feat: Introduce workspace `rust-analyzer.toml`s
In order to globally configure a project it was, prior to this PR, possible to have a `ratoml` at the root path of a project. This is not the case anymore. Instead we now let ratoml files that are placed at the root of any workspace have a new scope called `workspace`. Although there is not a difference between a `workspace` scope and and a `global` scope, future PRs will change that.
feat: Use spans for builtin and declarative macro expansion errors
This should generally improve some error reporting for macro expansion errors. Especially for `compile_error!` within proc-macros
feat(ide-completion): explictly show `async` keyword on `impl trait` methods
OLD:
<img width="676" alt="image" src="https://github.com/user-attachments/assets/f6fa626f-6b6d-4c22-af27-b0755e7a6bf8">
Now:
<img width="684" alt="image" src="https://github.com/user-attachments/assets/efbaac0e-c805-4dd2-859d-3e44b2886dbb">
---
This is an preparation for https://github.com/rust-lang/rust-analyzer/issues/17719.
```rust
use std::future::Future;
trait DesugaredAsyncTrait {
fn foo(&self) -> impl Future<Output = usize> + Send;
fn bar(&self) -> impl Future<Output = usize> + Send;
}
struct Foo;
impl DesugaredAsyncTrait for Foo {
fn foo(&self) -> impl Future<Output = usize> + Send {
async { 1 }
}
//
async fn bar(&self) -> usize {
1
}
}
fn main() {
let fut = Foo.bar();
fn _assert_send<T: Send>(_: T) {}
_assert_send(fut);
}
```
If we don't distinguish `async` or not. It would be confusing to generate sugared version `async fn foo ....` and original form `fn foo` for `async fn in trait` that is defined in desugar form.
fix: let glob imports override other globs' visibility
Follow up to #14930Fixes#11858Fixes#14902Fixes#17704
I haven't reworked the code here at all - I don't feel confident in the codebase to do so - just rebased it onto the current main branch and fixed conflicts.
I'm not _entirely_ sure I understand the structure of the `check` function in `crates/hir-def/src/nameres` tests. I think the change to the test expectation from #14930 is correct, marking the `crate::reexport::inner` imports with `i`, and I understand it to mean there's a specific token in the import that we can match it to (in this case, `Trait`, `function` and `makro` of `pub use crate::defs::{Trait, function, makro};` respectively), but I had some trouble understanding the meaning of the different parts of `PerNs` to be sure.
Does this make sense?
I tested building and using RA locally with `cargo xtask install` and after this change the documentation for `arrow_array::ArrowPrimitiveType` seems to be picked up correctly!
feat: use vscode log format for client logs
This change updates the log format to use the vscode log format instead
of the custom log format, by replacing the `OutputChannel` with a
`LogOutputChannel` and using the `debug`, `info`, `warn`, and `error`
methods on it. This has the following benefits:
- Each log level now has its own color and the timestamp is in a more
standard format
- Inspect output (e.g. the log of the config object) is now colored
- Error stack traces are now shown in the output
- The log level is now controlled on the output tab by clicking the gear
icon and selecting "Debug" or by passing the `--log` parameter to
vscode. The `trace.extension` setting has been marked as deprecated.
Motivation:
The large uncolored unformatted log output with a large config object logged whenever it changes has always dominated the logs. This subjectively has made it that looking to see what the client is doing has always been a bit disappointing. That said, there's only 17 log messages total in the client. Hopefully by making the logs more visually useful this will encourage adding more appropriate debug level messages in future.
Incidentally, it might be worth only logging the config change message at a debug level instead of an info level to reduce the noise.
Update compiler_builtins to 0.1.114
The `weak-intrinsics` feature was removed from compiler_builtins in https://github.com/rust-lang/compiler-builtins/pull/598, so dropped the `compiler-builtins-weak-intrinsics` feature from alloc/std/sysroot.
In https://github.com/rust-lang/compiler-builtins/pull/593, some builtins for f16/f128 were added. These don't work for all compiler backends, so add a `compiler-builtins-no-f16-f128` feature and disable it for cranelift and gcc.
Selects a rust-toolchain declared RA based on its date. The earliest (oldest) RA wins and becomes the one that the workspace uses as a whole.
In terms of precedence:
nightly > stable-with-version > stable
With stable-with-version, we invoke the RA with a `--version` arg and attempt to extract a date. Given the same date as a nightly, the nightly RA will win.
This change updates the log format to use the vscode log format instead
of the custom log format, by replacing the `OutputChannel` with a
`LogOutputChannel` and using the `debug`, `info`, `warn`, and `error`
methods on it. This has the following benefits:
- Each log level now has its own color and the timestamp is in a more
standard format
- Inspect output (e.g. the log of the config object) is now colored
- Error stack traces are now shown in the output
- The log level is now controlled on the output tab by clicking the gear
icon and selecting "Debug" or by passing the `--log` parameter to
vscode. The `trace.extension` setting has been marked as deprecated.
And that is due to a case where we have two ratomls in a source root, one of which
is a `workspace_ratoml` and the other one is simple old ratoml. Since we are not checking to see if
the source root is already populated with workspace ratoml, this test fails. Due to principles of clear
code I believe it is reasonable to not have two HashMaps that are almost for the exact same thing.
So next commit should remove `workspace_ratoml` and merge it with `krate_ratoml`s.
#17497 - Invalid RA diagnostic error: expected 2 arguments, found 1
Fix for #17497
The issue occurs because in some configurations of traits where one of them has `Deref` as a supertrait, RA's type inference algorithm fails to resolve the `Deref::Target` type, and instead uses a `TyKind::BoundVar` (i.e. an unknown type). This "autoderefed" type then incorrectly acts as if it implements all traits in scope.
The fix is to re-apply the same sanity-check that is done in [`iterate_method_candidates_with_autoref()`](9463d9eea4/crates/hir-ty/src/method_resolution.rs (L1008)), that is: don't try to resolve methods on unknown types. This same sanity-check is now done on each autoderefed type for which trait methods are about to be checked. If the autoderefed type is unknown, then the iterating of the trait methods for that type is skipped.
Includes a unit test that only passes after applying the fixes in this commit.
Includes a change to the assertion count in test `syntax_highlighting::tests::benchmark_syntax_highlighting_parser` as suggested by Lukas Wirth during review.
Includes a change to the sanity-check code as suggested by Florian Diebold during review.
The issue occurs because in some configurations of traits where one of them has Deref as a supertrait, RA's type inference algorithm fails to resolve the Deref::Target type, and instead uses a TyKind::BoundVar (i.e. an unknown type). This "autoderefed" type then incorrectly acts as if it implements all traits in scope.
The fix is to re-apply the same sanity-check that is done in iterate_method_candidates_with_autoref(), that is: don't try to resolve methods on unknown types. This same sanity-check is now done on each autoderefed type for which trait methods are about to be checked. If the autoderefed type is unknown, then the iterating of the trait methods for that type is skipped.
Includes a unit test that only passes after applying the fixes in this commit.
Includes a change to the assertion count in test syntax_highlighting::tests::benchmark_syntax_highlighting_parser as suggested by Lukas Wirth during review.
Includes a change to the sanity-check code as suggested by Florian Diebold during review.
