Fix "Directly go to item in search if there is only one result" setting
Part of #66181.
The setting was actually broken, so I fixed it when I added the GUI test.
r? `@notriddle`
rustdoc: Optimize impl sorting during rendering
This should fix the perf regression on [bitmaps-3.1.0](https://github.com/rust-lang/rustc-perf/tree/master/collector/compile-benchmarks/bitmaps-3.1.0) from https://github.com/rust-lang/rust/pull/107765.
The bitmaps crate has a lot of impls:
```rust
impl Bits for BitsImpl<1> { ... }
impl Bits for BitsImpl<2> { ... }
// ...
impl Bits for BitsImpl<1023> { ... }
impl Bits for BitsImpl<1024> { ... }
```
and the logic in `fn print_item` sorts them in natural order.
Before https://github.com/rust-lang/rust/pull/107765 the impls came in source order, which happened to be already sorted in the necessary way.
So the comparison function was called fewer times.
After https://github.com/rust-lang/rust/pull/107765 the impls came in "stable" order (based on def path hash).
So the comparison function was called more times to sort them.
The comparison function was terribly inefficient, so it caused a large perf regression.
This PR attempts to make it more efficient by using cached keys during sorting.
rustdoc: add support for type filters in arguments and generics
This makes sense, since the search index has the information in it, and it's more useful for function signature searches since a function signature search's item type is, by definition, some type of function (there's more than one, but not very many).
Render source page layout with Askama
~~I was looking at making `code_html` render into the buffer instead of in advance, but it turned out to need a pretty big refactor, so starting with rearranging the high-level layout.~~
Found another approach which required much less changes
cc #108868
This makes sense, since the search index has the information in it,
and it's more useful for function signature searches since a
function signature search's item type is, by definition, some type
of function (there's more than one, but not very many).
rustdoc: Remove footnote references from doc summary
Since it's one line, we don't have the footnote definition so it doesn't make sense to have the reference.
Part of https://github.com/rust-lang/rust/issues/109024.
r? `@notriddle`
rustdoc: implement bag semantics for function parameter search
This tweak to the function signature search engine makes things so that, if a type is repeated in the search query, it'll only match if the function actually includes it that many times.
This tweak to the function signature search engine makes things so that,
if a type is repeated in the search query, it'll only match if the
function actually includes it that many times.
rustdoc: reduce allocs in FnDecl::inner_full_print
Instead of maintaining parallel buffers for both HTML and non-HTML output, follow the idiom from the rest of format.rs that f.alternate() == true means textual output. Also, add an argument to control line wrapping explicitly.
This allows the caller to render once with textual output and no line wrapping, to decide whether line wrapping should be applied in the final HTML output.
Also, remove some format! and " ".repeat calls, and remove a dependency on calling `String::replace` to switch from newlines to spaces.
This coincidentally fixes some minor bugs where the old code was undercounting the number of characters for a declaration in text mode.
rustdoc: use restricted Damerau-Levenshtein distance for search
Based on https://github.com/rust-lang/rust/pull/108200, for the same rationale.
> This replaces the existing Levenshtein algorithm with the Damerau-Levenshtein algorithm. This means that "ab" to "ba" is one change (a transposition) instead of two (a deletion and insertion). More specifically, this is a restricted implementation, in that "ca" to "abc" cannot be performed as "ca" → "ac" → "abc", as there is an insertion in the middle of a transposition. I believe that errors like that are sufficiently rare that it's not worth taking into account.
Before this change, searching [`prinltn!`] listed `print!` first, followed by `println!`. With this change, `println!` matches more closely.
[`prinltn!`]: https://doc.rust-lang.org/nightly/std/?search=prinltn!
Instead of maintaining parallel buffers for both HTML and non-HTML output,
follow the idiom from the rest of format.rs that f.alternate() == true means
textual output. Also, add an argument to control line wrapping explicitly.
This allows the caller to render once with textual output and no line wrapping,
to decide whether line wrapping should be applied in the final HTML output.
Also, remove some format! and " ".repeat calls, and remove a dependency on
calling `String::replace` to switch from newlines to spaces.
This coincidentally fixes some minor bugs where the old code was undercounting
the number of characters for a declaration in text mode.
Based on https://github.com/rust-lang/rust/pull/108200, for the same
rationale.
> This replaces the existing Levenshtein algorithm with the
> Damerau-Levenshtein algorithm. This means that "ab" to "ba" is one change
> (a transposition) instead of two (a deletion and insertion). More
> specifically, this is a restricted implementation, in that "ca" to "abc"
> cannot be performed as "ca" → "ac" → "abc", as there is an insertion in the
> middle of a transposition. I believe that errors like that are sufficiently
> rare that it's not worth taking into account.
Before this change, searching `prinltn!` listed `print!` first, followed
by `println!`. With this change, `println!` matches more closely.
Add more license annotations
This PR updates the `.reuse/dep5` file to include more accurate licensing data for everything in the repository (*excluding* submodules and dependencies). Some decisions were made in this PR:
* The standard copyright attribution for files maintained by us is "The Rust Project Developers (see https://thanks.rust-lang.org)", to avoid having to maintain an in-tree `AUTHORS` file.
* For files that have specific licensing terms, we added the terms to the `.reuse/dep5` rather than adding SPDX comments in the files themselves.
* REUSE picks up any comment/text line with `Copyright` on it, so I had to sprinkle around `REUSE-IgnoreStart` and `REUSE-IgnoreEnd` comments.
The rendered `COPYRIGHT` file is available at https://gist.github.com/pietroalbini/efb81103f69596d39758114f3f6a8688.
r? `@pnkfelix`
This commit makes the `clean::Type::is_same` non-commutative, so
that a generic `impl` matches a concrete return, but a generic return
does not match a concrete `impl`. It makes slice and vector Write
for `u8` not match on every generic return value.
