Given a :: T, what should the return type be of that expression? m T or m ()?
I think "do a <- return 1" should be legal in Haskell, even with nothing after it, merely that the "a" is an unused variable warning. That makes incrementally typed code valid more often, which helps with immediate-feedback IDEs.
-
-
-
I'd go for m T - it's an unused variable warning, there's no () written anywhere.
- Još 2 druga odgovora
Novi razgovor -
-
-
I have found that I tend to put `undefined` at the end of my `do` blocks while writing code for this very reason. And, yes, my "solution" is kinda not very satisfying.
-
Tweet je nedostupan.
Novi razgovor -
-
-
This is something that shouldn't matter but actually does, this actually slows me down yes mhm.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
What would it desugar to?
-
return 1
- Još 3 druga odgovora
Novi razgovor -
-
-
I don't know what you intend it to mean. And since it's not obvious what it means, it makes me nervous.
-
It means the compiler can implicitly insert a `return ()` at the end. Heck, even make the entire do expression be bottom, but at least make it parse and make it typecheck.
Kraj razgovora
Novi razgovor -
-
-
I'd be ok with a flag that makes ghc more lenient in allowing this, but would rather consider it invalid code by default.
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.