[Debian-science-sagemath] Upgrade Cython in Debian / sage 8.2 status

[Tobias Hansen]

Yes, I already uploaded it to see if it gets rid of the 600 additional failing tests, which it does. Thanks!

The number is more or less arbitrary, and everything that is an improvement is good. The 3 month old sagemath 8.1 in unstable is more or less broken (for sure I don't expect it to build anymore, due to at least networkx). And we should free up experimental to prepare for sagemath 8.3.

The only reason I didn't already upload everything are the build failures of flint-arb. And I don't have much time at the moment, and go on vacation soon. I could also just upload everything to unstable anyway.

Best,
Tobias

On 06/25/2018 04:38 PM, Julian Rüth wrote:
> Hello Tobias,
>
> I noticed that you have already picked up my networkx patch from Sage's
> trac. I created
> https://salsa.debian.org/science-team/sagemath/merge_requests/4 with the
> presumably final version that should fix another 6 doctests.
>
> Out of curiosity, is there an explicit limit before this can go into
> unstable? The wiki page says "At present, we have a threshold of 40
> failing doctests (out of ~300K total), below which we call it a
> `successful build`". Is this number still current (some of the columns
> in https://people.debian.org/~thansen/sage-test-status.html show higher
> numbers.)
>
> julian
>
> * Tobias Hansen <thansen at debian.org> [2018-06-21 14:52:41 +0200]:
>
>> On 06/21/2018 02:46 PM, Julian Rüth wrote:
>>> * Tobias Hansen <thansen at debian.org> [2018-06-21 14:02:14 +0200]:
>>>> thanks for the help, it is really appreciated!
>>>>
>>>> Somehow we are not used to these pull requests yet. I didn't notice
>>>> the one for sagemath until now. Usually I'm notified by mail about
>>>> most things that are going on because I'm in the tracker group [1],
>>>> but not about pull requests.
>>> Ok. I was not aware that no notifications are sent out for MRs. I'll
>>> ping here or on the bug tracker in the future then.
>>>
>>>> What is stopping me from uploading sagemath 8.2 to unstable at the
>>>> moment is mostly that flint-arb 2.12 is failing to build on half the
>>>> release architectures [2] and that the number of failing tests is over
>>>> 600 at the moment on most architectures [3].
>>> I would like to help with these. Is there a good way to debug these
>>> architectures? Or should I just try to install Debian in QEMU?
>>>
>>> Some 200 tests seem to be failing because of the wrong networkx version
>>> I think (2.* instead of 1.11.) Should I create an MR for this even
>>> though I could not test it? Is there some sort of CI setup that I can
>>> use for my merge requests or would I have to roll my own at the moment?
>>>
>>> Thanks for your guidance with this!
>>>
>>> julian
>> Actually, I think most of these new failures are not architecture specific. It's just that I uploaded on June 6 and most architectures were waiting for dependencies to build for a few days. In the meantime other packages like networkx were uploaded, leading to the new failures. So you can just test on amd64 and for the most part we see what happens on the other architectures when we upload. Normally there are just a few additional failures here or there.
>>
>>>> [1] https://tracker.debian.org/teams/debian-sagemath/
>>>> [2] https://buildd.debian.org/status/package.php?p=flint-arb&suite=experimental
>>>> [3] https://people.debian.org/~thansen/sage-test-status-experimental.html
>>>>
>>>> On 06/21/2018 01:53 PM, Julian Rüth wrote:
>>>>> Hello Tobias,
>>>>>
>>>>> I started to work on the failing doctests.
>>>>>
>>>>> * Tobias Hansen <thansen at debian.org> [2018-06-07 00:35:40 +0200]:
>>>>>> […]
>>>>>>
>>>>>> I did a first upload of sagemath 8.2 to experimental today. There are
>>>>>> 65 failing doctests and I could really need some help to get this
>>>>>> number down. If you have any questions on how to get set up for
>>>>>> building sage let me know. Apart from the Wiki page there is also the
>>>>>> README.Debian file with some tips for debugging doctests.
>>>>> I created
>>>>> https://salsa.debian.org/science-team/sagemath/merge_requests/1 which
>>>>> fixes 8 doctests for me and
>>>>> https://salsa.debian.org/science-team/sympow/merge_requests/1 for sympow
>>>>> which fixes a few more doctests that I think used to fail already in
>>>>> 8.1.
>>>>>
>>>>> Jerome, would you mind to have a look at my sympow patch?
>>>>>
>>>>> cheers,
>>>>>
>>>>> julian
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Debian-science-sagemath mailing list
>>>>> Debian-science-sagemath at alioth-lists.debian.net
>>>>> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-science-sagemath
>>>> _______________________________________________
>>>> Debian-science-sagemath mailing list
>>>> Debian-science-sagemath at alioth-lists.debian.net
>>>> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-science-sagemath
>>>
>>> _______________________________________________
>>> Debian-science-sagemath mailing list
>>> Debian-science-sagemath at alioth-lists.debian.net
>>> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-science-sagemath
>>
>> _______________________________________________
>> Debian-science-sagemath mailing list
>> Debian-science-sagemath at alioth-lists.debian.net
>> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-science-sagemath
>
>
> _______________________________________________
> Debian-science-sagemath mailing list
> Debian-science-sagemath at alioth-lists.debian.net
> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-science-sagemath


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/debian-science-sagemath/attachments/20180625/a8692bc0/attachment.html>