Skip to content

Coq Call 2022 02 09

Matthieu Sozeau edited this page Feb 9, 2022 · 8 revisions

Topics

  • Status of the mind_finite field of inductive types (reflect the keyword use at definition time or tell if the type is really non-recursive, independently of the keyword) see e.g. #8226 [HH]
  • Unifying the name of functions for destructing sequences of lambda/prod and letins (#15582) [HH]
  • Satisfied with intermediate status of parametricity-based implementation of Boolean equality (#15527)?
  • Need an assignee for https://github.com/coq/coq/pull/15549 (Ali)
  • Display git hash in non-release versions: https://github.com/coq/coq/pull/15639 (Ali)
    • Pros: Easy to copy and paste for users of opam master etc. Makes debugging "version master" from ancient issues easier. Everybody else does it.
    • Cons: Not clear how to make reproducible. Depends on git. Messes up build order.
  • If Emilio is present, what to do about https://github.com/coq/coq/pull/15645 (Ali)
  • Making Ltac2 independent (Ali)
    • Would be nice for HoTT to use the Ltac2 library without any of the stdlib.
  • Hackathon: (Ali)

Notes

  • [mind_finite] Decision to have the flag reflect the actual recursivity but keeping compatibility as much as possible.
  • PR 15582: let's continue uniformizing the names
  • https://github.com/coq/coq/pull/15527: would need more manpower to go beyond, but a good first step.
Clone this wiki locally