#14214: Cythoned homsets
----------------------------------------------+-----------------------------
Reporter: SimonKing | Owner: tbd
Type: enhancement | Status: needs_work
Priority: major | Milestone: sage-5.10
Component: performance | Resolution:
Keywords: Hom, cython, cached method | Work issues: How to store
domain and codomain?
Report Upstream: N/A | Reviewers:
Authors: Simon King | Merged in:
Dependencies: #14159, #12951, #13184 | Stopgaps:
----------------------------------------------+-----------------------------
Comment (by nbruin):
I think you want the following code:
{{{
from sage.all import Sets
from sage.structure.parent cimport Parent
cdef class CachedAttributeType(object):
cdef public str name
def __init__(self, name):
self.name = name
def __get__(self,instance,cls):
try:
return instance.__cached_methods[self.name]
except KeyError:
raise AttributeError("%s instance has no attribute
'%s'"%(cls,self.name))
cdef class Bla(Parent):
def __init__(self, D,C):
try:
self.__dict__['_domain'] = D
self.__dict__['_codomain'] = C
except AttributeError:
if self.__cached_methods is None:
self.__cached_methods = {'_domain':D, '_codomain':C}
else:
self.__cached_methods['_domain'] = D
self.__cached_methods['_codomain'] = C
Parent.__init__(self, category = Sets())
_domain=CachedAttributeType("_domain")
_codomain=CachedAttributeType("_codomain")
}}}
Timings we'll be getting with the cython upgrade (with Python's attribute
lookup cache enabled)
{{{
sage: b=Bla("dommi","codommi")
sage: b._domain
'dommi'
sage: %timeit b._domain
10000000 loops, best of 3: 97.6 ns per loop
sage: c=Cla("dommi","codommi")
sage: %timeit c._domain
10000000 loops, best of 3: 50 ns per loop
}}}
With the old cython (where MRO depth is a big factor in attribute lookup):
{{{
sage: %timeit b._domain
10000000 loops, best of 3: 130 ns per loop
sage: %timeit c._domain
10000000 loops, best of 3: 78.5 ns per loop
}}}
What we should really do is get rid of `__cached_methods` make sure these
objects have a `__dict__` instead, where such information can be stored
much more effectively. There is an "empty dict" value one can use as a
place holder for empty dictionaries. There's really no good reason why
cython types cannot have a dictionary.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/14214#comment:40>
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?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
