On Feb 10, 2:45 am, "Nicolas M. Thiery" <nicolas.thi...@u-psud.fr>
wrote:
> Hi Robert,
>
> > > It's quite tied up into the category framework, so I don't see a way
> > > to merge it into Sage without the category framework. As for the
> > > category framework itself: it has been experimented with quite much
> > > during the sage-combinat workshop, and sounds reasonably stable and
> > > definitely useful. So it would make sense to start thinking about the
> > > merge. At this stage, I'd love to get comments and pre-reviews on the
> > > design from the Sage community.
>
> > Yes, let's get some non-combinat people looking at this! Could the
> > relevant patches be posted to a ticket?
>
> Will do. Though I'd recommend anyone wanting to look seriously at this
> to just run:
>
> sage -combinat install
Nicolas, now I am seeing your intention, you are just trying to get
people to use combiant :)
> The relevant patches (in .hg/patches) are:
>
> categories-nt.patch
> hopf_algebras-nt.patch
> semi-group-monoids-category-fh.patch
>
> (sorry, this is big, if not just because there are about 80 categories)
Yes, and my suspicion that hopf_algebras-nt.patch was "just" in there
to get code into Sage is wrong - it seems to be somewhat mislabeled.
Overall this is an amazing amount of hard work.
> > > There is one show-stopper though: the class pickling issue. I still
> > > did not get any answer from my post in comp.lang.python and feel
> > > pretty much stuck.
>
> > I personally have no idea on this, but pickling is something that
> > certainly needs to work well.
>
> YES. And currently my patch breaks pickling for many, if not most,
> sage parents and elements (those that are not cythoned).
Yeah, that is a major issue.
> > Is this because you're still making
> > classes on the fly to parallel all the existing classes?
>
> Indeed. There is an easy and robust fix, which is to exchange two
> lines in pickle.py and cpickle.py. But this means changing plain
> python libraries with all the debates that this should raise.
Well, we ship our own Python for a reason, but this causes trouble for
distributions. Could you open another thread and describe the problem
in detail - maybe something good will come out of it.
> Cheers,
> Nicolas
Cheers,
Michael
> --
> Nicolas M. Thiéry "Isil" <nthi...@users.sf.net>http://Nicolas.Thiery.name/
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
