apt/test, branch 3.1.14Debians commandline package managerhttps://git.kalnischkies.de/apt/atom?h=3.1.142026-01-23T11:04:31Zvariants: Do not report unsupported variants in repositories2026-01-23T11:04:31ZJulian Andres Klodejak@debian.org2026-01-23T11:03:04Zurn:sha1:49e80c2d9061372e0ec697be43bcba4c451b7752
We configure all variants the CPU supports as active, but repositories
usually do not provide all of them, so let's not be noisy.
why: Render info about all providers on virtual packages2026-01-20T13:07:15ZJulian Andres Klodejak@debian.org2026-01-20T13:04:20Zurn:sha1:34876ed2e5cdf504e2a4f9721486d279ae5f3be6
Refactor code into a doOne helper lambda, and then add loop
over the providers to use it.
This fixes a crash seen by Benjamin in a nice way than just
failing :)
Reported-by: Benjamin Drung
solver3: Rename decision to assignment2026-01-05T21:20:24ZJulian Andres Klodejak@debian.org2025-12-28T15:56:28Zurn:sha1:21d099878ed8c34f3b13747bcec380e0402e57a3
The previous use of decision here conflicted with the use
of decision level and the general notion of having made a
decision, because the assignment might have been propagated
as a matter of fact.
Correctly copy provided version in m-a lazy-init2025-11-10T10:01:02ZJulian Andres Klodejak@debian.org2025-11-10T09:41:20Zurn:sha1:2f4299bf39c85b694876cb8030fae0887e8b5edc
When a provides was copied, we inadvertently copied the package
version instead of the provided version. Copy the provided version
instead.
Closes: #1120463
Revert "depcache: Allow changing multi-arch versions for removed siblings"2025-10-26T08:31:20ZJulian Andres Klodejak@debian.org2025-10-26T08:14:20Zurn:sha1:22786e32b613518ae3a5cd07e9e2f1419b2c6552
This reverts commit 74636b64a7ee56637a3ce4ddc1d060ca9f8b77e1.
Adjust the test case accordingly, as it is no longer possible to
install libgcc-s1:s390x. We need to sort this out in the ToDepCache
translation layer at a later time.
Sadly the depcache rejects valid solutions like this; solving this
isn't trivial. Technically we need to split up the FromUser and
auto-bit handling to solve this, but even then, MarkDelete() with
FromUser=0 moves the auto-bit on metapackages; this doesn't happen
when doing FromUser=1, so we can't always do FromUser=1.
Fixes: a3fca052ca21ad222ac7f2fdd7f3fe84b44beb60
test: Correctly skip tests when not using -j2025-10-26T07:28:35ZJulian Andres Klodejak@debian.org2025-10-26T06:59:52Zurn:sha1:487ce0983c2008991566b4cfb269fd7d29a08e98
We never implemented comment support in the skip test file for
-j as we only run with -j, but the debian/tests/run-tests runs
without parallelism.
tests: Do not skip tests for --solver internal2025-10-26T07:28:35ZJulian Andres Klodejak@debian.org2025-10-26T06:58:43Zurn:sha1:78da58ec0c1cdbda6cdd22c63f85ddac703bb0d6
Remove the exports for the default as we don't want those
inherited, they can just be reset. Set the skip file to
/dev/null when a solver is selected, to make sure that
tests don't fall back to the default skip file.
Adjust test cases to treat 3.0 as default APT_SOLVER if not set2025-10-26T07:28:08ZJulian Andres Klodejak@debian.org2025-10-26T07:24:39Zurn:sha1:5e8e69cdd6b512480f3208298725e1e44c80f06e
Not needed so far because we export a default 3.0 APT_SOLVER, but
quite useful if we stop doing so (next commit).
solver3: Allow removing manually installed packages2025-10-25T20:16:54ZJulian Andres Klodejak@debian.org2025-10-25T19:31:34Zurn:sha1:a3fca052ca21ad222ac7f2fdd7f3fe84b44beb60Enable the new solver by default in 0.31, 1.21, 2.11, 3.12025-10-25T20:16:54ZJulian Andres Klodejak@debian.org2025-10-25T15:42:11Zurn:sha1:c5c5d10d1b2ac0a589102315840c4f2a075b18c1
Enable the new solver by default and remove the vendor-specific
configuration change in Ubuntu's drop-in.