[issue24018] add a Generator ABC

Guido van Rossum added the comment: Fixed. On Fri, May 8, 2015 at 10:28 PM, Stefan Behnel wrote: > > Stefan Behnel added the comment: > > Thanks! Minor grouch: it should say "collections.*abc*.Generator" in the > NEWS entry. > > -- > > ___ > Python t

[issue24018] add a Generator ABC

Roundup Robot added the comment: New changeset f7cc54086cd2 by Guido van Rossum in branch 'default': Fix news entry for issue 24018. https://hg.python.org/cpython/rev/f7cc54086cd2 -- ___ Python tracker

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Thanks! Minor grouch: it should say "collections.*abc*.Generator" in the NEWS entry. -- ___ Python tracker ___ _

[issue24018] add a Generator ABC

Changes by Raymond Hettinger : -- resolution: -> fixed stage: -> resolved status: open -> closed ___ Python tracker ___ ___ Python-b

[issue24018] add a Generator ABC

Roundup Robot added the comment: New changeset ba5d7041e2f5 by Raymond Hettinger in branch 'default': Issue #24018: Add a collections.Generator abstract base class. https://hg.python.org/cpython/rev/ba5d7041e2f5 -- nosy: +python-dev ___ Python tracker

[issue24018] add a Generator ABC

Changes by Raymond Hettinger : -- assignee: lukasz.langa -> rhettinger ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubs

[issue24018] add a Generator ABC

Guido van Rossum added the comment: Ask Yury if he'll commit it for you. It's ready. -- ___ Python tracker ___ ___ Python-bugs-list ma

[issue24018] add a Generator ABC

Stefan Behnel added the comment: This is blocking issue 24017 (async/await syntax). -- ___ Python tracker ___ ___ Python-bugs-list mai

[issue24018] add a Generator ABC

Łukasz Langa added the comment: Yup, will do. -- ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.p

[issue24018] add a Generator ABC

Guido van Rossum added the comment: Yeah, looks good -- Łuke, can you commit this? -- ___ Python tracker ___ ___ Python-bugs-list mail

[issue24018] add a Generator ABC

Raymond Hettinger added the comment: The latest patch looks good overall. Łukasz, assigning back to you. -- assignee: rhettinger -> lukasz.langa ___ Python tracker ___ __

[issue24018] add a Generator ABC

Stefan Behnel added the comment: > Please either > 1) drop the throw() method entirely or > 2) make throw an abstractmethod() Ok, as I already said, I think it's important to provide the complete protocol as code will usually expect that. Also, close() has a helpful implementation, but it dep

[issue24018] add a Generator ABC

Guido van Rossum added the comment: I'm with Raymond. -- ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https:

[issue24018] add a Generator ABC

Raymond Hettinger added the comment: > Therefore, I think it's important to cover the complete protocol > in the Generator ABC. I also think it's helpful to not require > users to override throw() in a subclass, as they might not need it. Sorry, but I think you're fighting the fundament nature

[issue24018] add a Generator ABC

Stefan Behnel added the comment: PEP 342 isn't really conclusive here as it intended to define the protocol based on the de-facto design of a yield-based generator function. Trying to abstract from that poses the question how a class based generator implementation should look like. Specificall

[issue24018] add a Generator ABC

Antoine Pitrou added the comment: I think the throw() method should be required. If you don't need throw(), send() or close(), then you aren't really asking for a full-blown generator: you are asking for an iterator, so you can just check for collections.Iterator. (PS: why is this bug assigned

[issue24018] add a Generator ABC

Raymond Hettinger added the comment: In the lastest patch, the close() method is now a valid mixin method. However, the throw() method should be made abstract because it doesn't provide the required operation (it doesn't even use the "self" argument) or it should be left out entirely (i.e. N

[issue24018] add a Generator ABC

Ludovic Gasc added the comment: Sorry guys to be basic for you, but if I take my "AsyncIO end-user" hat, I'm not sure to understand the potential end-user source code impacts to use Cython with Python 3.5 and AsyncIO. In concrete terms, it's only a low-level change, Cython will monkeypatch CP

[issue24018] add a Generator ABC

Stefan Behnel added the comment: FYI, here's the patch implementation for Cython: https://github.com/cython/cython/blob/cf63ff71c06b16c3a30facdc7859743f4cd495f6/Cython/Utility/Generator.c#L849 The only difference is that it takes care of changing "__next__" to "next" in Py2.x. -- ___

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Searching for Python code that seems to implement the Generator protocol doesn't return much: https://code.openhub.net/search?s=%22def%20throw%28%22%20%22def%20send%28%22%20%22def%20close%28%22&pp=0&fl=Python&mp=1&ml=1&me=1&md=1&ff=1&filterChecked=true But at l

[issue24018] add a Generator ABC

Changes by Stefan Behnel : Added file: http://bugs.python.org/file39180/inspect_docs.patch ___ Python tracker ___ ___ Python-bugs-list mailing

[issue24018] add a Generator ABC

Changes by Stefan Behnel : Removed file: http://bugs.python.org/file39179/inspect_docs.patch ___ Python tracker ___ ___ Python-bugs-list maili

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Adding a patch for the inspect docs that refers to the Right Way To Do It. -- Added file: http://bugs.python.org/file39179/inspect_docs.patch ___ Python tracker

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Yes, and there are certainly other compilers and tools. However, it's going to be a short list, so if Py3.5 takes the lead, they can all just agree on the one way to do it and let the first patcher win. -- ___ Python

[issue24018] add a Generator ABC

Antoine Pitrou added the comment: In some distant future, Numba might also care too :) (right now, we only support compiling basic generators, i.e. no send() and no throw()) -- ___ Python tracker _

[issue24018] add a Generator ABC

Guido van Rossum added the comment: Realistically only Cython will care much about this, so I'm okay if Cython just monkeypatches the collections package. -- ___ Python tracker

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Yes, code usually doesn't fall from the sky with a new CPython release. If something wants to make use of this ABC, it still has to find ways to also work with older CPythons. What would be a good fallback? A backport on PyPI? I wouldn't even mind shipping this

[issue24018] add a Generator ABC

Guido van Rossum added the comment: Is it an option to leave inspect alone? Its definition and use of generators is very focused on the builtin implementation. Although this means that code that wants to do the right thing but also be backwards compatible has to use something like ``` def isge

[issue24018] add a Generator ABC

Stefan Behnel added the comment: > should inspect.isgenerator() be changed? Replying to myself, one thing that speaks against this approach is that "inspect" also has the functions "getgeneratorlocals()" and "getgeneratorstate()", which depend on "gen.gi_frame". Cython could emulate that, but

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Good catch with the RuntimeError. Patch updated. -- Added file: http://bugs.python.org/file39157/generator_abc.patch ___ Python tracker ___ _

[issue24018] add a Generator ABC

Martin Panter added the comment: I agree that there is no big reason why we should force generators to stop working after close(). Your new default implementation of close() is probably the right thing too. I added a few new comments on Reitveld. -- ___

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Next question: should inspect.isgenerator() be changed? Or should its usages be replaced by isinstance(obj, Generator) ? -- ___ Python tracker _

[issue24018] add a Generator ABC

Stefan Behnel added the comment: > Is it a problem that the check can't be done in a fast way from C code? C code can still quickly special case the generator type in the positive case, and it will usually be interested in an exact type match anyway. Determining that an object is *not* (compat

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Here's a new patch that addresses the review comments. I kept throw() and close() non-abstract and added an example to the tests instead that implements a minimal generator by inheriting these methods from the Generator base class, using it as a mixin. It only

[issue24018] add a Generator ABC

Raymond Hettinger added the comment: For the other ABCs, if you define the required abstract methods, you get working versions of all the mixin methods. In the case of the Generator ABC, throw() and close() are useless empty stub methods. In the other ABCs, we leave optional methods out entir

[issue24018] add a Generator ABC

Changes by Martin Panter : -- nosy: +vadmium ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: https://mail.pytho

[issue24018] add a Generator ABC

Changes by Łukasz Langa : -- assignee: -> lukasz.langa nosy: +lukasz.langa ___ Python tracker ___ ___ Python-bugs-list mailing list U

[issue24018] add a Generator ABC

Antoine Pitrou added the comment: Is it a problem that the check can't be done in a fast way from C code? Other than that, sounds good to me. -- nosy: +pitrou ___ Python tracker ___

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Sorry, here's another doc fix. -- Added file: http://bugs.python.org/file39151/generator_abc.patch ___ Python tracker ___ ___

[issue24018] add a Generator ABC

Changes by Stefan Behnel : Removed file: http://bugs.python.org/file39150/generator_abc.patch ___ Python tracker ___ ___ Python-bugs-list mail

[issue24018] add a Generator ABC

Stefan Behnel added the comment: Ok, sure. Here's a new patch that adds tests and docs. -- Added file: http://bugs.python.org/file39150/generator_abc.patch ___ Python tracker ___

[issue24018] add a Generator ABC

Changes by Stefan Behnel : Removed file: http://bugs.python.org/file39146/generator_abc.patch ___ Python tracker ___ ___ Python-bugs-list mail

[issue24018] add a Generator ABC

Guido van Rossum added the comment: It's missing tests. :-) Otherwise looks quite sensible. Also, shouldn't you override __subclasshook__ so you don't inherit it from Iterator? -- ___ Python tracker

[issue24018] add a Generator ABC

New submission from Stefan Behnel: Currently, CPython tends to assume that generators are always implemented by a Python function that uses the "yield" keyword. However, it is absolutely possible to implement generators as a protocol by adding send(), throw() and close() methods to an iterator