Hi Simon and Florent,
Let me answer you both in one post, since I think your solutions might
be related. I agree with Simon that this is the offending line:
tester.assert_(isinstance(self,
self.parent().category().element_class))
Now, by adding the line
Element = DifferentialForm
to my parent class as Florent suggested, I tried to make it so that
element_class refers to DifferentialForm. However, the assert still
fails. Is there an example of a minimal code using the categories
framework that I could adapt?
I tried to find out where the problem might be by trying a standard
type of element, but these problems persist:
sage: u = polygen(QQ)
sage: isinstance(u, u.parent().category().element_class)
False
In the meantime, I see that SteenrodAlgebra has also this problem....
Any ideas?
All the best,
Joris
On 29 jul, 11:32, Florent Hivert <florent.hiv...@univ-rouen.fr> wrote:
> > everything works well (no output), but when I run F._test_elements() I
> > get an error in ... _test_category(). The precise error is listed
> > below, but the traceback tells me that the problem arose in
> > element.pyx:
>
> > if not is_extension_type(self.__class__):
> > # For usual Python classes, that should be done with
> > # standard inheritance
> > tester.assert_(isinstance(self,
> > self.parent().category().element_class))
>
> > Did I mess up with my use of the sage categories somewhere? Any help
> > would be much appreciated, and I can readily supply additional
> > information.
>
> Did you tell the parent what is the class of its elements ? Specifically when
> creating a parent with the category framework you need either to write
>
> class MyParentClass(...):
> ...
> class Element(...):
> ...
>
> Or else
>
> class MyElementClass(...):
> ...
> class MyParentClass(...):
> ...
> Element = MyElementClass
>
> Then the category framework create for you a dynamic class called
>
> MyParentClass.element_class
>
> Hope this helps,
>
> Cheers,
>
> Florent
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
