Bug#950147: sagemath ftbfs with python 3.8

[Tobias Hansen]

Control: block -1 by 953633

We have a patch for this but I'm waiting for a fix of a bug in pari that causes test failures and a test timeout.

Best,
Tobias

On Wed, 29 Jan 2020 17:46:47 +0100 Tobias Hansen <thansen at debian.org> wrote:
> It does if applied first. But ok, the ipython 7 patch needed some small changes. I applied it in the branch python3.8 on salsa.
> 
> Best,
> Tobias
> 
> On 1/29/20 5:35 PM, Matthias Klose wrote:
> > On 1/29/20 5:20 PM, Tobias Hansen wrote:
> >> Hi,
> >>
> >> Arch Linux has a patch for that:
> >> https://git.archlinux.org/svntogit/community.git/tree/trunk/sagemath-python-3.8.patch?h=packages/sagemath
> >>
> >> We plan to apply it when the default Python switches to 3.8. If we apply it now, the doctests will fail with Python 3.7. I haven't yet tried building with Python 3.8 though.
> > yes, I'm aware of that patch. It doesn't apply cleanly. Could you help updating it?
> >
> >
> >> Best,
> >> Tobias
> >>
> >> On 1/29/20 3:07 PM, Matthias Klose wrote:
> >>> Package: src:sagemath
> >>> Version: 9.0-1
> >>> Severity: important
> >>> Tags: sid bullseye
> >>> User: debian-python at lists.debian.org
> >>> Usertags: python3.8
> >>>
> >>> sagemath ftbfs with python 3.8.  Not sure if you already can check that in
> >>> Debian, however in Ubuntu, there seem to be a lot of failing tests, plus the
> >>> tests failing because of too many open files (googling about that, I found some
> >>> hints, but just for MacOSX).
> >>>
> >>> Build logs at
> >>> https://launchpad.net/ubuntu/+source/sagemath/9.0-1build1
> >>>
> >>> Any help would be appreciated.
> >>>
> 
> 
>