Associated types are highlighted as unresolved references

[lunacookies]

The dark red in the following screenshot represents unresolved references:

Curiously, this example from #6995 works fine in the method definition, but not the implementation:

Possibly related: #5003, #6995, #8283

Output of Show RA Version: nightly-50-g354151df3

[Veykril]

The Self::Bar in the fn foo(_: Self::Bar) line in the latter screenshot does resolve for me, just not to i32 but to the general type alias of the trait definition(which I think is another issue with us not being able to resolve back to trait impls yet).

The code in your first screenshot uses inherent type aliases which aren't stable, I don't think RA even handles those at all aside from parsing. At least for name classification we don't expect those.

[lunacookies]

@Veykril

does resolve for me, just not to i32 but to the general type alias of the trait definition

That is curious. What version of r-a are you on?

The code in your first screenshot uses inherent type aliases which aren’t stable

Oops, you’re right! Here’s a version that does compile but still has unresolved references:

[Veykril]

I'm on the latest nightly.

Your other screenshot seems to be the same as #8282, though I guess the title there can be improved

[lunacookies]

Your other screenshot seems to be the same as #8282

Hmm, you’re right, I didn’t see that issue. Should I close this because there seem to already be issues for a number of different cases regarding associated types?

[Veykril]

If that's the only problem left then I'd say yes let's close this