Skip to content

Latest commit

 

History

History
78 lines (70 loc) · 4.08 KB

2021-04-steering-council-update.md

File metadata and controls

78 lines (70 loc) · 4.08 KB

April 5

  • The Steering Council discussed PEP 647: User-Defined Type Guards and approved it. An acceptance notice was sent. It was discussed that after Python 3.10, the Steering Council should start a larger discussion around the typing language and how it relates to the Python language.
  • The Steering Council discussed PEP 654: Exception Groups and except* and it was decided that some members needed more time to review some of the existing discussion as well as Nathaniel's feedback.
  • The Steering Council will inform the community around expectations for scheduling. No new features (unless straight forward) will be approved after the beta freeze.
  • The Steering Council discussed accepting the clarification around PyPA members being PEP sponsors. The commit with the proposal was merged and the issue was closed.
  • The group discussed a response to Debian following the ongoing discussion around the future of the Python package in the Debian distribution and it was decided that Pablo would send the email to the Debian group.

April 12

  • The Steering Council discussed at length PEP 654: Exception Groups and except* The group decided an email should be sent to python-dev@ asking for more discussion and Thomas would draft that email.
  • The group worked on the PyCon US 2021 Steering Council keynote
  • Pablo mentioned he would send out the email to the Debian folks soon (which was sent April 13).

April 19

  • The Steering Council discussed PEP 646: Variadic Generics and was decided that a decision will be postponed to after Python 3.10 as there was not enough time to properly evaluate the PEP given the other, more time pressing issues regarding PEP 649: Deferred Evaluation Of Annotations Using Descriptors and type annotations.
  • The group discussed at length some concerns raised by a group of users and library authors regarding that the current release of 3.10 is going to make life very difficult to “pydantic”, users of the library and similar libraries that use type annotations at runtime. This is due to some limitations in PEP 563: Postponed Evaluation of Annotations that affect users of runtime annotations (and is also related with PEP 649: Deferred Evaluation Of Annotations Using Descriptors). The Steering Council discussed at length different possibilities with a careful evaluation of costs, risks and work to be done and decided that we simply can’t risk the compatibility breakage of PEP 563. The group decided that for stringified annotations the default must be reverted, at least for Python 3.10. Thomas will send an email to python-dev with the decision.

April 26

  • The Steering Council discussed PEP 654: Exception Groups and except* deciding to postpone the PEP to 3.11, and postpone their decision for 3.11 until after the Language Summit.
  • The SC discussed PEP 648: Extensible customizations of the interpreter at startup, deciding to postpone it to 3.11.
  • The SC reviewed PEP 467: Minor API improvements for binary sequences, which was postponed to 3.11 by its author.
  • The SC planned for the PyCon US keynote and Q&A, which will be recorded next week.
  • The SC briefly discussed the notion of adding sponsor logos to docs.python.org.
  • The SC discussed the proposal to delegate typing PEPs to a typing WG or BDFL, but deferred further discussion until later.
  • The SC discussed the role of CODEOWNERS in CPython and how different core devs use the file, and the consensus was that setting expectations for both reviewers and PR authors could be beneficial and could help to avoid misunderstandings. It was decided that some devguide changes could be proposed.