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.
Strengthen state tracking in const-prop
Some/many of the changes are replicated between both the const-prop lint and the const-prop optimization.
Behaviour changes:
- const-prop opt does not give a span to propagated values. This was useless as that span's primary purpose is to diagnose evaluation failure in codegen.
- we remove the `OnlyPropagateInto` mode. It was only used for function arguments, which are better modeled by a write before entry.
- the tracking of assignments and discriminants make clearer that we do nothing in `NoPropagation` mode or on indirect places.
Strengthen state tracking in const-prop
Some/many of the changes are replicated between both the const-prop lint and the const-prop optimization.
Behaviour changes:
- const-prop opt does not give a span to propagated values. This was useless as that span's primary purpose is to diagnose evaluation failure in codegen.
- we remove the `OnlyPropagateInto` mode. It was only used for function arguments, which are better modeled by a write before entry.
- the tracking of assignments and discriminants make clearer that we do nothing in `NoPropagation` mode or on indirect places.
Add utility macros to help with writing tests.
Adds two utility macros to help with testing:
* `external` expands to it's argument tokens, but makes them appear to come from an external macro. Helps make tests for `in_external_macro` much more readable.
* `inline_macros` is an attribute macro which allows the use of a pseudo `inline!` macro which expands to it's argument tokens, but makes them appear to be from a crate-local macro expansion. This removes the need to write `macro_rules` boilerplate when testing how lints interact with macros.
---
`external`'s usage is simple. `external!(struct Foo { x: u32});` will make the struct appear as though it came from an external macro. Individual tokens can be escaped if needed. `external!($x + 0 / 10)` will make everything except `x` appear as though it came from an external macro. Can also use `$literal` and `$(tokens...)` as well.
---
`inline_macros` is more complicated due to compiler constraints. Given:
```rust
#[inline_macros]
fn foo() {
inline!(5 + 5 / 10);
}
```
`inline!(5 + 5 / 10)` will be replace with a call to a generated macro which expands to the contained tokens.
Tokens can be escaped by prefixing them with `$`:
```rust
#[inline_macros]
fn foo() {
let x = 5;
inline!($x + 5 / $10);
}
```
This will pass `x` as an `ident` argument and `10` as a `literal` argument.
Token sequences can also be passed with `$(...)`:
```rust
#[inline_macros]
fn foo() {
let mut x = 5;
inline!(if $(x >= 5) {
$x = 5;
});
}
```
This will pass `x >= 5` as `tt` arguments, and `x` as an `ident` argument.
---
Not 100% sure `inline_macros` is actually worth having. It does make the tests a little easier to read once you're used to it and it becomes more useful once there are multiple macro tests. The verbosity of declaring single use macros starts to hurt at that point.
changelog: None
Rollup of 8 pull requests
Successful merges:
- #108651 (Forbid the use of `#[target_feature]` on `main`)
- #109009 (rustdoc: use restricted Damerau-Levenshtein distance for search)
- #109026 (Introduce `Rc::into_inner`, as a parallel to `Arc::into_inner`)
- #109029 (Gate usages of `dyn*` and const closures in macros)
- #109031 (Rename `config.toml.example` to `config.example.toml`)
- #109032 (Use `TyCtxt::trait_solver_next` in some places)
- #109047 (typo)
- #109052 (Add eslint check for rustdoc-gui tester)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
Rename `config.toml.example` to `config.example.toml`
This had bothered me for a while as it leads to bad (missing) syntax highlighting in most editors I've used, and `@jyn514` suggested I just make the change and that the compatibility concerns I had don't really matter.
I suspect it will be a contentious one, so will not be offended if the outcome of this is to close the PR.
Gate usages of `dyn*` and const closures in macros
We silently accepted `dyn*` and const closures in macros as long as they didn't expand to anything containing these experimental features, unlike other gated features such as `for<'a>` binders on closures, etc. Let's not do that, to make sure nobody begins relying on this.
Introduce `Rc::into_inner`, as a parallel to `Arc::into_inner`
Unlike `Arc`, `Rc` doesn't have the same race condition to avoid, but
maintaining an equivalent API still makes it easier to work with both
`Rc` and `Arc`.
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!
Check that a query has not completed and is not executing before starting it
This fixes a race in the query system where we only checked if the query was currently executing, but not if it was already completed, causing queries to re-execute.
r? `@cjgillot`
Ensure value is on the on-disk cache before returning from `ensure()`.
The current logic for `ensure()` a query just checks that the node is green in the dependency graph.
However, a lot of places use `ensure()` to prevent the query from being called later. This is the case before stealing a query result.
If the query is actually green but the value is not available in the on-disk cache, `ensure` would return, but a subsequent call to the full query would run the code, and attempt to read from a stolen value.
This PR conforms the query system to the usage by checking whether the queried value is loadable from disk before returning.
Sadly, I can't manage to craft a proper test...
Should fix all instances of "attempted to read from stolen value".
