#14102: Nonsymmetric Macdonald Polynomials for all affine types
-------------------------------------+-------------------------------------
Reporter: bump | Owner: sage-combinat
Type: enhancement | Status: needs_review
Priority: major | Milestone: sage-pending
Component: combinatorics | Resolution:
Keywords: Nonsymmetric | Merged in:
Macdonald polynomials, days40, | Reviewers: Anne Schilling,
days45, days49, days54 | Nicolas M. Thiéry, Mark Shimozono,
Authors: Nicolas M. | Bogdan Ion
Thiéry, Anne Schilling | Work issues:
Report Upstream: N/A | Commit:
Branch: | 96ae730a0475ac99cb6c0119d58c9f659e612f1e
public/combinat/nonsymmetric_macdonald-14102| Stopgaps:
Dependencies: #4327, #14143, |
#13589, #10963, #14673, #14610, |
#14775 |
-------------------------------------+-------------------------------------
Comment (by nthiery):
Hi!
Thanks to Jean-Baptiste's mac, I have been able to reproduce and
analyze the bug reported by Volker. At the end of the day, it's a
combination of:
- CombinatorialFreeModule still computing hash values for elements
from their _repr_ (yikes)
- CombinatorialFreeModule being somewhat lax in input checking,
leading to things like mixes of int's and Integer's in the support
of elements.
- The root system code introducing those int's and Integer's in the
first place
- sort giving different results for mixed lists containing strings and
integers, depending on whether some of those integers are int's or
Integer's
Thus the odds of things going wrong were low, which is why this
remained unnoticed for so long. Why did this only appear on certain
platforms?
Well, the result of
{{{
sage: sorted([int(0), 'delta', 1])
[0, 1, 'delta']
}}}
is not the same on my Linux and Jean-Baptiste's Mac.
Probably, with some other doctests, the problem would have instead
appeared on my machine and not on others.
I assume that (finally!) implementing a proper hash function for
CombinatorialFreeModule elements should do the job. But that will be
for another day. Yawn ....
In any cases, no worry to have: as I expected this has nothing to do
with #10963.
Cheers,
Nicolas
--
Ticket URL: <http://trac.sagemath.org/ticket/14102#comment:58>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" 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-trac.
For more options, visit https://groups.google.com/d/optout.
