On Sat, Jul 5, 2014 at 1:41 AM, Nathaniel Smith <n...@pobox.com> wrote:

> On 5 Jul 2014 00:07, "Charles R Harris" <charlesr.har...@gmail.com> wrote:
> > I've filed issues with all the affected projects. Here is the current
> status.
> >
> > matplotlib -- Reported, being fixed, should be in 1.4 in a few days.
> > skimage -- Reported.
> > scikit-learn -- Reported.
> > tables -- Reported.
> > statsmodels -- Reported, fixed in master.
> > bottleneck -- Reported. IIRC, kwgoodman already knew of the changes.
> > pyfits -- Reported to astropy.
> > milk -- Reported.
> > pandas -- Reportedly fixed in master.
>
> That is a massive pile of affected projects :-(.
>
> My worry is that if all these projects we know about are broken, then how
> many other codebases that we aren't testing are also broken?
>
Same worry here. If a major change in numpy breaks ~half of the projects
that make up a typical scipy stack, that change should not be made without
at least one release that emits warnings first.

We would have caught this much earlier had we had something like
https://github.com/matthew-brett/scipy-stack-osx-testing. Maybe a good idea
to have that as a separate repo in the numpy org, add a few more projects
to it, and then regularly run numpy master (or a PR) against the latest
releases of those projects.

Ralf

> If the issues are fixed in matplotlib and pandas I'd be inclined to
> release as is with a mention of versions in the release notes.
>
> Even if it's fixed in pandas master, how long until it's in user's hands?
>
> -n
>
> > Chuck
>
> >
> >
> > _______________________________________________
> > NumPy-Discussion mailing list
> > NumPy-Discussion@scipy.org
> > http://mail.scipy.org/mailman/listinfo/numpy-discussion
> >
>
>
> _______________________________________________
> NumPy-Discussion mailing list
> NumPy-Discussion@scipy.org
> http://mail.scipy.org/mailman/listinfo/numpy-discussion
>
>
_______________________________________________
NumPy-Discussion mailing list
NumPy-Discussion@scipy.org
http://mail.scipy.org/mailman/listinfo/numpy-discussion

Reply via email to