#13204: Element construction for CrystalOfTableaux should be robust versus
int's
----------------------------------+-----------------------------------------
Reporter: nthiery | Owner: sage-combinat
Type: defect | Status: needs_info
Priority: major | Milestone: sage-5.6
Component: combinatorics | Resolution:
Keywords: | Work issues:
Report Upstream: N/A | Reviewers: Nicolas M. Thiéry
Authors: Anne Schilling | Merged in:
Dependencies: | Stopgaps:
----------------------------------+-----------------------------------------
Changes (by tscrim):
* status: new => needs_info
Comment:
Another (more subtle) bug which comes from the same issue:
{{{
sage: C = KirillovReshetikhinCrystal(['A',int(4),1], 1,1)
sage: C[0].e0()
# Boom
AttributeError: 'int' object has no attribute 'epsilon'
}}}
The problem seems to be in that elements are created from
`C.cartan_type().n` somewhere (I didn't track this down).
I've uploaded my fix for this which automatically calls the
`CrystalOfLetters` on each element in the list (which should just type
check and return the element back if it already is a letter). There is a
slight performance hit (see below) for doing this rather than just
checking the leading element. Also this takes care of input like:
{{{
sage: t = T(list=[T.letters(3),1,4,2])
sage: t = T(list=[CrystalOfLetters(['D',6])(3),1,4,2])
}}}
(Note CrystalOfLetters checks (well lack thereof) are not robust enough to
catch the second input as an error, but that's a separate issue.) Really
the question boils down to how pathological do we think the users might be
with this versus the speed hit? If people think this is a good fix, this
is ready to review.
Thanks,[[BR]] Travis
Performance data before patch:
{{{
sage: B = KirillovReshetikhinCrystal(['A',4,1], 3,3)
sage: %timeit L = [x for x in B]
5 loops, best of 3: 584 ms per loop
sage: %timeit L = [x for x in B]
5 loops, best of 3: 550 ms per loop
sage: %prun L = [x for x in B]
398355 function calls in 4.147 seconds
Ordered by: internal time
ncalls tottime percall cumtime percall filename:lineno(function)
45228 0.668 0.000 1.602 0.000 crystals.py:169(index_set)
45053 0.658 0.000 2.433 0.000 crystals.py:727(index_set)
45228 0.443 0.000 0.680 0.000
cartan_type.py:1822(index_set)
...
}}}
after patch:
{{{
sage: B = KirillovReshetikhinCrystal(['A',4,1], 3,3)
sage: %timeit L = [x for x in B]
5 loops, best of 3: 642 ms per loop
sage: %timeit L = [x for x in B]
5 loops, best of 3: 633 ms per loop
sage: %prun L = [x for x in B]
426299 function calls in 5.135 seconds
Ordered by: internal time
ncalls tottime percall cumtime percall filename:lineno(function)
45228 0.791 0.000 1.942 0.000 crystals.py:167(index_set)
45053 0.728 0.000 2.874 0.000 crystals.py:726(index_set)
45228 0.572 0.000 0.838 0.000
cartan_type.py:1822(index_set)
...
}}}
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/13204#comment:1>
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 post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
