apt/apt-pkg/algorithms.cc, branch 2.9.22

apt-pkg/algorithms.cc, apt-pkg/cachefilter.h: drop unused #include

2024-11-12T19:36:10Z

Return an error if ResolveByKeep() is unsuccessful

2024-11-11T11:30:01Z

In case of held packages or packages that otherwise can't be kept back, we should return an error and not require the client to check the broken_count after calling us. LP: #2078720

Fix keeping back removals of obsolete packages

2024-11-11T11:30:01Z

ResolveByKeepInternal() inadvertently considered packages that were marked for removals as not being possible to keep back, but that's not true, all they need is to have a currently installed version. The test case is a bit awkward to construct because our upgrade code doesn't get into a place where this happens that we know off. Or maybe it does with phased upgrades, but here we simplify this by using a helper binary that dumps state, marks for removal, and then calls ResolveByKeep and checks the removal is undone. LP: #2078720

Modernize standard library includes

2024-02-20T12:49:04Z

This was automated with sed and git-clang-format, and then I had to fix up the top of policy.cc by hand as git-clang-format accidentally indented it by two spaces.

Merge branch 'keep-kernel-size' into 'main'

2024-02-20T08:30:05Z

Configure the amount of kernels to keep See merge request apt-team/apt!324

Add public phased update API

2024-02-13T13:27:27Z

This moves the functions of the PhasedUpgrader class into various other classes so they can be publicly exposed. This introduces three new functions: pkgDepCache::PhasingApplied() tells you whether phasing should be applied to the package. pkgProblemResolver::KeepPhasedUpdates() keeps back updates that have phasing applied. pkgCache::VerIterator::IsSecurityUpdate() determines whether this version contains security fixes.

Configure the amount of kernels to keep

2024-01-24T17:23:33Z

This commit introduces the following configuration for keeping a configurable amount of kernels: APT::NeverAutoRemove::KernelCount The logic dictates that the running kernel and the latest kernel are not autoremoved. In case the running kernel is the latest kernel, the previous kernel is kept. Any count lower than two is therefore disregarded. This is in line with the previous behavior. The default is therefore similar to: APT::NeverAutoRemove::KernelCount 2; This will be ignored and we will still keep two: APT::NeverAutoRemove::KernelCount 1; This will keep 3 kernels (including the runnig, and most recent) APT::NeverAutoRemove::KernelCount 3; Signed-off-by: Wesley Schwengle

dist-upgrade: Revert phased updates using keeps only

2023-07-05T08:57:26Z

In the bug, mutter was kept back due to phasing and the new gnome-shell depended on that, and was therefore kept back as well, however, gnome-shell-common was not broken, and apt decided to continue upgrading it by removing gnome-shell and the ubuntu desktop meta packages. This is potentially a regression of LP#1990586 where we added keep back calls to the start of the dist-upgrade to ensure that we do not mark stuff for upgrade in the first place that depends on phasing updates, however it was generally allowed by the resolver to also do those removals. To fix this, we need to resolve the update normally and then use ResolveByKeepInternal to keep back any changes broken by held back packages. However, doing so breaks test-bug-591882-conkeror because ResolveByKeep keeps back packages for broken Recommends as well, which is not something we generally want to do in a dist-upgrade after we already decided to upgrade it. To circumvent that issue, extend the pkgProblemResolver to allow a package to be policy broken, and mark all packages that already were already going to be policy broken to be allowed to be that, such that we don't try to undo their installs. LP: #2025462

Keep "or group" when installing package to satisfy it

2023-05-02T11:59:45Z

Check state of dependency, not dependee in dependency keep back

2022-09-28T14:32:44Z

When iterating over I's dependencies (which are called Pkg), we accidentally checked if I was Protected() instead of Pkg when deciding whether Pkg can be kept back. LP: #1990684