Age | Commit message (Collapse) | Author |
|
- operate optional on gzip compressed pdiffs
* apt-pkg/acquire-item.cc:
- don't uncompress downloaded pdiff files before feeding it to rred
|
|
not causing installation/removal of other packages (#64141)
|
|
|
|
- do not segfault if cache is not build (Closes: #254770)
|
|
- let apt-utils work with encoded tar headers if uid/gid are large.
Thanks to Nobuhiro Hayashi for the patch! (Closes: #330162)
|
|
- add a RealFileExists method and check that your configuration files
are real files to avoid endless loops if not (Closes: #604401)
|
|
could not be installed from experimental from unstable instead
|
|
requested on the commandline, e.g. with a modifier
|
|
- add SetCandidateRelease() to set a candidate version and
the candidates of dependencies if needed to a specified
release (Closes: #572709)
- change pkg/release behavior to use the new SetCandidateRelease
so installing packages from experimental or backports is easier
|
|
in the autoremove section and installing those (Closes: #604222)
|
|
not exist to not generate sources.list entries later for them
|
|
of status and Packages files with a testcase
|
|
build library in the test environment
|
|
|
|
only if they cause the remove of this essential (Closes: #601961)
|
|
(a lot) cleaner than using this ps "magic" as this fails now and then…
|
|
|
|
interpreted if a package with this name doesn't exist: e.g. g++
|
|
|
|
actually building all these packages
|
|
- mark all installed packages first without auto installation in
a dist-upgrade to prefer upgrading packages instead of installing
new packages in versioned or-groups (Closes: #605394)
|
|
- Explicitly disable compressed indexes at the start. This ensures that we
will actually test uncompressed indexes regardless of the internal
default value of Acquire::GzipIndexes.
|
|
- don't install previously not installed providers in a try
to statisfy a "Breaks: provides" dependency by upgrade
|
|
|
|
- try fixing before removing even if the fix is hidden in
a provides, hidden in the #590438 testcase
|
|
- respect the none-force even in LANG=C (Closes: #602573)
|
|
- fallback always to a suitable description (Closes: #601016)
|
|
behaviour can be undefined… (or at least can't be tested against dpkg)
|
|
- Add a FileFd::FileSize() method to get the size of the underlying
file and not the size of the content in the file as FileFd::Size()
does - the sizes can differ since the direct gzip integration
* methods/{gzip,bzip2}.cc:
- use FileSize() to determine if the file is invalid (Closes: #600852)
|
|
|
|
|
|
funtions (bash complains)
|
|
- source /etc/default/locale (if available) so that the
apt-get update cron job fetches the right translated package
descriptions
* fix test failure on amd64
|
|
|
|
- do not remove packages which the user requested for installation
explicitly while satisfying other install requests (Closes: #598669)
|
|
- do not check endpointer packages instead of only those which prevented
NeverAutoRemove settings from having an effect (Closes: #598452)
|
|
|
|
|
|
conflicts/breaks against virtual packages (LP: #614993)
|
|
- support 100-pinning in Release file with ButAutomaticUpgrades
as requested by the backports crew (Closes: #596097)
* apt-pkg/deb/deblistparser.cc:
- overrule NotAutomatic in case of ButAutomaticUpgrades
|
|
in future that APT plays considerable well with empty archives
|
|
- support also Dir::Etc::Trusted so that apt-key works in the same
way as the library part which works with the trusted files
|
|
|
|
skip-* instead of test-* to avoid calling him with run-tests
|
|
|
|
the notices as they will not be shown otherwise
|
|
- create Dir::Log if needed to support /var/log as tmpfs or similar,
inspired by Thomas Bechtold, thanks! (Closes: #523919, LP: #220239)
Easily done by moving a private method from pkgAcquire into the public
area of fileutl.cc to be able to use it also in here
|
|
|
|
- fix autoremove by using correct config-option name and
don't make faulty assumptions in error handling (Closes: #594689)
|
|
- don't use ReadOnlyGzip mode for PDiffs as this mode doesn't work
in combination with the AddFd methods of our hashclasses
Add also 2 testcases: one to test pdiffs in general and
one to test the handling of compressed indexes.
|