Minutes January 26 2022 - math-comp/math-comp GitHub Wiki
Participants: Reynald, Cyril, Enrico, Kazuhiko, Laurent, Pierre, Assia, Yves, Christian
-
Selecting a mathcomp version for the Coq platform (cf this zulip thread). As of now I (Cyril) do not think we released the 4CT and OOT for mathcomp 1.14.0 yet (nor mathcomp 1.13.0 AFAIK).
- 1.14 was selected
- odd order and four color's CIs fixed
- there were broken by mathcomp 1.13 but no released were done
- the lack of tag ni four color prevents the update of Nix
- TODO: Cyril to tag and Pierre to do the opam packages (edit: Pierre: I'll update the Nix ones too)
-
Discuss potential migration of MathComp project to coq-community
- Shall try to move some packages?
- What are the reasons not to move a package?
- Difference between packages using or not using coq-community tools?
- We are already using the Nix tools from coq-community
- We can use meta.yml but not to generate opam files
- The change of urls would be the main reason
- URLs in papers, in CIs, in old packages
- need to change upstreams
- Setting of email notifications in github?
- Move the package, doubling it and leave a warning behind?
- https://docs.github.com/en/repositories/creating-and-managing-repositories/transferring-a-repository
- One needs to receive right to transfer to the target community
- Try one and see how it goes
- Let the webpage and the mathcomp behing (they are not Coq)
- maybe not analysis (too big)
- maybe not finmap or bigenough (not leaves)
- move Tarjan?
- TODOs:
- Assia do the experiment with Apery and report (maybe not for the next meeting)
- four-color would be a good candidate (Yves to be taking care of this one?)
- both covers different sets of test points
-
Number notations for rat, int, ring #841
- have a number notation to be able to write say
42
instead of42%:R
inring_scope
- as of today it would be a
nat
by default
- as of today it would be a
- the difficulty was to update the CI
- we need to first merge the overlays
- Add number notation for rat
3.14%Q
is ok,(-2)%Q
is fine- how many
%N
added? how many%R
removed? x + -(2%Q)
andx + (-2)%Q
are not the same,x + -(2)%Q
andx - 2%Q
are the same- what happens when we will add semi-rings?
- two files touched in
solvable
andextrema
, do we really need to open the ring scope?- because it uses zmodule-group compatibility layer?
- TODO: Pierre to try
- have a number notation to be able to write say