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 christian.heimes
Recipients christian.heimes, erlendaasland, pablogsal, petr.viktorin, rhettinger, shihai1991
Date 2021-10-23.12:45:24
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1634993124.85.0.241399215956.issue45113@roundup.psfhosted.org>
In-reply-to
Content
> Given that some essential third party modules are not going down this path, it is unlikely that general users would ever see any benefit.

I disagree with your statement. It does not reflect my experience, too.

Heap types are a prerequisite for abi3 wheels. In order to get abi3 binary wheels, extensions must use the limited API, which means they must use heap types. Stable ABI wheels are a great benefit for extension authors, because it reduces their build matrix substantially. For instance cryptography's stable abi3 wheels work with any Python version >=3.6,<4.

Heap types and stable abi3 are also a major benefit for general users. Once Cython and the NumPy stack supports limited API and stable ABI, users no longer have to wait on builds for a new Python release. Every new minor release we get several bug reports on BPO about NumPy not working on latest release. This will go away.
History
Date User Action Args
2021-10-23 12:45:24christian.heimessetrecipients: + christian.heimes, rhettinger, petr.viktorin, pablogsal, shihai1991, erlendaasland
2021-10-23 12:45:24christian.heimessetmessageid: <1634993124.85.0.241399215956.issue45113@roundup.psfhosted.org>
2021-10-23 12:45:24christian.heimeslinkissue45113 messages
2021-10-23 12:45:24christian.heimescreate