ben.knoble
2021-7-9 12:41:38

My (vague and probably wrong) understanding is that when you require a non-typed module in a typed module, you have to supply the types. (If it’s a module you use a lot, might be worth wrapping it in your own typed module.) I have no idea how this applies to macros, especially since -> is just reordering expressions.


agj_chile
2021-7-9 13:10:36

Yep, you’re right! Thanks. I’m not sure how that applies to such macros though, that’s what’s tripping me up :white_frowning_face:



agj_chile
2021-7-9 14:29:39

You’re right, Ben :confused:


ben.knoble
2021-7-9 16:26:56

@samth are you the “typed/racket” person? I forget


samth
2021-7-9 16:27:09

Yes


ben.knoble
2021-7-9 16:30:30

Do you know where it’s documented or any details about using untyped macros in typed code?


raoul.schorer
2021-7-9 20:03:22

Is it possible to use for/fold/derived same as in #lang racket to make a new for form in #lang typed/racket ? Is there some documentation somewhere regarding the interaction with the typechecker?


raoul.schorer
2021-7-9 20:08:33


samth
2021-7-10 01:37:08

But in general there isn’t necessarily going to be a way to make arbitrary macros work; I can take a look at threading later but it might be hard.


capfredf
2021-7-10 01:47:49

to my knowledge, Typed Racket currently doesn’t support for/fold/derived


agj_chile
2021-7-10 03:47:23

Thank you, Sam! I was able to use ann to annotate the output of λ~>> , which is what was giving me trouble it seems. Sadly makes it a bit unwieldy to use…