Bug#1132621: unblock: rust-rand/0.10.0+dfsg-2 and friends

[Paul Gevers]

Hi,

On 4/6/26 20:12, Peter Michael Green wrote:

On 06/04/2026 08:57, Paul Gevers wrote:

So, I went ahead and scheduled tests with one of the latest sets shown for rust-indieweb but without src:rust-swc-core. They seem to have passed, so that seems to have been a good call. I applied the same set for the autopkgtest of rus-swc-core on s390x which also just passed. I think things are no longer blocked on autopkgtest failures.

It's no longer blocked on autopkgtests but it still seems to be having problems migrating.

I noticed too.

python-libcst was scheduled for autoremoval over a week ago, but doesn't seem to be actually getting removed. It looks like there is an inconsitency between what

britney thinks will be broken and what autoremovals wants to remove.

trying: -python-libcst
skipped: -python-libcst (0, 109, 22)
     got: 43+0: a-19:a-8:a-0:i-15:p-0:r-0:s-1
     * amd64: isort-faux-build-depends, orange-canvas-core-faux-build-depends, 
python-hypothesmith-faux-build-depends, python-trubar-faux-build-depends, 
python3-hypothesmith, python3-trubar

Right, src:isort is a key-package with a <!nocheck> annotation. So that means that removing it for real requires Release Team intervention (e.g. filing bugs against src:isort to drop the B-D, waiting some time, and then acting). It isn't so visible as usual because the list of to-be-removed-but-not-yet package list is so long (due to the r-cran-* situation), hence it wasn't on my radar yet. src:python-hypothesmith and src:python-trubar are also involved with B-D of isort. src:orange-canvas-core is listed for removal on 2026-04-16. Unfortunately autoremoval dates can get out of sync for dependency chains due to uploads after the counter starts ticking.

I *think* adding fragments/3.0.1-13, papers/49.3-2 and rust-cotp/1.9.7-2 to the easy hint will result

in a succesfull migration of rust-rand and friends.

Paul

I've never looked at the easy autohinter, so it might take some time to figure out how it works and *why* it didn't find a solution that works. Hmm, maybe I should just take a snapshot of the state, add a manual easy hint and work on solving that riddle afterwards.

OpenPGP_signature.asc
Description: OpenPGP digital signature