Description
The other PEPs that look like good candidates for being marked final are listed as follows. I've excluded PEPs that were approved for >3.11, refer to removals/changes planned for future version (e.g. PEP 594) or are otherwise known to be not yet final, and list special cases, PEPs I'm unsure about and typing PEPs (which may be contingent on implementation in mainstream type checkers) separately.
@gustavgransbo If you'd like to help, feel free to go ahead and submit a PR for each PEP, pinging any authors not already listed as codeowners.
The PEPs that are implemented in released versions and have no outstanding considerations I'm aware of:
- PEP-560 Core support for typing module and generic types 1: PEP 560: Mark as Final #3570
- PEP-578 Python Runtime Audit Hooks: PEP 578: Mark as Final #3811
- PEP-590 Vectorcall: a fast calling protocol for CPython: PEP 590: Mark as Final #3598
- PEP-615 Support for the IANA Time Zone Database in the Standard Library: PEP 615: Mark as Final #3592
- PEP-617 New PEG parser for CPython: PEP 617: Mark as Final #3596
- PEP-634 Structural Pattern Matching: Specification
- PEP-654 Exception Groups and except*
- PEP-678 Enriching Exceptions with Notes
- PEP-680 tomllib: Support for Parsing TOML in the Standard Library
- PEP-3121 Extension Module Initialization and Finalization
There are also some special cases:
- PEP-602 Annual Release Cycle for Python: Should be recategorized as an Active Process PEP: PEP 602: Mark as Active Process #3736
- PEP-581 Using GitHub Issues for CPython: Since the PEP's content is tightly focused on the rationale for choosing GitHub rather than the current, active process for using it and is thus now of historical rather than procedural value, it should be marked Final (rather than Active) like other transition-oriented Process PEPs have: PEP 581: Mark as Final #3807
I'm not sure about the following:
- PEP-590 Vectorcall: a fast calling protocol for CPython:
Implemented, but there is currently active work in 3.12 to better expose both ends of the protocol via the CPython API, and have the changes in the Finalizing the API section been addressed? @encukou @markshannon any insight here?Confirmed by @encukou to be fully implemented and should be marked Final; added above. - PEP-3121 Extension Module Initialization and Finalization:
Accepted from all the way back in 3.0, but I'm not sure whether this should be considered Final yet, since its been a long time and there is still ongoing work in PEP 687 that might be related. @encukou @erlend-aasland any insight?Confirmed by @encukou to be Final, as the API it describes is implemented and building on top of it is a topic for future PEPs.
There's also the Accepted typing PEPs, which might also depend on implementation status in the major typecheckers as well as being formally documented the appropriate places, since that is generally the majority of actual implementation work; @JelleZijlstra would be much better informed about that. Here's the raw list:
544 Protocols: Structural subtyping (static duck typing) typing
563 Postponed Evaluation of Annotations typing
586 Literal Types typing
589 TypedDict: Type Hints for Dictionaries with a Fixed Set of Keys typing
591 Adding a final qualifier to typing typing
593 Flexible function and variable annotations typing
604 Allow writing union types as ``X | Y`` typing
612 Parameter Specification Variables typing
613 Explicit Type Aliases typing
646 Variadic Generics typing
647 User-Defined Type Guards typing
655 Marking individual TypedDict items as required or potentially-missing typing
673 Self Type typing
675 Arbitrary Literal String Type typing
681 Data Class Transforms typing
Finally, I've left the 8000-series PEPs for a separate issue.
Footnotes
-
Typing-related, but principally deals with core CPython, not type checkers ↩