Paul Moore wrote: > 2009/3/1 Nick Coghlan <ncogh...@gmail.com>: >> As much as I'd like to get a simple generic implementation into >> functools, the lack of support for ABCs still bothers me (despite my >> last post about that on the tracker item). I'd be a -0 on it going in as >> is, but if someone can figure out a clever way of supporting ABCs >> without completing killing the invocation speed for generics, that would >> go up to a +1. > > Nick, > Armin Ronacher pointed out that it's not likely to be possible to > enumerate ABCs even in theory, as ABCs can do semantic checks (e.g. > checking for the presence of special methods like __iter__) without > needing either inheritance or explicit registration. > > As you had a genuine use case for simplegeneric, how badly does this > limitation damage it?
That's a very good point that Armin raises (i.e. that enumerating the ABCs a given class implements is impossible even in theory, let alone in practice). It is actually enough to put me back to the perspective that I posted on the tracker item: the failure to inherit generic function implementations when a class doesn't actually have the ABC in its MRO is similar in nature to the failure to inherit the ABCs own method implementations. It still bugs me a little, but I'll get over it :) Cheers, Nick. -- Nick Coghlan | ncogh...@gmail.com | Brisbane, Australia --------------------------------------------------------------- _______________________________________________ Python-Dev mailing list Python-Dev@python.org http://mail.python.org/mailman/listinfo/python-dev Unsubscribe: http://mail.python.org/mailman/options/python-dev/archive%40mail-archive.com
