On Monday, 15 June 2015 00:57:44 UTC+1, vdelecroix wrote:
>
> All right. I recompiled Sage without chomp. Now, beyond the binary tree
> issues with bliss (#18698), I have failing doctests related to
> gap_database (all of them are marked with #option - database_gap) in
>
> - sage/groups/perm_gps/permgroup_named.py
> - sage/tests/gap_packages.py
> - sage/groups/perm_gps/permgroup_named.py
>
> Looks to a failure similar to the one reported by
> Emanuel Charpentier on sage-release. The gap packages fail to be imported
>
> sage: import sage.tests.gap_packages
> sage: sage.tests.gap_packages.test_packages(['atlasrep'])
> Status Package GAP Output
> +---------+----------+------------+
> Failure atlasrep fail
> sage: sage.tests.gap_packages.test_packages(['tomlib'])
> Status Package GAP Output
> +---------+---------+------------+
> Failure tomlib fail
>
>
for these you (also) need database_gap
The error message in
sage: gap.load_package("tomlib")
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
<ipython-input-1-cc487c826f42> in <module>()
----> 1 gap.load_package("tomlib")
/home/dima/software/sage/local/lib/python2.7/site-packages/sage/interfaces/gap.pyc
in load_package(self, pkg, verbose)
504 if x == 'fail':
505 raise RuntimeError("Error loading Gap package
"+str(pkg)+". "+
--> 506 "You may want to install the
gap_packages SPKG.")
507
508 def eval(self, x, newlines=False, strip=True, split_lines=True,
**kwds):
RuntimeError: Error loading Gap package tomlib. You may want to install the
gap_packages SPKG.
After installing database_gap, atlasrep loads, too
Let me see how to fix this.
Vincent
>
> On 14/06/15 10:32, Frédéric Chapoton wrote:
> > according to #16364, chomp could lead to this kind of things. But it is
> not
> > in your list !!
> >
> > Le dimanche 14 juin 2015 10:29:33 UTC+2, Frédéric Chapoton a écrit :
> >>
> >> For binary trees, I have made http://trac.sagemath.org/ticket/18698
> that
> >> needs review.
> >>
> >> Le dimanche 14 juin 2015 10:19:22 UTC+2, Frédéric Chapoton a écrit :
> >>>
> >>> Indeed, the failure for binary trees should come from canonical
> labeling.
> >>> It would be enough to
> >>> change the doctest so that they are no longer sensitive to that.
> >>>
> >>> I am no longer that sure that TOPCOM is the other problem. What other
> >>> package is delaing with homology ? chomp ?
> >>>
> >>> Frederic
> >>>
> >>> Le dimanche 14 juin 2015 10:09:23 UTC+2, Nathann Cohen a écrit :
> >>>>
> >>>>> I would suspect that topcom should be responsible for the homology
> >>>> errors.
> >>>>>
> >>>>> No idea for what happens to binary trees. They seem to be reversed
> >>>> (left-right symmetry ?)
> >>>>
> >>>> That could be 'bliss'. It becomes the default algorithm to compute
> >>>> automorphism groups of graphs and canonical representatives. Those
> >>>> canonical representatives will be different from the previous ones,
> >>>> though.
> >>>>
> >>>> Nathann
> >>>>
> >>>
> >
>
--
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 [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-devel.
For more options, visit https://groups.google.com/d/optout.
