rustdoc: Don't inline #[doc(hidden)] pub use
Currently if a `#[doc(hidden)] pub use` item is inlined the `hidden`
attribute is ignored so the item can appear in the docs. By never inlining
such imports, they can be stripped.
An example in `std` is [`__OsLocalKeyInner`](https://doc.rust-lang.org/nightly/std/thread/struct.__OsLocalKeyInner.html) which clearly should not be documented.
It turns out that the subsequent lines of the error message comment
should be aligned like this.
The "turns the corresponding compiler warning" language may not be
strictly the most accurate—a lint check isn't the same as a compiler
warning; it emits a compiler warning if it's set to the `warn` level—
but it may be worth glossing over such distinctions in favor of simple,
familar phrasings for the sake of pedagogy; thanks to Guillaume Gomez
for the wording suggestion.
Let's also fix up the introductory clauses of the sentences about how to
fix the error to put a little more emphasis on the fact that the
`forbid` setting was probably there for a reason.
Add explanations for E0503 and E0508.
(cannot use `..` because it was mutably borrowed, cannot move out of type `..`, a non-copy fixed-size array)
Part of #32777.
add a test case for issue #32031
I propose a test case to finish the fix for issue #32031. Please review this commit thoroughly, as I have never written a codegen test before.
r? @eddyb
rustdoc: Don't generate empty files for stripped items
We need to traverse stripped modules to generate redirect pages, but we shouldn't generate
anything else for them.
This now renders the file contents to a Vec before writing it to a file in one go. I think
that's probably a better strategy anyway.
Fixes: #34025
Currently if a `#[doc(hidden)] pub use` item is inlined the `hidden`
attribute is ignored so the item can appear in the docs. By never inlining
such imports, they can be stripped.
Remove linking and intrinsics code made dead by only supporting LLVM 3.7 and up
This is mostly based on Alex's throwaway comment:
> probably reject those that LLVM just doesn't support...
So I'm more than happy to adjust the PR based on how you thought this should look. Also happy to split it into two PRs, one for linking and one for intrinsics.
r? @alexcrichton
/cc @nagisa @brson
Treat `#[test]` like `#[cfg(test)]` in non-test builds
This PR treats `#[test]` like `#[cfg(test)]` in non-test builds. In particular, like `#[cfg(test)]`,
- `#[test]` nodes are stripped during `cfg` processing, and
- `#[test]` is disallowed on non-optional expressions.
Closes#33946.
r? @nrc
