On Mon, 12 Dec 2016, Sandro Tosi wrote:

> On Sat, 26 Nov 2016 12:21:29 +0200 Adrian Bunk <b...@stusta.de> wrote:
> > Source: pandas
> > Version: 0.19.1-1
> > Severity: serious

> > https://buildd.debian.org/status/fetch.php?pkg=pandas&arch=i386&ver=0.19.1-1&stamp=1479504883

> the failures are now on more arch and even with 0.19.1-3 and
> 0.19.1+git174-g81a2f79-1 (sid / exp) - is there a bug upstream
> tracking this? is there something interested parties can do to help
> pandas reach testing?

well -- we have two major ways to "resolve" them:

1. fix bugs related to any exotic architecture ourselves 

1.1 upstream would love to have those fixed but overall neither it is their
priority, nor they have access to those platforms, see e.g
https://github.com/pandas-dev/pandas/issues/11282#issuecomment-181882770

NB I used to run CI for them on sparc but that box died and I had no time to
resurrect it...  may be that contributed to delayed detection of issues etc

1.2 NaT issues are numpy's issues, reported upstream and slowly being
dealt with: https://github.com/numpy/numpy/issues/8325
So probably the most impactful to help resolving this one if interested

1.3  I have no clue yet about 
SystemError: ../Objects/listobject.c:290: bad argument to internal function 
[i386,armel,...] 
https://github.com/pandas-dev/pandas/issues/14852
since can't reproduce locally and then when tried to do it earlier today
on one of the devel boxes -- it needed a reboot and never came back up
(may be did by now ;) )


2.  skip all of those failing on "interesting" platforms and I think we should
upload 0.19.1+git174-g81a2f79-1 (not 0.19.1) into sid/unstable. I have
already skipped (just across all architectures) a few...

I can run test of reverse depends meanwhile (on amd64) to see if anyone
is affected


I am happy for any kind of help!

-- 
Yaroslav O. Halchenko
Center for Open Neuroscience     http://centerforopenneuroscience.org
Dartmouth College, 419 Moore Hall, Hinman Box 6207, Hanover, NH 03755
Phone: +1 (603) 646-9834                       Fax: +1 (603) 646-1419
WWW:   http://www.linkedin.com/in/yarik        

Reply via email to