On Nov 22, 2:46 pm, Florent Hivert <florent.hiv...@lri.fr> wrote:
>       Hi there,
>
> I want to report a segfault in DiGraph.automorphism_group.
>
> > On my machine the following code creates a segfault if
> > finite_semigroup-nt.patch is applied and works if not.
>
> >     sage: G = DiGraph()
> >     sage: G.add_vertices([1, 6, 9])
> >     sage: G.add_edges([(9, 1, None), (9, 6, None)])
> >     sage: G.automorphism_group()
> >     ****BBBBOOOOMMMM****
>
> > Can someone confirm that the segfault happen on other machine ?
>
> > For Nicolas: could it be related to the recent work on categories ?
>
> We (Nicolas & myself) tracked back the segfault and found that it was
> triggered by the category system but it should have nothing to do with it. The
> problem seems to been raised by an early import of DiGraph.
>
> Here is a simple way to trigger the segfault:
>
> Starts from vanilla sage-4.7.2 together with the following patch:
>
> """
> # HG changeset patch
> # Parent 9e29a3d84c48c399daaf3920bcb8b17273a0e876
> diff --git a/sage/categories/all.py b/sage/categories/all.py
> --- a/sage/categories/all.py
> +++ b/sage/categories/all.py
> @@ -39,6 +39,8 @@ from finite_monoids import FiniteMonoids
>  from finite_groups import FiniteGroups
>  from finite_permutation_groups import FinitePermutationGroups
>
> +from sage.graphs.digraph import DiGraph
> +
>  # fields
>  from number_fields import NumberFields
> """
>
> Then,
>
>     sage: DiGraph().automorphism_group()
>
> Kaboom !!!
>
> I don't think I'll have time to investigate this further. Should I open a
> ticket ?
>
> Cheers,
>
> Florent

Hi Florent,
I tested your code with the patch and without - it worked both times:
Permutation Group with generators [(1,3)]

But maybe this is related, I had also problems with a crash in module
digraph.py and have created a ticket here
http://trac.sagemath.org/sage_trac/ticket/12054

cheers
emil

-- 
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to 
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Reply via email to