Message 379325 - Python tracker

➜

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	steve.dower
Recipients	ejguan, paul.moore, steve.dower, tim.golden, zach.ware
Date	2020-10-22.19:30:37
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<1603395037.91.0.511106382595.issue42120@roundup.psfhosted.org>
In-reply-to

Content
Sounds good, but I don't think we can backport it. It's not _our_ public API, but it's still going to impact compiling modules. Also, looking at the docs link above, copysign() is for floats and _copysign() is for doubles, which means it's an actual functional change. If POSIX/libc copysign() is floats, it's probably fine for a 3.10 change, but if we need to preserve the type then it needs to become _Py_copysign throughout.

Sounds good, but I don't think we can backport it. It's not _our_ public API, but it's still going to impact compiling modules.

Also, looking at the docs link above, copysign() is for floats and _copysign() is for doubles, which means it's an actual functional change. If POSIX/libc copysign() is floats, it's probably fine for a 3.10 change, but if we need to preserve the type then it needs to become _Py_copysign throughout.

History
Date	User	Action	Args
2020-10-22 19:30:37	steve.dower	set	recipients: + steve.dower, paul.moore, tim.golden, zach.ware, ejguan
2020-10-22 19:30:37	steve.dower	set	messageid: <1603395037.91.0.511106382595.issue42120@roundup.psfhosted.org>
2020-10-22 19:30:37	steve.dower	link	issue42120 messages
2020-10-22 19:30:37	steve.dower	create