Tweak chained operators diagnostic
Use more selective spans
Improve suggestion output
Be more selective when displaying suggestions
Silence some knock-down type errors
r? @Centril
Fix smaller issues with invalid placeholder type errors
Follow up to #70294.
- Fix placement of suggested generic param when bounds are present.
- Reduce error duplication for invalid placeholder types in `fn` types.
r? @Centril
resolve: Remove `rustc_attrs` as a standalone feature gate
Now it only gates specific built-in attributes.
So if you want to make a rustc attribute, make it a built-in (this was already the case in practice for some time).
Miri: move ModifiedStatic to ConstEval errors
Also generally adjust terminology from "static" to "global" where appropriate (to avoid confusion with specifically `static` items).
Store idents for `DefPathData` into crate metadata
Previously, we threw away the `Span` associated with a definition's
identifier when we encoded crate metadata, causing us to lose location
and hygiene information.
We now store the identifier's `Span` in a side table, which gets encoded
into the crate metadata. When we decode items from the metadata, we
combine the name and span back into an `Ident`.
This improves the output of several tests, which previously had messages
suppressed due to dummy spans.
This is a prerequisite for #68686, since throwing away a `Span` means
that we lose hygiene information.
Expand: nix all fatal errors
Basically, we go after all `.span_fatal` / `FatalError.raise()` and similar things and remove them one by one until there are no fatal errors left.
r? @petrochenkov
Add regression test for #70155.
With #70166 merged, `RangeInclusive` now derives `PartialEq` and `Eq`, implementing structural equality and as a side effect the range is now usable with const generics, closing #70155.
As per [#70166 (comment)](https://github.com/rust-lang/rust/pull/70166#issuecomment-601872201) a test is added to avoid a change to the private fields or the equality implementation of the range from subtly reverting #70155.
Fix invalid suggestion on `&mut` iterators yielding `&` references
Fixes#69789.
rustc suggested an invalid code when `&` reference from `&mut` iterator is mutated. The compiler knew we're mutating a value behind `&` reference, but as the assignment RHS is from desugaring, it could only see the iterator expression from source and inserted `mut` there.
r? @estebank
