#15303: Coercion discovery fails to be transitive
-------------------------------------+-------------------------------------
Reporter: nbruin | Owner:
Type: defect | Status: needs_work
Priority: major | Milestone: sage-5.13
Component: coercion | Resolution:
Keywords: | Merged in:
Authors: Simon King | Reviewers:
Report Upstream: N/A | Work issues: analyse recursion
Branch: | error
u/SimonKing/ticket/15303 | Commit:
Dependencies: #14711 | f837cbee8f81c4946a92193c73e86449c53515d9
| Stopgaps:
-------------------------------------+-------------------------------------
Comment (by SimonKing):
It seems to me that the `._coerce_costs` of a map are largely ignored.
Namely, in `.discover_coercion()`, we have a parameter `num_paths`: If
`num_paths` paths in the coercion graph are found by backtracking, then
the path with the least `._coerce_costs` is returned. But since
`num_paths=1`, in fact ''the first'' found coercion is returned.
The only exception is the user provided morphism. Here, the rule is: Even
if the user provides a coerce morphism, then one still searches a coercion
by backtracking, and in case of success the coerce costs of the user-
provided morphism is compared with the other morphism.
This just isn't fair!!! I believe that the user-provided morphism should
at least have the same priority as a morphism found by backtracking.
Two approaches, that are not mutually exclusive:
1. Amend the coerce costs, so that a simple map has less costs than a
composite map, unless there is a very good reason.
2. Let the user-provided morphism count for the number of morphisms found
in `.discover_coercion()`. Hence, if `num_paths=1` and the user provides a
morphism, then the maximal number of paths-to-be-considered is attained
and hence the user-provided morphism is returned without backtracking.
I think the second point is more important, and I will implement it now.
--
Ticket URL: <http://trac.sagemath.org/ticket/15303#comment:27>
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 unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac.
For more options, visit https://groups.google.com/groups/opt_out.
