On 03/01/2012 04:00 PM, Julien Puydt wrote:
I'm sorry, but I still fail to see how situation :
sage contains all its deps (even if they're already there)
will be any better for subtle bugs than situation :
sage has the same deps, but doesn't manage them itself more than by
declaring them (and having the system packaging satisfy them).
This will probably just be a losing battle, until we can demonstrate
that we've won. The way forward is the sage-on-gentoo/prefix approach:
spkg-install scripts are just a poor excuse for ebuilds already.
We ship and maintain half of a linux distro, with bugs popping up around
the arbitrary cut-off point (sage doesn't ship dvipng, so it's
perpetually busted on my machine thanks to sage's gd).
We don't need to do any of this work. There's absolutely no reason why
shipping maxima-5.24.0.spkg is any better than an ebuild specifying
=sci-math/maxima-5.24.0.
What (very) few packages need sage-specific patches can be applied with
USE="sage". Or, hell, even installed the old way with spkg-install.
Perhaps we can just execute the people who want to patch upstream
projects? =) I will concede this necessity in extreme circumstances.
Once sage-on-gentoo gets back upstream and into Gentoo proper, we'll be
able to do version bumps in seconds rather than weeks. At that point, no
one is going to waste their time duplicating in a week what can be done
in seconds, and sage will be properly packages from then on.
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
