Re: [sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-23 Thread Robert Miller
Vincent, > Thanks Robert for pointing this. But is the following the expected > behavior ? Of course not, that is clearly wrong. I've tracked down the bug, and posted a patch at: http://trac.sagemath.org/sage_trac/ticket/11620 This should fix the problem. ... However, when running doctests I

[sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-14 Thread Vincent D
On 14 juil, 08:55, Robert Miller wrote: > On Wed, Jul 13, 2011 at 11:47 PM, Robert Miller wrote: > > See > > >http://trac.sagemath.org/sage_trac/ticket/10549 > > ... and its dependencies. Thanks Robert for pointing this. But is the following the expected behavior ? {{{ sage: G1 = DiGraph(loops=T

Re: [sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-13 Thread Robert Miller
On Wed, Jul 13, 2011 at 11:47 PM, Robert Miller wrote: > See > > http://trac.sagemath.org/sage_trac/ticket/10549 ... and its dependencies. -- 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

Re: [sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-13 Thread Robert Miller
See http://trac.sagemath.org/sage_trac/ticket/10549 On Wed, Jul 13, 2011 at 11:01 AM, Maarten Derickx wrote: > Spending about an hour on debugging I got really close to the source of the > problem. > The main problem is that DiGraph (and probably Graph also but I did not test > this) is making

[sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-13 Thread Maarten Derickx
Spending about an hour on debugging I got really close to the source of the problem. The main problem is that DiGraph (and probably Graph also but I did not test this) is making the labels unique, even if the labels are not hashable. This is very wrong and might lead to really hard to find bugs

[sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-12 Thread Stefan
Hi all, Intrigued and worried by this report, I did the tests myself (on a plain and unpatched 4.7). See and shiver. sage: G1 = DiGraph(loops=True,multiedges=True) sage: G1.add_edges([(0, 0, 'c'), (0, 4, 'b'), (0, 5, 'c'), (0, 5, 't'), (1, 1,'c'),(1, 3,'c'), (1, 3, 't'), (1, 5, 'b'), (2, 2, 'c'

[sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-08 Thread Vincent D
On 8 juil, 16:05, Jeroen Demeyer wrote: > On 2011-07-08 15:15, Vincent D wrote: > > > Actually, the problem is even worst... the function relabel does not > > do the right job! The graph G2 obtained as above has two loops at the > > vertex 4. > > Well, to what does to vertex 4 get mapped to under

Re: [sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-08 Thread Jeroen Demeyer
On 2011-07-08 15:15, Vincent D wrote: > Actually, the problem is even worst... the function relabel does not > do the right job! The graph G2 obtained as above has two loops at the > vertex 4. Well, to what does to vertex 4 get mapped to under your "permutation"? -- To post to this group, send a

[sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-08 Thread Vincent D
> > How did you get simply "False"?  I always get (False, False) on the last > line, with sage-4.6, sage-4.6.2, sage-4.7 and sage-4.7.1.alpha3. > I also get "(False,False)" but expected "True"! I did a wrong copy- paste operation. -- To post to this group, send an email to sage-devel@googlegroup

[sage-devel] Re: bug in graph isomorphism in sage-4.7

2011-07-08 Thread Vincent D
Actually, the problem is even worst... the function relabel does not do the right job! The graph G2 obtained as above has two loops at the vertex 4. -- 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...@goog