#14711: Weak references in the coercion graph
-------------------------------------+-------------------------------------
Reporter: jpflori | Owner: davidloeffler
Type: defect | Status: needs_review
Priority: critical | Milestone: sage-6.1
Component: number fields | Resolution:
Keywords: memleak, number | Merged in:
field, QuadraticField | Reviewers: Jean-Pierre Flori
Authors: Simon King | Work issues:
Report Upstream: N/A | Commit:
Branch: | 0af59ea93689cb6abb9d3fae0f1cf11f2aee5cca
u/jpflori/ticket/14711 | Stopgaps:
Dependencies: |
-------------------------------------+-------------------------------------
Comment (by nbruin):
Replying to [comment:177 tscrim]:
> > Would this really be so difficult?
I don't think it would be particularly difficult
> I was thinking it was outside of the scope of this ticket, but wasn't
sure from the discussion if something like that was needed as a
dependency. Although that's basically what we have currently (if I'm
reading the code right). To do more of the graph algorithms, it might be
worthwhile to also store the data in other formats (ex. a map of parents
to an array of their "neighbors") depending on what is useful.
Simon's tutorial example for how to write parents is a very good
illustration of the problem: Localizations of Z at a finite set of primes
S. A parent may have so many coercions into it that it's inefficient,
impractical, or impossible to list them all (the localization at any
subset T would coerce naturally). The one (almost superfluous) thing his
example is forgetting is to test if the asked ring coerces into ZZ and via
that into the localization (i.e. the depth-first search). This example
convinced me that the hybrid we have now, ugly, confusing and complicated
as it may be, might actually be the best we can do. With things like Qbar,
SR, ComplexField etc., I don't think it's possible to separate the
programmatic and the recursible parts.
--
Ticket URL: <http://trac.sagemath.org/ticket/14711#comment:178>
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.
