I subscribed to the Debian tracker for all packages where I introduced
delta, so I can sync them when Debian incorporates the changes or
fixes the issue in some other way, or keep merging.
# audacity
ftbfs with libavcodec59
Audacity was removed from debian testing on August 8th, 2022
Debian RC bug: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1004598
Filed Ubuntu bug:
https://bugs.launchpad.net/debian/+source/audacity/+bug/1983862
Latest upstream is 3.1.3, released in December 23rd, 2021. Does not
have ffmpeg 5.0 patches
FFmpeg 5.0 support: https://github.com/audacity/audacity/pull/3121
Resolves: https://github.com/audacity/audacity/issues/2445
Upgrading audacity to 3.1.x requires wxwidgets >= 3.1.x
Needs debian to accept wxwidgets 3.2, then we can update audacity to
3.1.x and apply patch to fix build with ffmpeg 5.0
ITP bug for wxwidgets3.2:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=919903
wxwidgets 3.2.0 was just released
(https://www.wxwidgets.org/news/2022/07/wxwidgets-3.2.0-final-release/)
and is in the debian NEW queue:
https://ftp-master.debian.org/new/wxwidgets3.2_3.2.0+dfsg-1.html
Summary: first wxwidgets 3.2, then update audacity to 3.1.3+, and then
apply the ffmpeg5 patches if they are not released yed.
Plan B: try to backport the ffmpeg5.0 patches from that github PR into
the old audacity version we have.
# pytest
I closed https://bugs.launchpad.net/ubuntu/+source/pygments/+bug/1980296
as pygments and pytest are in the archive (pytest still in
kinetic-proposed only, due to introducing regressions in a few other
packages).
pytest 7.1.2-2 induced regressions:
pygmsh pygmsh/7.1.17-2
https://github.com/nschloe/pygmsh 7.1.17 is latest upstream release
segfault during test
pygmsh -> gmsh -> (opencascade transition) -> netgen FTBFS with new opencascade
netgen uploaded and fixed via
https://launchpad.net/ubuntu/+source/netgen/6.2.2006+really6.2.1905+dfsg-5ubuntu1
rebuild of netgen and gmsh needed. gmsh rebuild done:
https://launchpad.net/ubuntu/+source/gmsh/4.8.4+ds2-2build2
netgen upstream patch to fix ftbfs with new opencascade:
https://github.com/NGSolve/netgen/commit/08993ae5e2b687e6c21d045b0a9a3f7e8722f17d
debian bug for above: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1014964
Filed ubuntu bug:
https://bugs.launchpad.net/ubuntu/+source/netgen/+bug/1983997,
uploaded package to kinetic
Forwarded to debian via Salsa PR:
https://salsa.debian.org/science-team/netgen/-/merge_requests/12
pytest DEP8 error:
unittests-3 FAIL stderr: lsof version information:
Filed https://bugs.launchpad.net/ubuntu/+source/pytest/+bug/1984135
In debian it doesn't fail because lsof is not installed
Uploaded https://launchpad.net/ubuntu/+source/pytest/7.1.2-2ubuntu1 to
ubuntu, forwarded fix to debian via Salsa PR:
https://salsa.debian.org/python-team/packages/pytest/-/merge_requests/1
# netgen
uploaded https://launchpad.net/ubuntu/+source/netgen/6.2.2006+really6.2.1905+dfsg-5ubuntu1
to fix ftbfs with new opencascade
FTBFS bug: https://bugs.launchpad.net/ubuntu/+source/netgen/+bug/1983997
Debian: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1014964
Forwarded to debian via salsa pr:
https://salsa.debian.org/science-team/netgen/-/merge_requests/12
# deal.ii
Failed test with new opencascade
It needs a no-change rebuild with new opencascade.
Done via https://launchpad.net/ubuntu/+source/deal.ii/9.4.0-1ubuntu2
Failed on riscv64, where it takes about a day and a half to build.
Currently rebuilding again. I don't have the log of the previous build
anymore, but it looks like the compiler was killed. Maybe due to a
timeout limit, or an OOM issue.
# kicad
Rebuild with new opencascade:
https://launchpad.net/ubuntu/+source/kicad/6.0.7+dfsg-1build1
# gmsh
Rebuild with new opencascade:
https://launchpad.net/ubuntu/+source/gmsh/4.8.4+ds2-2build2
# Misc retries
Retries for failed dep8 tests under psmisc:
booth arm64
drumkv1 armhf
openmsx-catapult armhd
All passed.
Some other random retries
zarr armhf
tiff/libreoffice on armhf
opentsne
# tiff component mismatch
New debian upload enabled lerc support, which is in universe and
triggered a component mismatch.
I filed https://bugs.launchpad.net/ubuntu/+source/tiff/+bug/1984327
before finding an existing MIR for it, and uploaded a delta which
removes lerc support from the ubuntu package:
https://launchpad.net/ubuntu/+source/tiff/4.4.0-4ubuntu1
I subscribed to the MIR bug at https://bugs.launchpad.net/bugs/1977551
and am ready to undo the delta once that is approved, if noone beats
me to it.
# pax-utils
lddtree switched from sh to py, it's missing python dependencies. I
filed https://bugs.launchpad.net/ubuntu/+source/pax-utils/+bug/1985093,
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1017021 and
submitted a salsa PR at
https://salsa.debian.org/debian/pax-utils/-/merge_requests/1.
I uploaded https://launchpad.net/ubuntu/+source/pax-utils/1.3.5-1ubuntu1
to kinetic, and just saw that the salsa PR was merged, so I'm ready to
sync the package again as soon as launchpad ingests the new debian
upload.
# freecad
Our delta is no more needed in the latest debian upload, so I synced
the package: https://launchpad.net/ubuntu/+source/freecad/0.20+dfsg1-4.
The newer version was needed to fix an ftbfs with opencascade 7.6.
It's currently failing dep8.
# python-etelemetry
It's failing DEP8 test, but there are bigger problems. It seems to
require a server that has been offline for quite some months now:
rig.mit.edu
First public hint about it is this github issue that was filed in Aug
2021: https://github.com/sensein/etelemetry-client/issues/35
There is a PR from Feb 2022 to allow for custom endpoints which could
perhaps be used to not depend on that hardcoded MIT server:
https://github.com/sensein/etelemetry-client/pull/39
I filed https://bugs.launchpad.net/ubuntu/+source/python-etelemetry/+bug/1986381
and https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1017070 but I
don't have a real solution.
--
ubuntu-devel mailing list
ubuntu-devel@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
