Bug in .subgroup?

[B. L.]

Hello,
to me it seems that with 10.3 there might be a bug in ".subgroup", see example below. The subgroup cardinality is wrong and the equality test of the group and the subgroup generated by all group elements yields "False". In previous versionls of SAGE this worked as expected.
Can anybody help?
Thanks and kind regards,
Barbara


sage: g = AbelianGroup([4, 4])
sage: gg = g.subgroup(g.list())
sage: g
Multiplicative Abelian group isomorphic to C4 x C4
sage: gg
Multiplicative Abelian subgroup isomorphic to C4 x C4 generated by {f1, f1^2, f1^3, f0, f0*f1, f0*f1^2, f0*f1^3, f0^2, f0^2*f1, f0^2*f1^2, f0^2*f1^3, f0^3, f0^3*f1, f0^3*f1^2, f0^3*f1^3}
sage: g.cardinality()
16
sage: gg.cardinality()
134217728
sage: g==gg
False

[Dima Pasechnik]

Yes, it's surely a bug; this example gives |gg|=2^27 (wow!), or a smaller one:

sage: g=AbelianGroup([2,2])
sage: gg=g.subgroup(g.list())
sage: gg.cardinality().factor() # what?
2^3
sage: g.cardinality().factor()
2^2

Maybe related to https://github.com/sagemath/sage/pull/36986 ?

> --
> You received this message because you are subscribed to the Google Groups "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/55ecbb00-3927-4339-8cb3-f68347358446n%40googlegroups.com.

[Dima Pasechnik]

On Thu, Apr 4, 2024 at 11:07 AM Dima Pasechnik <dim...@gmail.com> wrote:
>
> Yes, it's surely a bug; this example gives |gg|=2^27 (wow!), or a smaller one:
>
> sage: g=AbelianGroup([2,2])
> sage: gg=g.subgroup(g.list())
> sage: gg.cardinality().factor() # what?
> 2^3
> sage: g.cardinality().factor()
> 2^2
>
> Maybe related to https://github.com/sagemath/sage/pull/36986 ?

yes, indeed, it came from https://github.com/sagemath/sage/pull/36986
I've opened https://github.com/sagemath/sage/issues/37744