ok, that makes a bit more sense as to why it happens. but there ought
to be indeed a better solution...

On Sep 7, 10:41 am, "Nicolas M. Thiery" <[email protected]>
wrote:
> On Sat, Sep 05, 2009 at 11:02:21AM -0700, Paul-Olivier Dehaye wrote:
> > I don't know enough python to tell for sure, but this looks suspicious
> > to me. The printout below shows that a lot of time is spent on
> > JoinCategory._repr_ (1 underscore) in categories.py and
> > SymmetricFunctionAlgebra_classical.__repr__ (2 underscores), and I don
> > t understand why it should be that way. Is it normal? (it seems to get
> > worse with Big going bigger too)
>
> Thanks for the report.
>
> Each time a tensor([a,b]) is calculated, it has, for example, to
> retrieve the parent of the result from those of a and b. This is
> cached, and looking up this cache involves computing a hash value.
>
> My bet is that this comes from the default implementation of the
> hashing function for SageObject, which is to compute a hash of the
> string representation of the object. That's not very safe and is *slow*.
>
> I have been meaning for a while to use 'id' instead has default
> implementation for __hash__ for objects with unique representation.
>
> Thoughts anyone?
>
> Should we go for it? Or at least cache the hash function?
>
> That this repr thing appears for categories rather than parents is
> further suspicious. There might be another missing cache somewhere in
> the tensor product code. I'll keep this in mind next time I play
> around with this!
>
> Cheers,
>                                 Nicolas
> --
> Nicolas M. Thiéry "Isil" <[email protected]>http://Nicolas.Thiery.name/
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"sage-combinat-devel" 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-combinat-devel?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to