#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):
The coerce costs of a (generic) map with exact domain and codomain are 10,
but 10000 if one of them is inexact. However, the coerce costs of a
`DefaultConvertMap` are between 100 and 400. And it seems to me that this
is not well balanced.
Namely, if the coercion model has to choose between a single default
convert map and a composition of 10 generic maps, it would take the
composition---which would of course normally be slower than a single map!
This decision has been made 5 years ago. Perhaps the coerce costs should
be refactored?
It makes sense that the coerce costs for inexact (co)domain will be high.
But a single map should certainly be better than a composite map with the
same domain and codomain.
--
Ticket URL: <http://trac.sagemath.org/ticket/15303#comment:25>
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.
