| Age | Commit message (Collapse) | Author |
|---|
|
Run `apt-get autoremove -o Debug::pkgAutoRemove=yes` and confirm the
logged reason for packages to be kept is correct.
Only check for specific debug lines containing 'MarkPackage:' in order
to prevent new debug logging to break the test case.
|
|
When running with Debug::pkgAutoRemove=yes, explain why certain packages
are being marked, either because they're marked essential/important or
because they match the blacklist from APT::NeverAutoRemove.
This should help troubleshoot cases where autoremove is not proposing
removal of packages expected to be up for removal.
Tested manually with `apt-get autoremove -o Debug::pkgAutoRemove=yes`.
|
|
id: '': no such user
./test-bug-611729-mark-as-manual: 59: [: Illegal number:
Regression-of: 68842e1741a5005b1e3f0a07deffd737c65e3294
Gbp-Dch: Ignore
|
|
Closes: #898886
|
|
Closes: 898797
|
|
Closes: 892792
|
|
Update .gitignore
See merge request apt-team/apt!16
|
|
Pu/gitlab ci
See merge request apt-team/apt!17
|
|
Salsa has support for GitLab CI. This introduces a test setup
for it, derived from the ones for shippable and travis. It is
not optimal yet: The build is run in the test stage. Fixing this
requires us to separate test from build dependencies, and storing
build/ as an artifact of the build stage; since build and test
stage run on different runners with fresh checkouts and images.
|
|
If sudo was invoked by root, SUDO_UID will be 0, and apt
will not print a Requested-By line.
|
|
|
|
Downloading from the same mirror we got a Release file from makes sense
for non-unique URIs as their content changes between mirror states, but
if we ask for an index via by-hash we can be sure that we either get the
file we wanted or a 404 for which we can perform a fallback for which
allows us to pull indexes from different mirror in parallel.
|
|
Individual items shouldn't concern themselves with these alternative
locations, we can deal with this more efficiently within the
infrastructure created for other alternative URIs now avoiding the need
to implement this in each item.
|
|
If we got a file but it produced a hash error, mismatched size or
similar we shouldn't fallback to alternative URIs as they likely result
in the same error. If we can we should instead use another mirror.
We used to be a lot stricter by stopping all trys for this file if we
got a non-404 (or a hash-based) failure, but that is too hard as we
really want to try other mirrors (if we have them) in the hope that they
have the expected and correct files.
|
|
Now that --with-source is supported in show we can go a little further
and add the "syntactic sugar" of supporting deb-files on the commandline
directly to give users an alternative to remembering dpkg -I for deb
files & as a bonus apt also works on changes files.
Most of the code churn is actually to deal with cases probably not too
common in reality like mixing packages and deb-files on the commandline
and getting the right order for these multiple records.
Closes: 883206
|
|
|
|
With the advent of compressed files and especially with in-memory
post-processed files the simple assumptions made in IsOk are no longer
true. Worse, they are at best duplicates of checks performed by the
cache generation (and validation) earlier and isn't used in too many
places. It is hence best to simply get right of these calls instead of
trying to fix them.
|
|
It is easier to prepend our fields, but that results in confusion for
things working on the so generated records as they don't start with the
usual "Package" – that shouldn't be a problem in theory, but in practice
e.g. "apt-cache show" shows these records directly to the user who
will probably be more confused by it than tools.
|
|
The syntax is a bit awkward, but it is the same as for a package name
and introducing another syntax wouldn't really help usability, so with
apt install ./foo.deb/experimental you will get the dependencies of foo
satisfied by your default release, but if this wouldn't satisfy the
version requirements the candidate for this dependency is switched to
the version from the experimental release. The same applies for apt
build-dep ./foo.dsc/stable-backports which was the initial request.
|
|
apt install pkg/release follows versioned dependencies in the candidate
switching if the current candidate does not satisfy the dependency,
so for uniformity the same should be supported in build-dep.
|
|
This shouldn't make a practical difference for most people, but for edge
cases it avoids DNS lookups and additionally prevents us from perfoming
unneeded SRV requests, too.
|
|
IP addresses are by definition not a domain so in the best case the
requests will just fail; we can do better than that on our own.
|
|
Gbp-Dch: Ignore
|
|
The boilerplate stems from automatic processing of these files as
most (with the exception of po/nl.po) are one-off (partial)
translations from years past.
Gbp-Dch: Ignore
|
|
Warned-by: i18nspector
Fixes: no-plural-forms-header-field
|
|
Repository cleanups
See merge request apt-team/apt!12
|
|
Warned-by: i18nspector
Fixes: no-plural-forms-header-field
|
|
Warned-by: i18nspector
Fixes: trailing-junk-in-plural-forms
|
|
Warned-by: i18nspector
Fixes: language-disparity
|
|
Warned-by: i18nspector
Fixes: c-format-string-argument-type-mismatch
Fixes: c-format-string-missing-arguments
|
|
Warned-by: i18nspector
Fixes: no-version-in-project-id-version
|
|
This is used from at least git shortlog.
|
|
Prompted-by: Jakub Wilk <jwilk@debian.org>
|
|
The formatting is very basic and more should probably be done, but we at
least get the files out of the root directory which in case of the
various READMEs was confusing salsa which one display as (central) README.
|
|
Reported-By: codespell & spellintian
Gbp-Dch: Ignore
|
|
Reported-By: gcc -Wdouble-promotion
Gbp-Dch: Ignore
|
|
GTest has a bunch of undefined macros which causes the compiler to spit
out warnings for each one on each test file. There isn't much we can do,
so we just disable the warning for the testcases. Other warnings like
sign-promo and sign-compare we can avoid by being more explicit about
our expected integer constants being unsigned.
As we are just changing testcases, there is no user visible change which
would deserve to be noted in the changelog.
Gbp-Dch: Ignore
Reported-By: gcc-8
|
|
Still allow the older one to be used.
Closes: #897149
|
|
|
|
Only use zstd defined variables if zstd was found.
|
|
|
|
We are opening 1.7, whoo
|
|
This makes cross-building a bit easier, and also porting to
other platforms.
|
|
|
|
Closes: 679580
|
|
|
|
[jak: Fix invalid empty line]
Closes: #895117
|
|
This reverts commit 57a00c50b14a49ed91816e3f4467e0f2e57ee772.
|
|
json-based hooks for apt cli tools
See merge request apt-team/apt!10
|
|
This allows third-party package managers like snap or flatpak
to hook in and suggest alternatives if packages could not be
found, for example.
This is still highly experimental and the protocol might change
in future versions.
|
