Home Original page

[Python-Dev] Pickling failure on Enums

Nick Coghlan ncoghlan at gmail.com
Wed May 15 00:16:36 CEST 2013 
On 15 May 2013 07:38, "Guido van Rossum" <guido at python.org> wrote:
>
> On Tue, May 14, 2013 at 2:13 PM, Ethan Furman <ethan at stoneleaf.us> wrote:
> > On 05/14/2013 01:58 PM, Guido van Rossum wrote:
> >>
> >> On Tue, May 14, 2013 at 1:09 PM, Ethan Furman <ethan at stoneleaf.us>
wrote:
> >>> I can get pickle failure on members created using the functional
syntax
> >>> with no module set;
> >>
> >>
> >> That's the case I care most about.
> >
> >
> > Good, 'cause that one is handled.  :)
>
> Then we're good.
>
> >>> I cannot get pickle failure on those same classes;
> >>
> >>
> >> I suppose you mean "if you create the same enums using class syntax"?
> >> Sounds fine to me.
> >
> >
> > No.  Example class:
> >
> > --> Example = Enum('Example', 'example ie eg')  # no module name given,
> > frame hack fails
> >
> > --> pickle(Example.ie)
> > # blows up
> >
> > --# pickle(Example)
> > # succeeds here, but unpickle will fail
>
> Not great, but (a) few people pickle classes, and (b) there's probably
> something you can do to the metaclass to sabotage this. But it's fine
> to punt on this now.

It may be a bug in pickle - it sounds like it is sanity checking type(obj),
but not checking for cases where obj itself is a class.

Cheers,
Nick.

>
> >>> I cannot get pickle failure on class syntax enums that inherit complex
> >>> types
> >>> (such as the NEI class in the tests).
> >>
> >>
> >> Is the NEI base class picklable?
> >
> >
> > No.  If it is, then the derived enum is also picklable (at least the
> > variation I have tested, which is when the NEI base class has
> > __getnewargs__).
> >
> > I'm really hoping you'll say that can be a documentation issue.  ;)
>
> Essentially the same response -- with enough hackery you can probably
> get this to do what you want, but I wouldn't hold up a release for it.
>
> For example you could file low-priority bugs for both issues in the
> hope that someone else figures it out.
>
> --
> --Guido van Rossum (python.org/~guido)
> _______________________________________________
> Python-Dev mailing list
> Python-Dev at python.org
> http://mail.python.org/mailman/listinfo/python-dev
> Unsubscribe:
http://mail.python.org/mailman/options/python-dev/ncoghlan%40gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-dev/attachments/20130515/8eb9a00c/attachment-0001.html>
More information about the Python-Dev mailing list