PermutationGroup.cardinality is sometimes an int

[videlec]

sage: P = PermutationGroup(['(1,2)','(1,3)'])
sage: P.cardinality()
6
sage: type(P.cardinality())
<type 'int'>

This prevents using the .is_one() method or any other Integer specific method.

The problem comes from the shortcuts implemented in the _order method.

Component: group theory

Author: Vincent Delecroix

Branch/Commit: e9795a0

Reviewer: Travis Scrimshaw

Issue created by migration from https://trac.sagemath.org/ticket/23778

[videlec]

Commit: f441434

[videlec]

New commits:

f441434

23778: fix cardinality for PermutationGroup

[videlec]

Branch: u/vdelecroix/23778

[tscrim]

comment:2

Did you also check to see how

        cycle_tuples = []
        for g in gens:
            temp = g.cycle_tuples()
            if len(temp) != 2:
                return None
            cycle_tuples.append(temp)

compares timing-wise? At least it seems like it would be faster...

[videlec]

comment:3

I don't believe that cycle_tuples would be the culprit, but your version is cleaner.

[sagetrac-git]

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

1907f8b

23778: simpler loop

[sagetrac-git]

Changed commit from f441434 to 1907f8b

[tscrim]

Reviewer: Travis Scrimshaw

[tscrim]

comment:5

One failing test according to the patchbot:

sage -t --long src/sage/sets/set_from_iterator.py  # 1 doctest failed

[sagetrac-git]

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

e9795a0

23778: fix a doctest in set_from_iterator.py

[sagetrac-git]

Changed commit from 1907f8b to e9795a0

[videlec]

comment:8

Indeed... the source code had changed.

[videlec]

comment:9

Fixed!

[tscrim]

comment:11

Thanks.

[vbraun]

Changed branch from u/vdelecroix/23778 to e9795a0