered fragile, as it depends on the input
report not changing its structure. But this is no worse than using the
local prolog and epilog html templates.
|
|
This got disabled by default starting with lcov 1.10, enable it back
to get the full coverage report.
|
|
Switch from --checksum to --rc geninfo_checksum=1 so that it can be
passed to all commands, even the ones that do not recognize --checksum.
|
|
|
|
Print a text report to stdout, like with the Perl coverage.
|
|
|
|
This stops the madness of having to invent sequence numbers to be able
to order the test suite run. Instead we now honour the order of the
arguments passed to the runner.
|
|
The original name was confusing with Makefile.am, let's be more clear.
|
|
|
|
The ChangeLog file is distributed, and cannot be regenerated outside
of the git repository, so do not remove it in DISTCLEANFILES, and move
the generation code into dist-hook, which also avoids unnecessary
computation during normal builds.
|
|
Do not fix the shlibs conffiles to avoid possible useless prompts.
Fixes CodeLayout::ProhibitTrailingWhitespace.
Warned-by: perlcritic
|
|
Mark false positives in the perl code so that perlcritic ignores them,
and so that they are documented in-place.
|
|
This test is useful for all perl code in the project, not just for the
scripts directory. Move it to the top level directory so that we don't
have to do strange things referencing sibling directories and similar.
|
|
There are no packages using absolute paths in the Debian archive anymore.
|
|
It serves no purpose, it's Debian specific, it's usually outdated
relative to the main archive, and everyone should be using
dpkg-architecture anyway, or looking at the Debian archive for the
distribution supported architectures.
|
|
This sets up the environment to be able to run the local scripts using
local modules and data files.
|
|
For now document the database locking “interface”.
Closes: #670897
|
|
For architectures where the ABI changes some attributes, like MIPS n32
or AMD64 x32, the architecture bits do not match the ones from the cpu,
so we need to override them.
|
|
The standard way to select if a specific component of the build is to
be enabled or disabled is through --enable-foo and --disable-foo
options, --with-foo and --without-foo are used for selecting external
modules to be used.
|
|
Regression introduced in commit d7c001aa6b3d49f1542df05cde60406be2e981ec.
|