Hacker News Reader

Comment by elcapitan | Hacker News Reader

Is ReScript the same as ReasonML, and is it gaining real traction this time?

A couple of years ago I looked into Reason and really liked it, but the toolchain wasn't going anywhere and it didn't see a lot of adaption back then.

View on HN

Yes and no. The syntax has been updated to make it less OCaml and more JavaScript-like.

ReasonML can compile to native, rescript can’t. Reason still exists as a separate project, and projects that have been rewritten into reason will not necessarily switch.

In theory this is correct, but in practice Reason(ML) is dead. Or it was the last time I checked a month or two ago.

ReScript lives on though.

Such a pity they've become two projects. To have on lang for both BE and FE is really nice.

Well technically I guess you could still have that via node, just not native as with the Ocaml compiler?

And via Deno as well.

I rather work with OCaml/httpaf[0] than with JS/express or what have you.

[0]: https://github.com/inhabitedtype/httpaf

Honestly that was the selling point of ReasonML to me. Share business domain types and share common code but also get a native compiled backend. The pattern matching, variants, and syntax of Rescript are quite nice but unfortunately I don't find it enough of a selling point for frontend code that (in my experience) does not benefit as much from those previous points as say writing compilers or parsers because the majority of what youre writing is interacting with 3rd party libs and components, building forms etc in React, for which the code ends up looking pretty similar in rescript-react.

Exactly. I could take the whole "generate API consumer client lib from API specification" out of the equation.

We tried doing this with darklang. It ended up not actually working that well (it took a lot of effort to have even a few shared types), and it didn't get us very much either.

I think I read your postmortem on that actually. Good stuff

Reply on news.ycombinator.com