On Fri, Mar 30, 2012 at 03:16:15PM -0700, Anne Schilling wrote:
> To me it feels more natural to have the output on the basis of simple roots,
> rather than the ambient space.
> Is there now a simple way to convert a vector in the ambient space
> to the root basis?
Not yet implemented. Which is a
Anne,
> Is there now a simple way to convert
> a vector in the ambient space to the root basis?
Longish answer:
The "domain" of the Weyl group, the space on which it is defined to act,
can be at least one of the following:
1. Ambient space
2. Root lattice
3. Coroot lattice
4. Weight lattice
5.
Hi all,
Nicolas sez:
>>I have the feeling that it would be more natural if a method of
>> W (or of it's elements) returning some roots would return then as
>> elements of L.
Anne sez:
> To me it feels more natural to have the output on the basis of simple
> roots, rather than the ambient space.
> On Sat, Mar 24, 2012 at 08:52:04PM -0400, msh...@math.vt.edu wrote:
>> Now you should get
>>
>> sage: W=WeylGroup(['A',2])
>> sage: w=W.from_reduced_word([1,2,1])
>> sage: w.inversions()
>> [alpha[1], alpha[1] + alpha[2], alpha[2]]
>
> Sorry for the slow answer. Here W is a Weyl group implemente
Hi Marc,
On Sat, Mar 24, 2012 at 08:52:04PM -0400, msh...@math.vt.edu wrote:
> Now you should get
>
> sage: W=WeylGroup(['A',2])
> sage: w=W.from_reduced_word([1,2,1])
> sage: w.inversions()
> [alpha[1], alpha[1] + alpha[2], alpha[2]]
Sorry for the slow answer. Here W is a Weyl group imp
Ah, shoot; I had forgotten but there is currently a shortcoming with
cached methods which make them rather incompatible with super calls
in subclasses:
class A(object):
@cached_method
def f(self): print "I am A.f"
class B(A):
def f(self):
super(B, sel
On Thu, Mar 29, 2012 at 07:50:31PM -0400, msh...@math.vt.edu wrote:
> I need to call the Coxeter group method something different
> like reflection_inversions (and the solution becomes trivial)
alternative name: inversions_as_reflections
> or someone needs to tell me how to tell python not to run
On Thu, Mar 29, 2012 at 09:04:41PM -0400, msh...@math.vt.edu wrote:
> Is the following a decent way to determine whether "element" is an element
> of a Weyl group of a given Cartan type?
>
> hasattr(element, "domain") and hasattr(element.domain(), "cartan_type")
> and element.domain().cartan_type(
Christian and Nicolas,
I pushed trac_12774-coxeter-ms.patch.
I moved it way up the queue and
had no problems applying the rest of the queue.
I temporarily broke some rules for putting the patch that high on the queue.
E.g., I have a trac ticket but haven't put a version of the patch
on trac just
Christian and Nicolas,
Is the following a decent way to determine whether "element" is an element
of a Weyl group of a given Cartan type?
hasattr(element, "domain") and hasattr(element.domain(), "cartan_type")
and element.domain().cartan_type() == cartan_type
E.g. is this likely to change or are
Arrgh, my previous question obviously didn't make sense.
I need to call the Coxeter group method something different
like reflection_inversions (and the solution becomes trivial)
or someone needs to tell me how to tell python not to run over
the Coxeter group method (so the method is actually avai
Hi Mark,
If I understand correctly, you can you getattr, see
getattr(object, name[, default]) -> value
Get a named attribute from an object; getattr(x, 'y') is equivalent to
x.y. When a default argument is given, it is returned when the
attribute doesn't exist; without it, an exc
Nicolas,
But I want the Weyl group version to use the output of the
Coxeter version, which will be cached,
and then make the (co)root variants if
needed. How do I do that syntactically without
running over the Coxeter version?
--Mark
>> Since WeylGroups has CoxeterGroups as a super category, to
On Thu, Mar 29, 2012 at 09:03:59PM +0200, Christian Stump wrote:
> Since WeylGroups has CoxeterGroups as a super category, to me it seems
> be perfectly reasonable to name both methods "inversions", The one
> from WeylGroups then overwrite the one from CoxeterGroups
Indeed.
As for usual inheritan
Christian,
> Since WeylGroups has CoxeterGroups as a super category, to me it seems
> be perfectly reasonable to name both methods "inversions", The one
> from WeylGroups then overwrite the one from CoxeterGroups (or does it
> happen to be the other way round, Nicolas?). To avoid confusion, maybe
> Indeed, in trac_12774-coxeter-ms.patch, there is already a Coxeter group
> element method which returns inversions in reflection form:
>
> sage: W = WeylGroup(['A',2],prefix="s")
> sage: w = W.from_reduced_word([1,2,1])
> sage: w.length_decreasing_reflections_right()
> [s1, s1*s2*s1, s2]
>
> If e
Hi all,
> Jean Michel wrote:
>
>> The list of inversions, in my view, should preferably be a list
>> of reflections (which does not need the existence of roots and makes
>> sense for abstract Coxeter groups).
Indeed, in trac_12774-coxeter-ms.patch, there is already a Coxeter group
elemen
> In other words, something like this:
>
> sage: w.inversions()
> [alpha[1], alpha[1] + alpha[2], alpha[2]]
>
> sage: w.inversions(reflections=True)
> [s1,s1*s2*s1,s2]
+1
what was again the point of having a list rather than a set?
Christian
--
You received this message because you are subscri
> Frederic,
>
> I pushed a patch (coxeter_ms.patch) to the sage-combinat server.
>
> Now you should get
>
> sage: W=WeylGroup(['A',2])
> sage: w=W.from_reduced_word([1,2,1])
> sage: w.inversions()
> [alpha[1], alpha[1] + alpha[2], alpha[2]]
>
> which is much nicer.
>
> --Mark
Jean Michel wro
Frederic,
I pushed a patch (coxeter_ms.patch) to the sage-combinat server.
Now you should get
sage: W=WeylGroup(['A',2])
sage: w=W.from_reduced_word([1,2,1])
sage: w.inversions()
[alpha[1], alpha[1] + alpha[2], alpha[2]]
which is much nicer.
--Mark
> Trying to use inversions for elements of W
On Fri, Mar 23, 2012 at 01:14:00AM -0700, Frédéric Chapoton wrote:
>Trying to use inversions for elements of Weyl groups, I found that it
>exists, but does not work, and is badly documented and tested. In which
>patch is this located ?
sage: w.inversions??
Tells it is in sage/
Hello,
Trying to use inversions for elements of Weyl groups, I found that it
exists, but does not work, and is badly documented and tested. In which
patch is this located ?
Frederic
sage: W=WeylGroup(['A',3])
sage: w=W.from_reduced_word([1,2,1])
sage: w.inversions()
AttributeError: 'WeylGroup
22 matches
Mail list logo