Fix NLL typeck ICEs
* Don't ICE when a type containing a region is constrained by nothing
* Don't ICE trying to normalize a type in a `ParamEnv` containing global bounds.
To explain what was happening in the `issue-61311-normalize.rs` case:
* When borrow checking the `the_fn` in the last `impl` we would try to normalize `Self::Proj` (`<Unit as HasProjFn>::Proj`).
* We would find the `impl` that we're checking and and check its `where` clause.
* This would need us to check `<Box<dyn Obj + 'static> as HasProj>::Proj: Bound`
* We find two possible implementations, the blanket impl and the bound in our `ParamEnv`.
* The bound in our `ParamEnv` was canonicalized, so we don't see it as a global bound. As such we prefer it to the `impl`.
* This means that we cannot normalize `<Box<dyn Obj + 'static> as HasProj>::Proj` to `Unit`.
* The `<Box<dyn Obj + 'static> as HasProj>::Proj: Bound` bound, which looks like it should be in our `ParamEnv` has been normalized to `Unit: Bound`.
* We fail to prove `<Box<dyn Obj + 'static> as HasProj>::Proj: Bound`.
* We ICE, since we believe typeck have errored.
Closes#61311Closes#61315Closes#61320
r? @pnkfelix
cc @nikomatsakis
Add new diagnostic writer using annotate-snippet library
This adds a new diagnostic writer `AnnotateRsEmitterWriter` that uses
the [`annotate-snippet`][as] library to print out the human readable
diagnostics.
The goal of #59346 is to eventually switch over to using the library instead of
maintaining our own diagnostics output.
This PR does **not** add all the required features to the new
diagnostics writer. It is only meant as a starting point so that other
people can start contributing as well.
There are some FIXMEs in `librustc_errors/annotate_rs_emitter.rs` that
point at yet to be implemented features of the new diagnostic emitter, however
those are most likely not exhaustive.
[as]: https://github.com/rust-lang/annotate-snippets-rs
Make cannot move errors more consistent with other borrowck errors
* Note the type of the place being moved in all cases.
* Note the place being moved from.
* Simplify the search for overloaded place operators
* Extend the note for move from overloaded deref apply to all types.
* Add a note for moves from overloaded index.
* Special case moves for closure captures.
r? @pnkfelix
Re-implement async fn drop order lowering
This PR re-implements the async fn drop order lowering changes so
that it all takes place in HIR lowering, building atop the work done by
@eddyb to refactor `Res::Upvar`.
Previously, this types involved in the lowering were constructed in
libsyntax as they had to be used during name resolution and HIR
lowering. This was awful because none of that logic should have existed
in libsyntax.
This commit also changes `ArgSource` to keep a `HirId` to the original
argument pattern rather than a cloned copy of the pattern.
Only b7aa4ed and 71fb8fa should be reviewed, any other commits
are from #61276 (though 447e336 might end up staying in this PR).
As a nice side effect, it also fixes#61187 (cc #61192).
r? @eddyb
cc @cramertj
add support for unchecked math
add compiler support for
```rust
/// Returns the result of an unchecked addition, resulting in
/// undefined behavior when `x + y > T::max_value()` or `x + y < T::min_value()`.
pub fn unchecked_add<T>(x: T, y: T) -> T;
/// Returns the result of an unchecked substraction, resulting in
/// undefined behavior when `x - y > T::max_value()` or `x - y < T::min_value()`.
pub fn unchecked_sub<T>(x: T, y: T) -> T;
/// Returns the result of an unchecked multiplication, resulting in
/// undefined behavior when `x * y > T::max_value()` or `x * y < T::min_value()`.
pub fn unchecked_mul<T>(x: T, y: T) -> T;
```
cc https://github.com/rust-lang/rfcs/issues/2508
This commit removes the `HirId` from `ArgSource::AsyncFn`, relying on
the fact that only `simple_ident` is used in each of the locations that
previously took the original pattern from the `ArgSource::AsyncFn`.
This commit re-implements the async fn drop order lowering changes so
that it all takes place in HIR lowering, building atop the work done by
`@eddyb` to refactor `Res::Upvar`.
Previously, this types involved in the lowering were constructed in
libsyntax as they had to be used during name resolution and HIR
lowering. This was awful because none of that logic should have existed
in libsyntax.
This commit also changes `ArgSource` to keep a `HirId` to the original
argument pattern rather than a cloned copy of the pattern.
Add more detail to type inference error
When encountering code where type inference fails, add more actionable
information:
```
fn main() {
let foo = Vec::new();
}
```
```
error[E0282]: type annotations needed in `std::vec::Vec<T>`
--> $DIR/vector-no-ann.rs:2:16
|
LL | let foo = Vec::new();
| --- ^^^^^^^^ cannot infer type for `T` in `std::vec::Vec<T>`
| |
| consider giving `foo` a type
```
Fix#25633.
Point at individual type args on arg count mismatch
- Point at individual type arguments on arg count mismatch
- Make generics always have a valid span, even when there are no args
- Explain that `impl Trait` introduces an implicit type argument
Fix#55991.
