Hi Florent,
On 29 Mrz., 21:54, Florent Hivert <florent.hiv...@univ-rouen.fr>
wrote:
> So it seems that there is no difference... Well actually I just figured out
> that this has something to do with Cython: if I change
>
> class End(object):
>
> to
>
> class End(SageObject):
>
> Then:
>
> sage: timeit('obj_long.toto()',number=10^6)
> 1000000 loops, best of 3: 9.22 s per loop
> sage: timeit('obj_end.toto()',number=10^6)
> 1000000 loops, best of 3: 203 ns per loop
I think, the question is (at least on the short run), how one can work
around.
Taking your example:
class End(SageObject):
def toto(self):
return 1
sage: c_long = long_mro(500)
sage: obj_long = c_long()
sage: obj_end = End()
sage: timeit("obj_end.toto()")
625 loops, best of 3: 963 ns per loop
sage: timeit("obj_long.toto()")
625 loops, best of 3: 22.8 µs per loop
So, as you confirmed, a long mro matters, at least if Cython is
involved.
But it seems possible to speed things up by offering a short path:
sage: c_long.toto = End.toto
sage: timeit("obj_long.toto()")
625 loops, best of 3: 2.55 µs per loop
My question is:
Is it good or at least acceptable practice to do things like
class MPolynomialRing_polydict( MPolynomialRing_macaulay2_repr,
PolynomialRing_singular_repr, MPolynomialRing_generic):
...
base_ring = CategoryObject.base_ring
base = CategoryObject.base
...
cdef class MPolynomial(CommutativeRingElement):
...
parent = Element.parent
...
The example above indicates that it would very well speed things up.
Or is there a good reason to avoid that short path?
Cheers,
Simon
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel+unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
