SI-9248 Lint warning when inferring ?F[_] = Any/Nothing#4401
SI-9248 Lint warning when inferring ?F[_] = Any/Nothing#4401retronym wants to merge 1 commit intoscala:2.11.xfrom
Conversation
|
Review by @adriaanm |
There was a problem hiding this comment.
My Webster's says one r in inference. I didn't know you could spell it inferrible. Like enfant inferrible. Thx again for another interesting answer, not to take you for granted.
There was a problem hiding this comment.
The typo stemmed from a copy/paste/edit from the neighbouring warning. Just pushed a a Webster's certified version.
There was a problem hiding this comment.
That was a "terrible" pun.
Any and Nothing are kind polymorphic, meaning the following is valid:
def f[F[A]] = 0; f[Any]; f[Nothing]
This commit warns, under -Xlint:infer-any, when we infer such
type appliations.
|
Suggestion at retronym#11 I had an idea to push the warning check into |
|
I think we should defer this. It makes for a great test of how far we could push abide. If someone would like to pick this up, I'd suggest merging the second |
4 participants
Any and Nothing are kind polymorphic, meaning the following is valid:
This commit warns, under -Xlint:infer-any, when we infer such
type appliations.