[Numpy-discussion] Merging new (optional) build system in the trunk ?
david at ar.media.kyoto-u.ac.jp
Tue Nov 13 00:34:35 EST 2007
Travis E. Oliphant wrote:
> David Cournapeau wrote:
>> I would appreciate to get some comment on whether there is any
>> chance to get my numpy.scons branch merge into the trunk at some near
>> future. I feel to have reached the point where the only big thing
>> missing is more testing. I tried to test it on many platforms, but there
>> is a limit to what I can test just by myself. The branch has been
>> conceived such as by default, the current numpy.distutils is used to
>> build, and the scons-based build is used only by explicit request
>> (another setup.py), so this does not force any use now: except the work
>> related to numpyconfig (which can be tested separately, since it was
>> done in a different branch, and is only a few lines of code), everything
>> else is exactly the same than before.
> I think there is a chance. I'm generally favorable to the idea. I was
> mainly waiting for the 1.0.4 release. I think we should be able to
> merge it over now if there are no serious objections.
Ok, great. Just tell me when you want to merge it, because I did a few
things to make testing on the buildbot easier:
- I inverted the old and new setup.py (to be able to test my branch
on the buildbot).
- There is also some debugging on by default, which should be
disabled in the trunk.
This ia a 2 second change, but just to avoid any surprises for people
regularly using the trunk
Also, what is the timeline for 1.0.5 ? Ideally, once the scons work is
merged into the trunk, I would like to work on a scipy branch which uses
scons, so that both numpy and scipy next (1.0.5 and 0.7 ?) releases can
both be built using scons.
> I'm also interested in moving scipy.weave into numpy (without the blitz
> converters which will stay in scipy).
> For 1.0.5, I would also like to see aligned memory and some steps
> towards optimization using intrinsics.
Are you speaking about SIMD intrisics ?
More information about the NumPy-Discussion