#14711: Memleak when creating QuadraticField
-------------------------------------------------+-------------------------
Reporter: jpflori | Owner:
Type: defect | davidloeffler
Priority: critical | Status: new
Component: number fields | Milestone: sage-5.12
Keywords: memleak, number field, | Resolution:
QuadraticField | Merged in:
Authors: | Reviewers:
Report Upstream: N/A | Work issues:
Branch: u/SimonKing/ticket/14711 | Commit:
Dependencies: | Stopgaps:
-------------------------------------------------+-------------------------
Comment (by nbruin):
Replying to [comment:54 SimonKing]:
> I think the (unwritten, I am afraid) contract is that
`register_coercion()` is only called in `__init__`. So, perhaps it should
rather be `_register_coercion()`, to remove it from the user interface.
Absolutely not! I think it's a very important feature that coercions can
be discovered "lazily", i.e., be registered after the fact. It also means
(but this is just a fact of life) that, while parents are supposed to be
immutable, their relations in the (global state)! coercion graph can
evolve over time.
You could of course have a `_register_coercion` for internal use that
mandates being passed a map with the promise no-one else will keep a
reference to that map, but I'm pretty sure we have to keep an advertised
`register_coercion`. You could ask sage-devel, of course.
At some point there was even an idea to have a context manager to
temporarily modify the coercion graph:
{{{
K=QuadraticField(3)
with coercion(K.real_embeddings()[0]):
print 1+K.0
}}}
leading to `-0.732050807568877` (assuming the first embedding is the
negative one).
For basic computations these things are not so essential, but by the time
you're a couple levels deep, e.g., you want to compute the archimedean
period matrices of some abelian variety defined over a number field,
letting the coercion framework to the required conversions might be the
only quick way to actually get your data in the right ring. I think we
don't want to take away that possibility.
--
Ticket URL: <http://trac.sagemath.org/ticket/14711#comment:55>
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.
