On Tue, Apr 4, 2017 at 2:29 AM, Ethan Furman <et...@stoneleaf.us> wrote: > It this point I do not. If you can give us an example Enum and why it's > necessary to be built like that I might be swayed -- although the odds are > good that the change will go into aenum instead (it's already a mess with > the 2.7 compatibility code...).
Here's a counter-example that supports the current behaviour: >>> from enum import IntFlag, auto >>> class Spam(IntFlag): ... FOO = auto() ... BAR = auto() ... FOOBAR = FOO | BAR ... SPAM = auto() ... HAM = auto() ... SPAMHAM = SPAM | HAM ... >>> list(Spam) [<Spam.FOO: 1>, <Spam.BAR: 2>, <Spam.FOOBAR: 3>, <Spam.SPAM: 4>, <Spam.HAM: 8>, <Spam.SPAMHAM: 12>] It makes perfect sense to define combined flags in terms of the (previously-defined) individual flags. If auto() has to be deferred until the class is fully defined, this would be a lot harder. ChrisA -- https://mail.python.org/mailman/listinfo/python-list
