On Jun 20, 'Travis Scrimshaw' via sage-devel wrote:
That definitely is not a bug in CombinatorialFreeModule, so there shouldn't be anything changed there. It is doing what it is told: that its indexing set is in finite sets. I would say the bug is purely in the DisjointUnionEnumeratedSets not doing a full check of its member sets.
I think so too, but I thought that the behaviour for the disjoint union was intended as this. Now looking at the code it seems that it wasn't. The checks are for a couple of reasonable cases, but some corner cases make them fail on both ways.
An infinite union of empty sets is infinite: sage: F = Family(ZZ, lambda i: Set()) sage: C = DisjointUnionEnumeratedSets(F) sage: C.category() Category of facade infinite enumerated setsIn the case of finite families the following simple change seems to check correctly. But I suppose this can break something, there must've been a reason to check only the last set.
sage: C = DisjointUnionEnumeratedSets([NonNegativeIntegers(), Family([1,2,3])])
sage: C.category()
Category of facade infinite enumerated sets
$ git diff src/sage/sets/disjoint_union_enumerated_sets.py
diff --git a/src/sage/sets/disjoint_union_enumerated_sets.py
b/src/sage/sets/disjoint_union_enumerated_sets.py
index c6d35ea8bd..f5fdba7119 100644
--- a/src/sage/sets/disjoint_union_enumerated_sets.py
+++ b/src/sage/sets/disjoint_union_enumerated_sets.py
@@ -302,7 +302,7 @@ class DisjointUnionEnumeratedSets(UniqueRepresentation,
Parent):
# try to guess if the result is infinite or not.
if self._family in InfiniteEnumeratedSets():
category = InfiniteEnumeratedSets()
- elif self._family.last().cardinality() == Infinity:
+ elif any(x.cardinality() == Infinity for x in self._family):
category = InfiniteEnumeratedSets()
else:
category = FiniteEnumeratedSets()
I can't think of a reasonable way of checking against the infinite union of
empty sets.
Best,
R.
Best,
Travis
On Saturday, June 20, 2020 at 8:43:06 PM UTC+10 hel...@potuz.net wrote:
On Jun 20, Markus Wageringel wrote:
>I think the problem here is that enumerating over a disjoint union of an
>infinite set and a finite set will never reach the elements in the finite
set.
>There is this comment in the documentation of DisjointUnionEnumeratedSets
>
>
> Possible extensions: the current enumeration order is not suitable
> for unions of infinite enumerated sets (except possibly for the
> last one). One could add options to specify alternative enumeration
> orders (anti-diagonal, round robin, ...) to handle this case.
>
>
>so the workaround is to first enumerate over the finite set and then the
>infinite set, as the last enumerated set in the union will determine
whether
>the disjoint union is finite or not:
>
>sage: DisjointUnionEnumeratedSets([NonNegativeIntegers(), Family('x')]) in
>FiniteSets()
>True
>sage: DisjointUnionEnumeratedSets([Family('x'), NonNegativeIntegers()]) in
>FiniteSets()
>False
>
>
Nice! I hadn't seen that. I do think however that the check in
CombinatorialFreeModule should be changed or at least that taking the
disjoint union should print a warning that that this set is considered to
be finite.
Best,
R.
>--
>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 [1]sage-devel+...@googlegroups.com.
>To view this discussion on the web visit [2][1]https://groups.google.com/d
/msgid/
>sage-devel/aeb3c3ba-c804-428d-9119-2629950be6cbo%[2]40googlegroups.com.
>
>References:
>
>[1] mailto:sage-devel+...@googlegroups.com
>[2] [3]https://groups.google.com/d/msgid/sage-devel/
aeb3c3ba-c804-428d-9119-2629950be6cbo%40googlegroups.com?utm_medium=email&
utm_source=footer
--
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 [4]sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit [5]https://groups.google.com/d/msgid/
sage-devel/9460d061-39db-44b9-90e7-5d348e706bben%40googlegroups.com.
References:
[1] https://groups.google.com/d/msgid/
[2] http://40googlegroups.com/
[3]
https://groups.google.com/d/msgid/sage-devel/aeb3c3ba-c804-428d-9119-2629950be6cbo%40googlegroups.com?utm_medium=email&utm_source=footer
[4] mailto:sage-devel+unsubscr...@googlegroups.com
[5]
https://groups.google.com/d/msgid/sage-devel/9460d061-39db-44b9-90e7-5d348e706bben%40googlegroups.com?utm_medium=email&utm_source=footer
-- 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 sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/20200621141057.GA914605%40vertex.
signature.asc
Description: PGP signature
