This issue tracker has been migrated to GitHub, and is currently read-only.
For more information, see the GitHub FAQs in the Python's Developer Guide.

Author ncoghlan
Recipients brett.cannon, bronger, dcjim, gvanrossum, jhylton, loewis, ncoghlan, tim.peters
Date 2009-03-31.21:42:43
SpamBayes Score 0.0002534043
Marked as misclassified No
Message-id <1238535766.23.0.468487714769.issue992389@psf.upfronthosting.co.za>
In-reply-to
Content
I just had a thought: we may be able to eliminate this behaviour without
mucking about in the package globals.

What if the import semantics were adjusted so that, as a last gasp
effort before bailing out with an ImportError, the import process
checked sys.modules again with the full module name?

Not a fully fleshed out idea at this point (and possibly symptomatic of
not being fully awake yet), but I'll bring it up in the current
python-dev thread anyway.
History
Date User Action Args
2009-03-31 21:42:46ncoghlansetrecipients: + ncoghlan, gvanrossum, tim.peters, loewis, jhylton, brett.cannon, dcjim, bronger
2009-03-31 21:42:46ncoghlansetmessageid: <1238535766.23.0.468487714769.issue992389@psf.upfronthosting.co.za>
2009-03-31 21:42:44ncoghlanlinkissue992389 messages
2009-03-31 21:42:43ncoghlancreate