#12808: Optimize ClassCallMetaClass using Cython
--------------------------------------------------+-------------------------
Reporter: hivert | Owner: jason
Type: enhancement | Status: needs_review
Priority: major | Milestone: sage-5.0
Component: misc | Resolution:
Keywords: classcall UniqueRepresentation | Work issues:
Report Upstream: N/A | Reviewers:
Authors: Florent Hivert | Merged in:
Dependencies: | Stopgaps:
--------------------------------------------------+-------------------------
Comment (by hivert):
Hi Simon,
Thanks for your in depth review.
Replying to [comment:20 SimonKing]:
> Some further questions (in addition to my question about the rôle of
_included_private_doc_):
This is a tentative inclusion of the doc of the special methods (see
[http://groups.google.com/group/sage-
devel/browse_thread/thread/b1cf6b2243bba673?hl=en this thread] on sage
devel.
> Why is `__all__ = ['ClasscallType', 'ClasscallMetaclass', 'typecall',
'timeCall']` added? Isn't importing also possible without that? Or is it
needed to make `from sage.misc.classcall_metaclass import *` work?
I just want to hide the few class I wrote here for timing (CRef, C2, C3,
C2C). With this line, they are neither imported with import *, nor
documented.
> Why is there a two-step cythonisation? I mean, why is there a cdef class
> `ClasscallType` implementing the special methods, and then
> `ClasscallMetaclass` defined by double inheritance from `ClasscallType`
and
> `NestedClassMetaclass`? Wouldn't it be better/easier/faster to cdef
> `NestedClassMetaclass` as well, and then cdef `ClasscallMetaclass`
directly,
> without having the `ClasscallType`?
There is this discussion about cleaning up the way metaclass are defined
and
used in Sage. I just wanted to keep features separate. Should we decide to
merge or rewrite the metaclass architecture, I'd rather to keep this for a
different ticket.
> I guess the typecall function will be useful in other modules (e.g.,
when
> cythonising dynamic_class). But it only is def. Shouldn't it rather be
`cdef
> inline`?
You mean `cpdef inline` (we may want to call it from Python) ? I had the
impression from C that this inline doesn't do anything if the function is
defined in a different module. More precisely, in C/C++ inline only work
if
the function is defined in {{{.h}}} instead of {{{.c}}}. Does Cython do
anything for that ?
> Concerning the timing tools provided in the module: Is there still no
> "central" location for all aspects of timing? I thought there were
> occasional discussions on sage-devel about a benchmark/timing framework.
Not that I know of.
Florent
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/12808#comment:21>
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.
