Bug#426012: Anyone know why NumPy 1:1.0.3-1 conflicts with SciPy 0.5.2-7+b1?

2007-08-15 Thread Ondrej Certik
I'd rather advise to figure out why a Conflicts: is declared in the first place. It seems needed, given the changelog of python-numpy (the version gets upgraded), but undocumented there. You probably want to ask its maintainer. I don't know why it conflicts, but the attached patch makes scipy

Bug#426012: Anyone know why NumPy 1:1.0.3-1 conflicts with SciPy 0.5.2-7+b1?

2007-08-15 Thread luk
On Wed, Aug 15, 2007 at 10:50:21AM -0700, Ondrej Certik wrote: I'd rather advise to figure out why a Conflicts: is declared in the first place. It seems needed, given the changelog of python-numpy (the version gets upgraded), but undocumented there. You probably want to ask its maintainer.

Bug#426012: Anyone know why NumPy 1:1.0.3-1 conflicts with SciPy 0.5.2-7+b1?

2007-08-15 Thread Ondrej Certik
The change in debian/control is not needed, at least not the part about adding python-numpy build dependency: --- debian/control(revision 140) +++ debian/control(revision 144) -Build-Depends: debhelper (= 5.0.37.2), python-all-dev, sharutils, python-central (= 0.5),

Bug#426012: Anyone know why NumPy 1:1.0.3-1 conflicts with SciPy 0.5.2-7+b1?

2007-08-15 Thread luk
On Wed, Aug 15, 2007 at 11:26:13AM -0700, Ondrej Certik wrote: So changing python-numpy-dev (= 1:1.0.1) to python-numpy-dev (= 1:1.0.3-1) should suffice for debian/control. Did you try it in pbuilder? I did and it fails when importing from numpy.distutils.core import setup, because that is

Bug#426012: Anyone know why NumPy 1:1.0.3-1 conflicts with SciPy 0.5.2-7+b1?

2007-08-15 Thread Ondrej Certik
That would only mean that there is a missing dependency between python-numpy-dev and python-numpy AFAICS... I agree. But I prefer to have scipy working now. When numpy maintainer fixes this, the python-numpy dependecy could be removed. Ondrej -- To UNSUBSCRIBE, email to [EMAIL PROTECTED]

Bug#426012: Anyone know why NumPy 1:1.0.3-1 conflicts with SciPy 0.5.2-7+b1?

2007-07-15 Thread Cyril Brulebois
Tom Epperly [EMAIL PROTECTED] (06/07/2007): I am wondering if I can just build SciPy by hand. I'd rather advise to figure out why a Conflicts: is declared in the first place. It seems needed, given the changelog of python-numpy (the version gets upgraded), but undocumented there. You probably

Bug#426012: Anyone know why NumPy 1:1.0.3-1 conflicts with SciPy 0.5.2-7+b1?

2007-07-06 Thread Tom Epperly
I am wondering if I can just build SciPy by hand. Tom -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]