Message 336660 - 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	ncoghlan
Recipients	arigo, eric.snow, ncoghlan, njs, vstinner
Date	2019-02-26.12:59:14
SpamBayes Score	-1.0
Marked as misclassified	Yes
Message-id	<[email protected]>
In-reply-to

Content
(On closer inspection, that's actually be the same breakage as already mentioned above) However, what I'm not clear on is how this would affect projects that had already generated their cffi code, and include that in their sdist. Are all those sdists going to fail to build on Python 3.8 now? It's OK to require that extension modules be rebuilt for a new release, but breaking compatibility with a code generator that means a broad selection of projects are all going to fail to build is a much bigger problem.

(On closer inspection, that's actually be the same breakage as already mentioned above)

However, what I'm not clear on is how this would affect projects that had *already* generated their cffi code, and include that in their sdist. Are all those sdists going to fail to build on Python 3.8 now?

It's OK to require that extension modules be rebuilt for a new release, but breaking compatibility with a *code generator* that means a broad selection of projects are all going to fail to build is a much bigger problem.

History
Date	User	Action	Args
2019-02-26 12:59:14	ncoghlan	set	recipients: ncoghlan, arigo, vstinner, njs, eric.snow
2019-02-26 12:59:14	ncoghlan	set	messageid: <[email protected]>
2019-02-26 12:59:14	ncoghlan	link	issue35886 messages
2019-02-26 12:59:14	ncoghlan	create