rich comparison for matrices

[chapoton]

from _cmp_ to _richcmp_

[chapoton]

New commits:

​d72dcf7

rich comparison for matrices

[tscrim]

LGTM except you have a doctest failure in src/sage/categories/groups.py. So it seems like some order changed in a subtle way. It probably just is a doctest that we need to update.

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​95eacb0

trac 28694 fix doctest in Cayley table for SL(2,2)

[chapoton]

I have replaced the doctest. This change comes from a change of sorting for matrices in SL(2,F2). I think. May need to be checked ?

[chapoton]

Before

sage: sorted(SL(2,2))
[
[0 1]  [0 1]  [1 0]  [1 0]  [1 1]  [1 1]
[1 0], [1 1], [0 1], [1 1], [0 1], [1 0]
]

After

sage: sorted(SL(2,2))
[
[1 0]  [1 0]  [0 1]  [0 1]  [1 1]  [1 1]
[0 1], [1 1], [1 0], [1 1], [1 0], [0 1]
]

So indeed there is a change here.

[tscrim]

I actually like the sorting after a little better since the identity comes first and it follows the iteration order. I just don't understand why the iteration order did change. It really looks like equivalent code and tests. Unfortunately, I don't think it matches the docstring. Ah, perhaps there is a _cmp_ in the group code that gets used and gets used instead of the _richcmp_ since these are matrix group elements?

[chapoton]

Do you think one needs to investigate further, or one could proceed as it is ?

[tscrim]

It is okay to proceed as-is, it might just be something that we will again have to change later on.

[tscrim]

Let it be so.