`unescape_literal` becomes `unescape_unicode`, and `unescape_c_string`
becomes `unescape_mixed`. Because rfc3349 will mean that C string
literals will no longer be the only mixed utf8 literals.
- Rename it as `MixedUnit`, because it will soon be used in more than
just C string literals.
- Change the `Byte` variant to `HighByte` and use it only for
`\x80`..`\xff` cases. This fixes the old inexactness where ASCII chars
could be encoded with either `Byte` or `Char`.
- Add useful comments.
- Remove `is_ascii`, in favour of `u8::is_ascii`.
Use upstream exhaustiveness checker!
Because it has been librarified!
The extra `Apache-2.0 WITH LLVM-exception` license is for `rustc_apfloat`. Also this duplicates `rustc_index` because the other upstream deps are still on an earlier version. They should be bumpable now though. Good thing is that we don't need this new crate to be synchronized with the others, which will make our lives easier.
internal: add `max_line_length` to `.editorconfig`
This helps with setting proper visual guides/rulers for the right margin in some editors/IDEs (e.g. CLion)
fix panic with reference in macro
it panic at `builder.make_mut(segment)`, where segment is from macro expand. And the usage reference in orginal macro call isn't a `PathSegment` so we can't update it in `apply_references`, I can't find a way to deal with it properly so here just filter out the reference in macro. LMK if there are better way to fix this
try to close https://github.com/rust-lang/rust-analyzer/issues/16328
Work through temporarily allowed clippy lints, part 1
This is the first batch of not allowing but actually fixing the clippy lints. Each commit removes one lint from the lint table and then fixes the resulting warnings.
Follow-up to #16401
