Author gregory.p.smith
Recipients docs@python, gregory.p.smith, martin.panter, serhiy.storchaka
Date 2015-03-20.21:07:32
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1426885652.75.0.483005832185.issue22341@psf.upfronthosting.co.za>
In-reply-to
Content
I do not object to the removal of the & 0xfffffff from the stdlib library code if these functions have actually been fixed to always return unsigned now.  (double check the behavior, and if good, just do it!)

But I think the docs should still mention that the & 0xffffffff is a good practice if code needs to be compatible with Python versions prior to X.Y (list the last release before the behavior was corrected).  Possibly within a .. versionchanged: section.

People often use the latest docs when writing code in any version of Python as we continually improve docs and are pretty good about noting old behaviors and when behaviors changed.
History
Date User Action Args
2015-03-20 21:07:32gregory.p.smithsetrecipients: + gregory.p.smith, docs@python, martin.panter, serhiy.storchaka
2015-03-20 21:07:32gregory.p.smithsetmessageid: <1426885652.75.0.483005832185.issue22341@psf.upfronthosting.co.za>
2015-03-20 21:07:32gregory.p.smithlinkissue22341 messages
2015-03-20 21:07:32gregory.p.smithcreate