Nicolás Ojeda Bär

@Athishpranav2003 the simplest approach is to try mapping the absolute path to a relative path, relative to the workspace root, and then continue with the already-existing logic. I don't think...

`dune utop` works fine if the code being loaded into the toplevel depends on one or more of the libraries `utop` itself links against (eg `lwt`, `react`). On the other...

> I found some related code in the utop repo, I understand the suggested approach would be to add `React` to [this list](https://github.com/ocaml-community/utop/blob/fcb7903603cec5e8a2bb4371ea43b1c01f261170/src/top/expunge/modules.ml#L40-L42)? Yes. But to be clear, I am...

This PR - is coherent with the other simplifications that have been done in the functor syntax, - the actual diff is tiny, - it aligns the module language with...

There seems to be a consensus in favour, and no strong objections. @yallop: if you can - resolve the conflict, and - address @gasche's comment [above](https://github.com/ocaml/ocaml/pull/12828#discussion_r1430939354) then I think we...

I believe that m68k is a 32-bit architecture, in which case it is normal for `jumptbl_base` to be 0, see: https://github.com/ocaml/ocaml/blob/f1c28574accca8a8693b5097d32e0d1d4a19e906/runtime/interp.c#L56-L61

Not only they are not installed, they are not even built in the current build system.

> I'm not sure why bash is considered a fallback for command length limits however. Surely that is one thing you can't expect to find on windows? In practice, until...

> Historically, `flexlink` eschews using the Unix library, but this is daft @dra27 You are the expert, but won't this cause issues when bootstrapping flexlink?

Nit, but why not just `(open ...)`?