never_patterns: Parse match arms with no body
Never patterns are meant to signal unreachable cases, and thus don't take bodies:
```rust
let ptr: *const Option<!> = ...;
match *ptr {
None => { foo(); }
Some(!),
}
```
This PR makes rustc accept the above, and enforces that an arm has a body xor is a never pattern. This affects parsing of match arms even with the feature off, so this is delicate. (Plus this is my first non-trivial change to the parser).
~~The last commit is optional; it introduces a bit of churn to allow the new suggestions to be machine-applicable. There may be a better solution? I'm not sure.~~ EDIT: I removed that commit
r? `@compiler-errors`
Add `never_patterns` feature gate
This PR adds the feature gate and most basic parsing for the experimental `never_patterns` feature. See the tracking issue (https://github.com/rust-lang/rust/issues/118155) for details on the experiment.
`@scottmcm` has agreed to be my lang-team liaison for this experiment.
- Rename them both `as_str`, which is the typical name for a function
that returns a `&str`. (`to_string` is appropriate for functions
returning `String` or maybe `Cow<'a, str>`.)
- Change `UnOp::as_str` from an associated function (weird!) to a
method.
- Avoid needless `self` dereferences.
rustfmt currently has a runtime dependency on the sysroot. So when we
build a standalone rustfmt binary we need to set `LD_LIBRARY_PATH` so
each rustfmt binary knows where to find it's dependencies.
When running our Diff-Check job to test PRs for breaking changes it's
often the case that both the master rustfmt binary and the feature
branch binary have the same runtime dependencies so we only need to
set `LD_LIBRARY_PATH` once.
However, when running the diff-check job against a subtree sync PR that
assumption doesn't hold. The subtree sync PR bumps the required
toolchain used to build rustfmt and therefore the binary that gets built
for the subtree sync PR has a different runtime dependency than the
master rustfmt binary.
Now we set `LD_LIBRARY_PATH` twice to account for this potential
difference.
These changes mostly improve logging out the cargo version and version
of the two rustfmt binaries that are compiled. Some other minor logging
changes were made as well to add some whitespace to improve visual
clarity when looking at the logs in the GitHub Actions console.
Fixes 5912
When `control_brace_style = "AlwaysNextLine"`, the code seems to always assume that `body_prefix` is `{`. This is however not the case when `match_arm_blocks = false`. This causes `block_sep` to introduce extra white space that causes the error.
The fix was to check if `body_prefix` is empty before matching on `ControlBraceStyle::AlwaysNextLine`.
report `unused_import` for empty reexports even it is pub
Fixes#116032
An easy fix. r? `@petrochenkov`
(Discovered this issue while reviewing #115993.)
for now, let-chains can only be formatted on a single line if the chain
consits of 2 expressions where the first is an identifier proceeded by
any number of unary operators and the second is a let-expr.
