* jupyter-ydoc: build-depends on a package that was removed because it
build-depends on a package that doesn't exist. Removed from
oracular-proposed and added to extra-removals.
https://bugs.launchpad.net/ubuntu/+source/jupyter-ydoc/+bug/2072723
* libpsml: build-depends on non-existent libxmlf90-dev in Debian and Ubuntu.
Removed from oracular-proposed and added to extra-removals.
https://bugs.launchpad.net/ubuntu/+source/libpsml/+bug/2067729
* aevol: there was a proposed patch but it was incomplete. Fully fixed the
buffer overflow problem and uploaded, to fix the autopkgtests. (This
raced a removal of aevol from oracular as a broken revdep of boost1.74;
so it had to go through NEW again.)
* swiftlang: Debian bug was just closed today but it's still building in
unstable and not available yet for syncing. Removed the package from
oracular-proposed because it requires Debian source fixes to be buildable
and releasable.
https://bugs.launchpad.net/ubuntu/+source/swiftlang/+bug/2060818
* devicexlib: Fortran porting issues, no decision yet in Debian about
removal vs fixing. Skipping.
* octave-iso2mesh, octave-brain2mesh: had to be moved to multiverse
(binaries are in contrib in Debian). Done. However, oracular was then
for some reason unhappy copying the binaries of octave-iso2mesh to the
release pocket, so I did a no-change reupload.
* node-yarnpkg -> node-corepack: build failure was fixed but there was an
outstanding autopkgtest failure because the single test in the package
requires network access. Tests that depend on network resources to
succeed are bad and flaky tests; added a badtest hint to let this through.
* pytorch-vision: FTBFS in Debian and oracular, removed from Debian testing.
Removed from oracular-proposed as well. https://bugs.debian.org/1065722
* pytorch-text: source hard-codes python 3.11. This still works in Debian
but not Ubuntu. Removed, will get fixed via Debian eventually.
* fscrypt: fails to build because of the known issue that Launchpad ppc64el
builders have less locked memory available than on other archs, due to a
kernel difference. Provided a partial patch to let tests be skipped when
locked memory can't be allocated, but one test still fails. Uploaded, and
forwarded the patch to Debian. https://bugs.debian.org/1081220.
--
Steve Langasek Give me a lever long enough and a Free OS
Debian Developer to set it on, and I can move the world.
Ubuntu Developer https://www.debian.org/
slangasek@ubuntu.com vorlon@debian.org