On Mon, Feb 13, 2012 at 07:46:31PM +0100, Jean MICHEL wrote:
> -A naive (in the sense of straightforward) port of the needed parts of (a
> rather old version of) Chevie to plain Python. I do not know if much of it
> is easily usable for a Sage port.
It should be trivial to import. It might need a bit of refactoring to
interact nicely with other features of Sage, and to follow the Sage
guidelines for documentation, tests, ...
> By the way, I have a question: what is a Sage port? I ask this because it
> is my plan to eventually work in Sage (when GAP3 becomes too obsolete). But
> I must confess that I have not yet started on this plan since I want to
> finish some large projects on Chevie and I think it is not yet time for me
> to discard GAP3. I understand that there are several levels of porting:
>
> - The first is that Chevie can be used in Sage through the interface
> to GAP3, thanks to Saliola.
>
> -Next, one could map each Chevie type of object to an appropriate Sage
> class or category; since the design in Sage could be quite different this
> could be non trivial. Since I have not yet learned properly Sage this is
> the hardest for me, and is where I am stuck.
>
> -One could then write the high-level code of Chevie in Sage, calling GAP3 or
> GAP4 for the low-level stuff.
>
> -Or one could port all the code of Chevie in Sage. This requires in
> particular efficient cyclotomic numbers (mostly done, if I understand) and
> efficient permutations (anything done there?). This second question begs
> the question wether one would also port much of the permutation group
> library of GAP3/4 to Sage.
That's a tough question, especially since it's a lot about community:
who is likely to contribute and with which interest in mind? At the
moment, I foresee more interest from people from, let's say,
``combinatorial representation theory'' rather than ``group theory''
(whatever that means).
At the moment, I guess I would lean toward the following:
- Progressively migrate the "database part" of Chevie to Sage or
plain Python (with any combination of the GAP3 interface and
Meinholf's work as intermediate steps); Jean: a quick review on
what would need to be completed/updated on that side in PyCox
w.r.t. the latest version of Chevie would be useful.
- Use Sage for permutation arithmetic (possibly improving it if needed)
- Polish the integration of Coxeter 3 (for an alternative super fast
implementation of Coxeter group element arithmetic).
- Delegate all standard group theory calculations to GAP4.
- Implement in Sage the higher level algorithms, especially those of
«algebraic combinatorics» flavor (by higher level, I mean those
that require a combination of the tools provided by the above).
> Even if I do not do it myself, I would be happy to help anyone
> trying to do any of the above.
Thanks! Your involvement on this list has already proven super useful.
Btw: would you have some calculation suggestions for benchmarking the
universal cyclotomic field?
Cheers,
Nicolas
--
Nicolas M. Thiéry "Isil" <nthi...@users.sf.net>
http://Nicolas.Thiery.name/
--
You received this message because you are subscribed to the Google Groups
"sage-combinat-devel" group.
To post to this group, send email to sage-combinat-devel@googlegroups.com.
To unsubscribe from this group, send email to
sage-combinat-devel+unsubscr...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/sage-combinat-devel?hl=en.
