[issue17115] __loader__ = None should be fine

Roundup Robot added the comment: New changeset 97b7bd87c44c by Brett Cannon in branch 'default': #17115: I hate you MS for not supporting C99. http://hg.python.org/cpython/rev/97b7bd87c44c -- ___ Python tracker ___

[issue17115] __loader__ = None should be fine

Changes by Brett Cannon : -- resolution: -> fixed status: open -> closed ___ Python tracker ___ ___ Python-bugs-list mailing list Uns

[issue17115] __loader__ = None should be fine

Roundup Robot added the comment: New changeset bb023c3426bc by Brett Cannon in branch 'default': #17115: Remove what appears to be a useless chunk of code which broke http://hg.python.org/cpython/rev/bb023c3426bc -- ___ Python tracker

[issue17115] __loader__ = None should be fine

Brett Cannon added the comment: It's because test_pydoc is actively deleting the __loader__ of a test module. I'll have a look to figure out why it's doing that. -- resolution: fixed -> status: closed -> open ___ Python tracker

[issue17115] __loader__ = None should be fine

Ezio Melotti added the comment: Some buildbots are failing after this: http://buildbot.python.org/all/builders/x86%20Ubuntu%20Shared%203.x/builds/7840 http://buildbot.python.org/all/builders/AMD64%20FreeBSD%209.0%203.x/builds/4536 FAIL: test_everyone_has___loader__ (test.test_importlib.test_api.

[issue17115] __loader__ = None should be fine

Changes by Brett Cannon : -- resolution: -> fixed stage: test needed -> committed/rejected status: open -> closed ___ Python tracker ___

[issue17115] __loader__ = None should be fine

Roundup Robot added the comment: New changeset e39a8f8ceb9f by Brett Cannon in branch 'default': #17115,17116: Have modules initialize the __package__ and __loader__ http://hg.python.org/cpython/rev/e39a8f8ceb9f -- nosy: +python-dev ___ Python tracker

[issue17115] __loader__ = None should be fine

Changes by Brett Cannon : -- dependencies: -xml.parsers.expat.(errors|model) don't set the __loader__ attribute ___ Python tracker ___ _

[issue17115] __loader__ = None should be fine

Changes by Gökcen Eraslan : -- nosy: +gkcn ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.o

[issue17115] __loader__ = None should be fine

Eric Snow added the comment: > [document that] the language reference and importlib docs now supersede the > PEP Agreed. PEP 302 is even crustier now than it was a year ago and Barry's new import page in the language reference obviates the need for 302 as the de facto spec. --

[issue17115] __loader__ = None should be fine

Eric Snow added the comment: > In all honesty I would like to tweak imp.new_module()/PyModule_Create()... +1 -- nosy: +eric.snow ___ Python tracker ___ _

[issue17115] __loader__ = None should be fine

Brett Cannon added the comment: It will interpret its absence as None, but I would rather get the very common case of actual module instances just setting None automatically. Not having it set when None is already an accepted default value for __package__ seems inconsistent; either the lack of

[issue17115] __loader__ = None should be fine

Antoine Pitrou added the comment: Mmmh, what is the point of forcing people to adapt their module-like objects so that they have a __loader__ entry? Couldn't importlib just interpret the absence of __loader__ as a None? -- nosy: +pitrou ___ Python tr

[issue17115] __loader__ = None should be fine

Brett Cannon added the comment: Screw it, forget the "like" and make that a "will happen". A What's New entry telling people to update their code to use a getattr() with a None default value for transitioning should be enough (and a single line change for the few people who would care about th

[issue17115] __loader__ = None should be fine

Changes by Brett Cannon : -- dependencies: +Update importlib.util.module_for_loader/set_loader to set when __loader__ = None ___ Python tracker ___ _

[issue17115] __loader__ = None should be fine

Changes by Brett Cannon : -- dependencies: +Raise ValueError when __loader__ not defined for importlib.find_loader() ___ Python tracker ___ _

[issue17115] __loader__ = None should be fine

Changes by Brett Cannon : -- dependencies: +xml.parsers.expat.(errors|model) don't set the __loader__ attribute ___ Python tracker ___ __

[issue17115] __loader__ = None should be fine

New submission from Brett Cannon: There is no reason __loader__ can't be set to None like __package__. That means having xml.parsers.expat.(model|errors) set the attribute to None by default (and thus removing the exemption for those modules from test_importlib.test_api.StartupTests), updating