Hey everyone, how’s it going?
I discovered this language recently and I think it has a lot of potential. Turns out I’m in the position of rewriting legacy frontend code at my job, and although I would really like to go straight to ReScript, it seems like the language is not stable yet and things like using react-bootstrap are looking kinda rough right now. I’m by no means an advanced developer, so the lack of examples and updated bindings are going to give me a bad time.
Thing is, let’s say I would like to move to this language in the future: what would be the best decisions I could take upon the rewrite? Hope this isn’t too much of a silly question. Thanks!
Actually everything except the syntax is rock solid (build tooling, compiler) and the new syntax is also quite stable at the moment (with limited editor support). We have a pretty sizable codebase at my startup in it (still ReasonML). The only change is the name (Bucklescript to ReScript), and an improved syntax (Reason to ReScript) that makes it even more like JavaScript and is more elegant in a few ways.
If you don’t mind to not have autocomplete and type inference in your editor, you can go straight to the ReScript syntax. Otherwise you can work in ReasonML first and convert later (the syntaxes are really similar so the conversion would be only a small change).
The bindings ecosystem doesn’t matter that much in practice in my opinion. In fact most of the JS libraries we consume, we just write bindings to the functions that we use, it is a bit of work, but you have to learn the API surface anyway so I found that writing out the bindings helps me in better understanding the API I am working with. But for the bindings that are available, they are pretty stable and nothing’s really changed (they are all in ReasonML now, but 100% compatible with ReScript as it’s just a different syntax).
2 Likes