rustdoc: render doc(hidden) as a code attribute

[chojs23]

Move #[doc(hidden)] into the shared code-attribute renderer so it matches the styling and placement of other attributes in rustdoc HTML.

Closes #132304

[rustbot]

r? @GuillaumeGomez

rustbot has assigned @GuillaumeGomez.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[GuillaumeGomez]

In case open_tag is None, we could remove the need for the if check by creating two different closures:

Suggested change

	let mut render_attr = |attr: &str| -> fmt::Result {
	if !wrote_any {
	if let Some(open_tag) = open_tag {
	w.write_str(open_tag)?;
	}
	wrote_any = true;
	}
	render_code_attribute(prefix, attr, w)
	};
	let mut render_attr: FnMut(&str) -> fmt::Result = if let Some(open_tag) = open_tag {
	|attr: &str| -> fmt::Result {
	if !wrote_any {
	w.write_str(open_tag)?;
	wrote_any = true;
	}
	render_code_attribute(prefix, attr, w)
	}
	} else {
	|attr: &str| -> fmt::Result {
	render_code_attribute(prefix, attr, w)
	}
	};

Not sure it compiles but with a few tweaks it should (or use functions instead, it works too).

[chojs23]

Using two closures in an if expression does not compile because each closure has a unique anonymous type. So I need boxing or enum to make one concrete type which adds allocation or extra codes.

I used the empty string instead, WDYT?

    let open_tag = open_tag.unwrap_or("");
    let mut render_attr = |attr: &str| -> fmt::Result {
        if !wrote_any {
            w.write_str(open_tag)?;
            wrote_any = true;
        }
        render_code_attribute(prefix, attr, w)
    };

[GuillaumeGomez]

The problem is also that it adds more checks. Did you try with functions instead of closures?

[chojs23]

Hmm i tried it but using function does not reduce the checks, it only changes the style.

I moved the <dt><code> write into render_attributes_in_code_with_options by adding an open_tag parameter.

[rustbot]

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

[rust-bors]

☔ The latest upstream changes (presumably #151232) made this pull request unmergeable. Please resolve the merge conflicts.