fix: find Self reference
#15864
fix: find Self reference
#15864
Conversation
rustbot
added
the
S-waiting-on-review
|
@bors r+ |
|
☀️ Test successful - checks-actions |
| @@ -584,7 +584,7 @@ impl<'a> FindUsages<'a> { | |||
| ) -> bool { | |||
| match NameRefClass::classify(self.sema, name_ref) { | |||
| Some(NameRefClass::Definition(Definition::SelfType(impl_))) | |||
| if impl_.self_ty(self.sema.db) == *self_ty => | |||
| if impl_.self_ty(self.sema.db).as_adt() == self_ty.as_adt() => | |||
There was a problem hiding this comment.
Hmm, this doesn't seem right to me, now we discard any other type that could appear as the self type in an impl. I'm curious as to why this equality fails in the given example in the first place. Sounds like we are losing some type info along the way somewhere?
There was a problem hiding this comment.
I am curious too. debug to that point and found the unequality. Is there anyother type for Self in an impl beside Adt?
Actually I don't quite understand the whole implementation of find refenence, it‘s complicated for me :(
There was a problem hiding this comment.
trait T {}
impl T for fn() {}is an example. Don't think you can trigger references highlighting on that, but other code might require the Self resolution in search to work for that. I can take a look at that next week and see what's going on here
There was a problem hiding this comment.
And yes, the reference search is pretty messy looking and undocumented so I don't blame you for not being able to fully grasp the implementation there :)
There was a problem hiding this comment.
Ah the problem is that the self_ty here contains an {error} for the generic param where as the impl.self_ty() returns the struct with a var for the generic parameter, so the equality check fails. Hm
took a lot of time to debug to find the problem, here should compare the actual
Adttypeclose #12693