Louis Gesbert

As mentionned in private, this is more restrictive when running non-interactively but better reflects what I had in mind when adding the menu. I am open to discussion if it...

Note also that, while the current interface and cli flags may be confusing, the upcoming version should be pretty clear in how you can work around that lack of detection....

(also included a little bit of cleanup, like removing things like `Format.fprintf fmt "%a"`)

# Types in expressions Some expressions contain types. This is mandatory, since type constraints may come from the syntax and need to be passed through. This concerns: - `EAbs`, with...

# Structs vs tuples We now have: - for ScopeLang: ```ocaml | EStruct of StructName.t * expr StructFieldMap.t | EStructAcess of expr * StructFieldName.t * StructName.t - for Dcalc ```ocaml...

# Use inline records I think we have consensus on this one, but it's mostly cosmetic and not very high priority (a bit tedious to write too). Replace e.g. ```ocaml...

> ## Structs vs tuples Following proposition `#1`, encoding structs as `ECoerce (ETuple (_), TStruct struct_name)` and `ETupleAccess (ECoerce (_, TStruct struct_name), n)` could be a reasonable solution, depending on...

Nope, not yet!

Suggestion: - if $OPAMROOT is not defined: ``` In normal operation, opam only alters files within ~/.opam. ``` as before - if it is defined: ``` In normal operation, opam...

Ok, I see what's happening, we just need to refine how information is filtered out. Here is the message with more info and less filtering: ``` * No agreement on...