#9808: Upgrade numpy to 1.5.0 and scipy to 0.8
----------------------------------------------------------------------------------+
Reporter: maldun
| Owner: maldun
Type: task
| Status: needs_work
Priority: major
| Milestone: sage-4.6
Component: packages
| Keywords: numpy, scipy
Author: Stefan Reiterer, Francois Bissey, John Palmieri, David Kirkby
| Upstream: Fixed upstream, but not in a stable release.
Reviewer: Karl-Dieter Crisman, David Kirkby, Leif Leonhardy, Francois
Bissey | Merged:
Work_issues:
|
----------------------------------------------------------------------------------+
Comment(by kcrisman):
> > That's what I'm trying to figure out. Usually it gets added in the
spkg-install, but apparently that's not the case here - it's coming from
Scipy itself, [http://projects.scipy.org/numpy/ticket/1399 here] (first
noted by John).
>
> If you grep through the scipy directory, I don't think you'll find
"ppc64" anywhere. I think it's coming from numpy, and in particular the
file src/numpy/distutils/fcompiler/gnu.py. If you make the change
Yes, you're right - I realized that later last night, but had been
confused by the scipy.org address. And this is exactly the change
mentioned in all these links being bandied about.
> {{{
> #!diff
> --- gnu.py.old 2010-08-21 22:08:35.000000000 -0700
> +++ gnu.py 2010-10-08 20:59:29.000000000 -0700
> @@ -254,7 +254,7 @@
> if not sys.platform == 'darwin':
> return []
> arch_flags = []
> - for arch in ["ppc", "i686", "x86_64", "ppc64"]:
> + for arch in ["ppc", "i686", "x86_64"]:
> if _can_target(cmd, arch):
> arch_flags.extend(["-arch", arch])
> return arch_flags
> }}}
> to numpy and install it, then afterwards scipy seems to install
correctly. I don't have time this weekend to create a new spkg which
incorporates this patch, or even to figure out under what circumstances to
do it (do we have to detect DARWIN + x86? should we get rid of "ppc"
also?). If someone else wants to make a new spkg, that would be fine with
me.
I can't do that this weekend, but might be able to next week. I think
that detecting `os.uname()[2][:2]==10` (or however that works in shell
script) would be sufficient, since it seems to be a 10.6 problem with
Python (namely, that the Python we use wouldn't support ppc64 or whatever
with 10.6 - which makes sense, since 10.6 doesn't work on ppc). It
probably wouldn't hurt to remove ppc as well in that situation, but I
think that Python/Apple haven't removed that as much yet since they still
want to support universal binaries on PPC Leopard (10.5).
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/9808#comment:257>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
