Konstantin A. Olkhovskiy

> perhaps the conclusion is that js_of_ocaml should introduce its own file suffixes, not reuse .cmo/.cma, and explicitly requires re-compilation This sounds pretty inconvenient for users that have single codebase...

> Many .opam files have no name or version for the parent component. Can these be made mandatory? I don't think this can be made mandatory. Some folks add version...

One more link on opam supply chain https://github.com/ocaml/opam/wiki/Opam-supply-chain

So far I've just removed generating of code for group creation, so that command line only generates some `build(g *temple.Group)` function which puts all the pieces into the group. This...

Why not just glob proto files in dune recursively (typically they are organized as a hierarchy corresponding to package names etc), and produce one output by `ocaml-protoc` as the result,...

Since empty messages work just fine with our internal fork, I looked through the commits and found this one: ```diff diff --git a/vendor/ocaml-protoc/src/compilerlib/pb_codegen_backend.ml b/vendor/ocaml-protoc/src/compilerlib/pb_codegen_backend.ml index 84f12e2..d6dcf44 100644 --- a/vendor/ocaml-protoc/src/compilerlib/pb_codegen_backend.ml +++...

Actually that patch was already upstreamed by me some time ago to both `master` and `br-4.0` branches. I've checked master branch (e2cf9037), and for exactly the same input file: ```proto...

@zshipko what's the latest on OCaml 5 support for ocaml-rs? I can't seem to find relevant information in README...

Hm, looks like this is the tracking issue in ocaml-interop: https://github.com/tizoc/ocaml-interop/issues/58 which further mentions this MR: https://gitlab.com/ocaml-rust/ocaml-boxroot/-/merge_requests/75, which is still under review (seems like final stages, fingers crossed).

I've came up with one idea, and it seems to be working and solves this issue. Basically it boils down to piggyback on cargo vendoring and versioning capabilities to distribute...