Re: Semidirect product command as permgroup method or global function?

[Volker Braun]

If in doubt I would always prefer the method. Its OOP and more discoverable, for starters.

On Friday, August 3, 2012 6:21:24 PM UTC-4, Khalasz wrote:

I'm a student of Rob Beezer. I just developed a patch for computing the semi-direct product of two permutation groups given a homomorphism from one into the automorphism group of the other. I'm wondering if people have thoughts on whether it would be better implemented as a global function, a la direct_product_permgroups(), or as a method, like direct_product(). I wrote it as a method, RAB prefers the global function, but suggested I see what the larger community has to say before I make my final decision.

[Simon King]

Hi!

On 2012-08-03, Volker Braun <vbrau...@gmail.com> wrote:
> If in doubt I would always prefer the method. Its OOP and more
> discoverable, for starters.

> --

I'd prefer a method as well. Actually, a method for all groups (perhaps
defined as a parent method in sage/categories/groups.py?) that defaults
to raise a NotImplementedError and is replaced by something more
meaningful in special cases (such as in the case of permutation groups).

Cheers,
Simon