1743: Support `#[macro_use]` on `extern crate` r=matklad a=uHOOCCOOHu
Unfortunately, #1688 is still an issue. My guess is wrong :(
Co-authored-by: uHOOCCOOHu <hooccooh1896@gmail.com>
1757: Assoc type bindings r=flodiebold a=flodiebold
This adds support for type bindings (bounds like `where T: Iterator<Item = u32>`).
It doesn't yet work in as many situations as I'd like because of some [Chalk problems](https://github.com/rust-lang/chalk/issues/234). But it works in some situations, and will at least not bitrot this way ;)
(part of the problem is that we use `Normalize` to normalize associated types, but produce `ProjectionEq` goals from where clauses, so Chalk can't normalize using the environment; this would be fixed by using `ProjectionEq` for normalization, which I think is the 'proper' way, but then we'd run into those ambiguity problems everywhere...)
Co-authored-by: Florian Diebold <flodiebold@gmail.com>
It's a bit complicated because we basically have to 'undo' the desugaring, and
the result is very dependent on the specifics of the desugaring and will
probably produce weird results otherwise.
1752: Always wrap blocks into block expressions r=flodiebold a=matklad
This way, things like function bodies are expressions, and we don't have to single them out
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
1737: Report type mismatches in analysis-stats r=matklad a=flodiebold
Only the number usually; each one individually when running with `-v`.
Getting the file/line locations for the exprs was really annoying and I had to make some stuff public (that I didn't remember why it would be `pub(crate)`); maybe I missed some easier way? It would be nice to have some general way for mapping locations 🤔
This reports 1768 mismatches on RA currently; from skimming, this seems to be mostly various kinds of coercions, though there were also some other things.
Co-authored-by: Florian Diebold <flodiebold@gmail.com>
