[pypy-dev] idea for Summer of PyPy - call for feedback (multimethods, actually)
mrevelle at gmail.com
Thu Jan 19 15:22:09 CET 2006
I'm not sure you looked closely enough at PyProtocols dispatch.
Here's a good example from Bob Ippolito (make sure to read the comments):
Sorry if this discussion is bothering anyone on pypy-dev.
On 1/19/06, Aurélien Campéas <aurelien.campeas at logilab.fr> wrote:
> Le mercredi 18 janvier 2006 à 10:43 -0800, Drumheller, Michael a écrit :
> > >>> -----Original Message-----
> > >>> From: Aurelien Campeas [mailto:aurelien.campeas at logilab.fr]
> > >>> Sent: Wednesday, January 18, 2006 8:33 AM
> > >>> To: pypy-dev at codespeak.net
> > >>> Subject: Re: [pypy-dev] idea for Summer of PyPy - call for feedback
> > >>> It still does not seem to really support multiple dispatch
> > >>> (only the funky "predicative" dispatch). David Mertz's own
> > >>> multimethods.py is more complete than that. Whatever. None
> > >>> of them is complete.
> > It is nice to see that others are aware of David Mertz's multimethods.py.
> > I have the impression that not enough people know about it.
> > There is a decorator technique that you can use with it which gives it
> > practically transparent syntax, which I also think is not very well known.
> I have seen Guido's "multimethods in 5 minutes" which is by-the-way more
> a showcase for decorators than for multimethods.
> > Just curious: Can you say in what way multimethods.py is inadequate for your purposes?
> For my immediate usage, they lack the decorator stuff (I don't know yet
> how to do it).
> For WP10 purposes ('aspects'), they lack around/before/after methods (I
> can't see them in http://gnosis.cx/download/gnosis/magic/multimethods.py
> anyway) ... but Adrien here told me that there is more in aspects than
> this stuff.
> Also having fast dispatch in PyPy could be nice (isn't PyPy a bit
> concerned by performance ... ?).
> > Michael
> > >>> Regards,
> > >>> Aurélien
> pypy-dev at codespeak.net
More information about the Pypy-dev