Bug#1026119: transition: matplotlib

2022-12-28 Thread Sandro Tosi
> thanks! matplotlib/3.6.2-3 has been accepted in unstable

bugs have been filed:
https://udd.debian.org/cgi-bin/bts-usertags.cgi?user=debian-python%40lists.debian.org&tag=matplotlib3.6

-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
Twitter: https://twitter.com/sandrotosi



Bug#1026119: transition: matplotlib

2022-12-25 Thread Sandro Tosi
> > All in all, i think this transition is rather manageable and i'd like
> > to upload to unstable, if you're ok with it.
>
> Please go ahead.

thanks! matplotlib/3.6.2-3 has been accepted in unstable

Regards,
-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
Twitter: https://twitter.com/sandrotosi



Bug#1026119: transition: matplotlib

2022-12-25 Thread Graham Inggs
Control: tags -1 + confirmed

Hi Sandro

On Sat, 24 Dec 2022 at 19:03, Sandro Tosi  wrote:
> All in all, i think this transition is rather manageable and i'd like
> to upload to unstable, if you're ok with it.

Please go ahead.

Regards
Graham



Bug#1026119: transition: matplotlib

2022-12-24 Thread Sandro Tosi
> I'll update this report when i've uploaded a new release in experimental and
> with the results of some autopkgtests.

matplotlib/3.6.2-2 is in experimental and the results from autopkgtest
at https://qa.debian.org/excuses.php?experimental=1&package=matplotlib
are quite promising.

Looking at the errors:

* dolfin is a real bug, that needs to be correct:
https://bitbucket.org/fenics-project/dolfin/src/588d3acdb58727b6331588f9278d611534f4c93e/python/dolfin/common/plotting.py#lines-279
uses a now removed argument to `gca()`,
https://matplotlib.org/3.6.2/api/prev_api_changes/api_changes_3.6.0.html#:~:text=Keyword%20arguments%20to%20gca()

* lmfit-py, all failures seem unrelated to the new mpl, mostly missing modules

* mplcursors, seems like a valid bug, but it's unclear to me what the
failing test is trying to accomplish (catching warnings?) and so it
looks like an upstream fix (currently not available AFAICT) is neede

* poretools, not sure if this is a bug in this package or matplotlib

All in all, i think this transition is rather manageable and i'd like
to upload to unstable, if you're ok with it.

Regards,
-- 
Sandro "morph" Tosi
My website: http://sandrotosi.me/
Me at Debian: http://wiki.debian.org/SandroTosi
Twitter: https://twitter.com/sandrotosi



Bug#1026119: transition: matplotlib

2022-12-14 Thread Sandro Tosi
Package: release.debian.org
Severity: normal
User: release.debian@packages.debian.org
Usertags: transition
X-Debbugs-Cc: matplot...@packages.debian.org, mo...@debian.org
Control: affects -1 + src:matplotlib

Hello,
as suggested in #1025788, i'm filing this report to mark my intention to update
matplotlib from 3.5.2 (currently in unstable) to 3.6.2 (latest upstream
release).

I havent done any of the work needed for this task, so i'm not even sure if it's
feasable to achieve before the freeze (biggest unknown are possible NEW packages
either in mpl directly or in its b-d), but i'll give it a try.

I'll update this report when i've uploaded a new release in experimental and
with the results of some autopkgtests.

Regards,
Sandro