sylvain.thenault@logilab.fr [Wed, 18 Feb 2009 00:45:01 +0100] rev 737
fix traced_selection
sylvain.thenault@logilab.fr [Wed, 18 Feb 2009 00:30:55 +0100] rev 736
need safety belt for class registered twice (during test for instance)
Adrien Di Mascio <Adrien.DiMascio@logilab.fr> [Wed, 18 Feb 2009 00:22:47 +0100] rev 735
__select__ is not a classmethod anymore, it's a pure function (or a Selector instance),
we have to pass the class explicitly
sylvain.thenault@logilab.fr [Wed, 18 Feb 2009 00:12:36 +0100] rev 734
get kwargs as well
sylvain.thenault@logilab.fr [Wed, 18 Feb 2009 00:12:18 +0100] rev 733
ooooops
Adrien Di Mascio <Adrien.DiMascio@logilab.fr> [Wed, 18 Feb 2009 00:12:34 +0100] rev 732
__selectors__ must return a tuple
Adrien Di Mascio <Adrien.DiMascio@logilab.fr> [Wed, 18 Feb 2009 00:03:27 +0100] rev 731
some more s/__selectors__/__select__ but still more to come
sylvain.thenault@logilab.fr [Tue, 17 Feb 2009 23:54:30 +0100] rev 730
__select__ migration
sylvain.thenault@logilab.fr [Tue, 17 Feb 2009 23:54:15 +0100] rev 729
unnecessary registerrer
Adrien Di Mascio <Adrien.DiMascio@logilab.fr> [Tue, 17 Feb 2009 23:47:23 +0100] rev 728
first pass to use __select__ instead of __selectors__
Adrien Di Mascio <Adrien.DiMascio@logilab.fr> [Tue, 17 Feb 2009 23:46:48 +0100] rev 727
fix _instantiate_selector() mini bug (make sure obj is a class before calling issubclass)
Adrien Di Mascio <Adrien.DiMascio@logilab.fr> [Tue, 17 Feb 2009 23:45:51 +0100] rev 726
improve __selectors__() so that it always return a tuple
Adrien Di Mascio <Adrien.DiMascio@logilab.fr> [Tue, 17 Feb 2009 23:44:58 +0100] rev 725
safety belt inside _accept_interfaces() as it might be called on a class that doesn't provide (yet) __select__
sylvain.thenault@logilab.fr [Tue, 17 Feb 2009 23:43:08 +0100] rev 724
should no more be used, drop documentation