[Numpy-discussion] How many build systems do we need?
Charles R Harris
charlesr.harris@gmail....
Sun Jan 13 12:33:51 CST 2013
On Sun, Jan 13, 2013 at 10:26 AM, David Cournapeau <cournape@gmail.com>wrote:
> On Sun, Jan 13, 2013 at 11:11 AM, Ralf Gommers <ralf.gommers@gmail.com>
> wrote:
> >
> >
> >
> > On Sun, Jan 13, 2013 at 5:25 PM, David Cournapeau <cournape@gmail.com>
> > wrote:
> >>
> >> On Sun, Jan 13, 2013 at 9:50 AM, Charles R Harris
> >> <charlesr.harris@gmail.com> wrote:
> >> >
> >> >
> >> > On Sun, Jan 13, 2013 at 6:44 AM, David Cournapeau <cournape@gmail.com
> >
> >> > wrote:
> >> >>
> >> >> On Sun, Jan 13, 2013 at 7:29 AM, Ralf Gommers <
> ralf.gommers@gmail.com>
> >> >> wrote:
> >> >> >
> >> >> >
> >> >> >
> >> >> > On Sun, Jan 13, 2013 at 6:34 AM, Charles R Harris
> >> >> > <charlesr.harris@gmail.com> wrote:
> >> >> >>
> >> >> >> Hi All,
> >> >> >>
> >> >> >> In the continuing proposal for cleanups, note that we currently
> >> >> >> support
> >> >> >> three (3!) build systems, distutils, scons, and bento. That's a
> bit
> >> >> >> much to
> >> >> >> maintain when contemplating changes, and scons and bento both have
> >> >> >> external
> >> >> >> dependencies. Can we dispense with any of these? Thoughts?
> >> >> >
> >> >> >
> >> >> > Numscons is the only one that can be dropped. I'm still using it
> >> >> > regularly,
> >> >> > but the few things it does better than bento can be easily improved
> >> >> > in
> >> >> > bento. So if removing numscons support from master saves some
> >> >> > developer
> >> >> > hours, +1 from me.
> >> >>
> >> >> I think numscons was already scheduled to be dropped in 1.7 (and next
> >> >> version of scipy as well) ? I am certainly in favor of dropping it as
> >> >> well.
> >> >
> >> >
> >> > Is bento documented anywhere or can you commit to keeping it working
> for
> >> > numpy?
> >>
> >> Both. Doc: http://bento.readthedocs.org/en/latest/ and tests
> >> (https://travis-ci.org/cournape/Bento) are continuously run/updated.
> >
> >
> > That's bento's own test suite only - we should add a numpy build with
> Bento
> > for at least Python 2.7 to the numpy Travis config.
>
> Definitely. I was merely answering Chuck's worries that bento may just
> be a one man, undocumented, bitrotted thing :)
>
>
Tsk, tsk, I would never use such extreme language ;) I put up a PR
expunging SCons support from numpy, could you take a look at it? There was
also a file related to mingw builds on windows (in cpu_id I think) that has
no bento equivalent.
Chuck
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.scipy.org/pipermail/numpy-discussion/attachments/20130113/769c2773/attachment.html
More information about the NumPy-Discussion
mailing list