Fix doc cfg on reexports
Fixes#83428.
The problem was that the newly inlined item cfg propagation was not working since its real parent is different than its current one.
For the implementation, I decided to put it directly into `CfgPropagation` instead of inside `inline.rs` because I thought it would be simpler to maintain and to not forget if new kind of items are added if it's all done in one place.
r? `@notriddle`
rustdoc: strategic boxing to reduce the size of ItemKind and Type
The `Type` change redesigns `QPath` to box the entire data structure instead of boxing `self_type` and the `trait_`.
This reduces the size of several `ItemKind` variants, leaving `Impl` as the biggest variant. The `ItemKind` change boxes that variant's payload.
rustdoc: simplify highlight.rs
Split render_with_highlighting, which took many optional parameters, into three functions for specific purposes, which each take a smaller number of mostly required parameters.
Remove some plumbing to pass through an "edition" parameter, which was used solely to avoid highlighting some 2021 Edition keywords in non-2021 code.
I've tested a build of std docs before and after, and this does not change the generated HTML at all.
Followup from https://github.com/rust-lang/rust/pull/91264#discussion_r901151101
r? ```@GuillaumeGomez```
Split render_with_highlighting, which took many optional parameters, into three
functions for specific purposes, which each take a smaller number of mostly
required parameters.
Remove some plumbing to pass through an "edition" parameter, which was used
solely to avoid highlighting some 2021 Edition keywords in non-2021 code.
Replace boolean argument for print_where_clause with an enum to make code more clear
As you suggested ``@notriddle.`` Just not sure if the naming seems good to you?
r? ``@notriddle``
rustdoc: Censor certain complex unevaluated const exprs
Fixes#97933.
This is more of a hotfix for the aforementioned issue. By that, I mean that my proposed patch is
not the best solution but one that does not change as much existing code.
It treats symptoms rather than the root cause.
This PR “censors” certain complex unevaluated constant expressions like `match`es, blocks, function calls, struct literals etc. by pretty-printing them as `_` / `{ _ }` (number and string literals, paths and `()` are still printed as one would expect).
Resorting to this placeholder is preferable to printing the full expression verbatim since
they can be quite large and verbose resulting in an unreadable mess in the generated documentation.
Further, mindlessly printing the const would leak private and `doc(hidden)` struct fields (#97933), at least in the current
stable & nightly implementations which rely on `span_to_snippet` (!) and `rustc_hir_pretty::id_to_string`.
The censoring of _verbose_ expressions is probably going to stay longer term.
However, in regards to private and `doc(hidden)` struct fields, I have a more proper fix in mind
which I have already partially implemented locally and for which I am going to open a separate PR sometime soon.
For that, I was already in contact with `@GuillaumeGomez.`
The proper fix involves rustdoc not falling back on pretty-printing unevaluated consts so easily (what this PR is concerned about)
and instead preferring to print evaluated consts which contain more information allowing it to selectively hide private and `doc(hidden)` fields, create hyperlinks etc. generally making the output more granular and precise (compared to the brutal `_` placeholder).
Unfortunately, I was a bit too late and the issue just hit stable (1.62).
Should this be backported to beta or even a potential 1.62.1?
r? `@GuillaumeGomez`
This reduces the size of the function signature index, because
it's common to have many functions that operate on the same types.
$ wc -c search-index-old.js search-index-new.js
5224374 search-index-old.js
3932314 search-index-new.js
By my math, this reduces the uncompressed size of the search index by 32%.
On compressed signatures, the wins are less drastic, a mere 8%:
$ wc -c search-index-old.js.gz search-index-new.js.gz
404532 search-index-old.js.gz
371635 search-index-new.js.gz
Now that the "All Crates" dropdown is only rendered on the search results page,
there is no need to load crates.js on most pages. Load it only on crate pages.
Also, add the `defer` attribute so it does not block page rendering.
For sidebar-items.js, move the script tag to `<head>`. Since it already has the
defer attribute it won't block loading. The defer attribute does preserve
ordering between scripts, so instead of the callback on load, it can set a
global variable on load, which is slightly simpler. Also, since it is required
to finish rendering the page, beginning its load earlier is better.
Remove generation and handling of sidebar-vars. Everything there can be computed
with information available in JS via other means.
Remove the "other" wrapper in the sidebar. It was unnecessary.
Remove excess script fields
Hack: many traits and types in std are re-exported from core or alloc. In
general, rustdoc is capable of recognizing these implementations as being
on local types. However, in at least one case, rustdoc gets confused and
labels an implementation as being on a foreign type. To make sure that
confusion doesn't pass on to the reader, consider all implementations in
std, core, and alloc to be on local types.
Remove `#[rustc_deprecated]`
This removes `#[rustc_deprecated]` and introduces diagnostics to help users to the right direction (that being `#[deprecated]`). All uses of `#[rustc_deprecated]` have been converted. CI is expected to fail initially; this requires #95958, which includes converting `stdarch`.
I plan on following up in a short while (maybe a bootstrap cycle?) removing the diagnostics, as they're only intended to be short-term.
