#11935: Make parent/element classes independent of base rings
---------------------------+------------------------------------------------
Reporter: SimonKing | Owner: nthiery
Type: enhancement | Status: needs_review
Priority: major | Milestone: sage-4.8
Component: categories | Keywords: parent class, element class,
category
Work_issues: | Upstream: N/A
Reviewer: | Author: Simon King
Merged: | Dependencies: #9138 #11900 #11943 #11999
---------------------------+------------------------------------------------
Comment(by SimonKing):
Hi Nicolas,
Replying to [comment:32 nthiery]:
> Replying to [comment:31 SimonKing]:
> class CategoryWithClassesFromBases(Category): # TODO: find a better
name
Perhaps `CategoryEnsemble`? The name is short and refers to the fact that
the members of an ensemble (such as all categories of algebras over
fields) are sufficiently similar that they share important features (such
as their parent class).
Or (slightly longer) `CategoryWithParameters`? Again, if you have
parameters (such as one or two base rings) then still certain important
features may not depend on the parameters. Or `ParametrizedCategory`, but
I think you prefer if the name starts with the word "Category", isn't it?
> class JoinCategory(CategoryWithClassesFromBases): ...
I think join categories should have their own parent class. Reason:
The parent/element classes of a category ensemble are pickled by a weak
form of "pickling by construction": If C1,C2,... belong to an ensemble of
categories that share their parent class, then that class will be pickled
as getattr(C,'parent_class'), where C is any member of the ensemble.
But I think that `JoinCategory` should ''not'' use that "pickling by weak
construction". Reason: For `JoinCategory`, the "pickling by weak
construction" is equivalent to the default pickling of dynamic classes
(which is: the class is determined by name, bases, and potentially a new
class such as `ParentMethods`, which is empty in the case of a
`JoinCategory`). Hence, it would be a waste of time to construct the
pickle data for the `JoinCategory` ''while they are created''.
I will certainly test both approaches. But if I remember correctly what I
did yesterday, the difference between "pickling by weak construction" or
"default pickling" for join categories was 6% in the infamous elliptic
curve benchmark.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/11935#comment:33>
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.
