Somehow r-a believed that my sysroot was something weird with no rustc.
Probably a me issue, but it was impossible to diagnose since r-a just
gave me a plain "No such file or directory". Adding this error makes it
clear what happened and allows diagnosing the problem.
feat: resolve inherent and implemented associated items in docs
This partially fixes#9694.
Supported:
- Trait methods and constants.
* Due to resolution differences pointed out during the review of the PR, trait associated types are _not_ supported.
- Inherent methods, constants and associated types.
* Inherent associated types are a [nightly feature](https://github.com/rust-lang/rust/issues/8995), and are supported with no additional work in this PR.
Screenshot of VS Code running with the change:
<img width="513" alt="image" src="https://github.com/rust-lang/rust-analyzer/assets/7189784/c37ed8b7-b572-4684-8e81-2a817b0027c4">
You can see that the items are resolved (excl. trait associated types) since they are semantically highlighted in the doc comment.
fix: try obligation of `IndexMut` when infer
Closes#15842.
This issue arises because `K` is ambiguous if only inferred from `Index` trait, but is unique if inferred from `IndexMut`, but r-a doesn't use this info.
SymbolInformation::kind is finer-grained than the SCIP symbol suffix.
This also fixes a bug where all type aliases where treated like type
parameters.
```
trait SomeTrait {
type AssociatedType; // ← this is SomeTrait#[AssociatedType]
}
type MyTypeAlias = u8; // ← this used to be [MyTypeAlias]
// and now is MyTypeAlias#
```
To build the SymbolInformation::signature_documentation we need access
to the “label” when building the TokenStaticData, preferably without
any markdown markup.
Therefore this refactors ide::hover::render::definition and its helper
functions to give easier access to the label alone.
For local variables, this gets the moniker from the enclosing
definition and stores it into the TokenStaticData.
Then it builds the scip symbol for that moniker when building the
SymbolInformation.
This is meant to implement SymbolInformation::enclosing_symbol, so we
can build the enclosing symbol from the enclosing moniker without
having the full enclosing token's TokenStaticData.
fix: pick up new names when the name conflicts in 'introduce_named_generic'
Improve generation of names for generic parameters in `introduce_named_generics`.
fix#15731.
### Changes
- Modified `for_generic_parameter` function in `suggest_name.rs` to handle conflicts with existing generic parameters and generate unique names accordingly.
- Update `introduce_named_generic` function and pass existing params to `for_generic_parameter`, enabling the detection and handling of name collisions.
* Extracted the function `for_unique_generic_name` that handling generics with identical names for reusability.
* Renamed `for_generic_params` to `for_impl_trait_as_generic` for clarity
* Added documentations for `for_impl_trait_as_generic` and `for_unique_generic_name`
This commit changes how the expected type is calculated when working
with Fn pointers, making the parenthesis stop vanishing when completing
the function name.
I've been bugged by the behaviour on parenthesis completion for a long
while now. R-a assumes that the `LetStmt` type is the same as the
function type I've just written. Worse is that all parenthesis vanish,
even from functions that have completely different signatures. It will
now verify if the signature is the same.
While working on this, I noticed that record fields behave the same, so
I also made it prioritize the field type instead of the current
expression when possible, but I'm unsure if this is OK, so input is
appreciated.
ImplTraits as return types will still behave weirdly because lowering is
disallowed at the time it resolves the function types.
fix: rewrite code_action `generate_delegate_trait`
I've made substantial enhancements to the "generate delegate trait" code action in rust-analyzer. Here's a summary of the changes:
#### Resolved the "Can’t find CONST_ARG@158..159 in AstIdMap" error
Fix#15804, fix#15968, fix#15108
The issue stemmed from an incorrect application of PathTransform in the original code. Previously, a new 'impl' was generated first and then transformed, causing PathTransform to fail in locating the correct AST node, resulting in an error. I rectified this by performing the transformation before generating the new 'impl' (using make::impl_trait), ensuring a step-by-step transformation of associated items.
#### Rectified generation of `Self` type
`generate_delegate_trait` is unable to properly handle trait with `Self` type.
Let's take the following code as an example:
```rust
trait Trait {
fn f() -> Self;
}
struct B {}
impl Trait for B {
fn f() -> B { B{} }
}
struct S {
b: B,
}
```
Here, if we implement `Trait` for `S`, the type of `f` should be `() -> Self`, i.e. `() -> S`. However we cannot automatically generate a function that constructs `S`.
To ensure that the code action doesn't generate delegate traits for traits with Self types, I add a function named `has_self_type` to handle it.
#### Extended support for generics in structs and fields within this code action
The former version of `generate_delegate_trait` cannot handle structs with generics properly. Here's an example:
```rust
struct B<T> {
a: T
}
trait Trait<T> {
fn f(a: T);
}
impl<T1, T2> Trait<T1> for B<T2> {
fn f(a: T1) -> T2 { self.a }
}
struct A {}
struct S {
b$0 : B<A>,
}
```
The former version will generates improper code:
```rust
impl<T1, T2> Trait<T1, T2> for S {
fn f(&self, a: T1) -> T1 {
<B as Trait<T1, T2>>::f( &self.b , a)
}
}
```
The rewritten version can handle generics properly:
```rust
impl<T1> Trait<T1> for S {
fn f(&self, a: T1) -> T1 {
<B<A> as Trait<T1>>::f(&self.b, a)
}
}
```
See more examples in added unit tests.
I enabled support for generic structs in `generate_delegate_trait` through the following steps (using the code example provided):
1. Initially, to prevent conflicts between the generic parameters in struct `S` and the ones in the impl of `B`, I renamed the generic parameters of `S`.
2. Then, since `B`'s parameters are instantiated within `S`, the original generic parameters of `B` needed removal within `S` (to avoid errors from redundant parameters). An important consideration here arises when Trait and B share parameters in `B`'s impl. In such cases, these shared generic parameters cannot be removed.
3. Next, I addressed the matching of types between `B`'s type in `S` and its type in the impl. Given that some generic parameters in the impl are instantiated in `B`, I replaced these parameters with their instantiated results using PathTransform. For instance, in the example provided, matching `B<A>` and `B<T2>`, where `T2` is instantiated as `A`, I replaced all occurrences of `T2` in the impl with `A` (i.e. apply the instantiated generic arguments to the params).
4. Finally, I performed transformations on each assoc item (also to prevent the initial issue) and handled redundant where clauses.
For a more detailed explanation, please refer to the code and comments. I welcome suggestions and any further questions!
fix: self type replacement in inline-function
Fix#16113, fix#16091
The problem described in this issue actually involves three bugs.
Firstly, when using `ted` to modify the syntax tree, the offset of nodes on the tree changes, which causes the syntax range information from `hir` to become invalid. Therefore, we need to edit the AST after the last usage for `usages_for_locals`.
The second issue is that when inserting nodes, it's necessary to use `clone_subtree` for duplication because the `ted::replace` operation essentially moves a node.
The third issue is that we should use `ancestors_with_macros` instead of `ancestors` to handle impl definition in macros.
I have fixed the three bugs mentioned above and added unit tests.
internal: Migrate assists to the structured snippet API, part 5
Continuing from #15874
Migrates the following assists:
- `extract_variable`
- `generate_function`
- `replace_is_some_with_if_let_some`
- `replace_is_ok_with_if_let_ok`
Don't trim trailing whitespace from doc comments
Don't trim trailing whitespace from doc comments as multiple trailing spaces indicates a hard line break in Markdown.
I'd have liked to add a unit test for `docs_from_attrs`, but couldn't find a reasonable way to get an `&Attrs` object for use in the test.
Fixes#15877.
fix: make callable fields not complete in method access no parens case
Follow up PR for #15879
Fixes the callable field completion appearing in the method access with no parens case.
fix: no code action 'introduce_named_generic' for impl inside types
Fix#15734.
### Changes Made
- Find params in `ancestors` instead of just `parent`
- Added tests (`replace_impl_with_mut` and `replace_impl_inside`)
fix: Correct references from `rust-analyzer.cargo.check` to `rust-analyzer.check`
When reading the manual, I noticed that the documentation referenced configurations that have since been renamed. This PR updates those references to their new names.
While reading through the code base, I stumbled across a piece of code that I found hard to read despite its simple purpose. This is my attempt at making the code easier to understand for future readers.
I won't be offended if this is too minor and not worth your time.
internal: Update world symbols request definiton, prefer focus range for macros
Prior to this, the symbol search would always jump to the defining macro call, not it jumps to the name in the macro call input if possible. This is a large improvement for assoc items in an attribute impl or trait.
Complete exported macros in `#[macro_use($0)]`
Closes https://github.com/rust-lang/rust-analyzer/issues/15657.
Originally added a test case for incomplete input:
```rust
#[test]
fn completes_incomplete_syntax() {
check(
r#"
//- /dep.rs crate:dep
#[macro_export]
macro_rules! foo {
() => {};
}
//- /main.rs crate:main deps:dep
#[macro_use($0
extern crate dep;
"#,
expect![[r#"
ma foo
"#]],
)
}
```
but couldn't make it pass and removed it 😅 Our current recovering logic doesn't work for token trees and for this code:
```rust
#[macro_use(
extern crate lazy_static;
fn main() {}
```
we ended up with this syntax tree:
```
SOURCE_FILE@0..53
ATTR@0..52
POUND@0..1 "#"
L_BRACK@1..2 "["
META@2..52
PATH@2..11
PATH_SEGMENT@2..11
NAME_REF@2..11
IDENT@2..11 "macro_use"
TOKEN_TREE@11..52
L_PAREN@11..12 "("
WHITESPACE@12..13 "\n"
EXTERN_KW@13..19 "extern"
WHITESPACE@19..20 " "
CRATE_KW@20..25 "crate"
WHITESPACE@25..26 " "
IDENT@26..37 "lazy_static"
SEMICOLON@37..38 ";"
WHITESPACE@38..40 "\n\n"
FN_KW@40..42 "fn"
WHITESPACE@42..43 " "
IDENT@43..47 "main"
TOKEN_TREE@47..49
L_PAREN@47..48 "("
R_PAREN@48..49 ")"
WHITESPACE@49..50 " "
TOKEN_TREE@50..52
L_CURLY@50..51 "{"
R_CURLY@51..52 "}"
WHITESPACE@52..53 "\n"
```
Maybe we can try to parse the token tree in `crates/ide-completion/src/context/analysis.rs` but I'm not sure what's the best way forward.
fix: Correctly set and mark the proc-macro spans
This slows down analysis by 2-3s on self for me unfortunately (~2.5% slowdown)
Noisy diff due to two simple refactoring in the first 2 commits. Relevant changes are [7d762d1](7d762d18ed) and [1e1113c](1e1113cf5f) which introduce def site spans and correct marking for proc-macros respectively.
fix: Update metavariable expression implementation
Fixes https://github.com/rust-lang/rust-analyzer/issues/16154
This duplicates behavior of that before and after PR https://github.com/rust-lang/rust/pull/117050 based on the toolchain version. There are some 1.76 nightlies that are still broken (any before that PR basically) but fetching and storing the commit makes little sense to me (opposed to the toolchain version).
minor: Use reserve when removing markdown from text
After markdown syntax removal the length of the text is roughly the same so we can reserve memory beforehand
fix(mbe): desugar doc correctly for mbe
Fixes#16110.
The way rust desugars doc comments when expanding macros is rendering it as raw strings delimited with hashes. Rust-analyzer wasn't aware of this, so the desugared doc comments wouldn't match correctly when on the LHS of macro declarations.
This PR fixes this by porting the code used by rustc:
59096cdad0/compiler/rustc_ast/src/tokenstream.rs (L662-L671)
Fixes#16110.
The way rust desugars doc comments when expanding macros
is rendering it as raw strings delimited with hashes.
Rust-analyzer wasn't aware of this, so the desugared doc
comments wouldn't match correctly when on the LHS of macro
declarations.
This PR fixes this by porting the code used by rustc: 4cfdbd328b/compiler/rustc_ast/src/tokenstream.rs (L6837)
internal: Move proc-macro knowledge out of base-db into hir-expand
It does not make much sense to me to have that live in base-db, additionally, it kind of conflicts with moving span things out into a separate crate
Fix incorrectly replacing references in macro invocation in "Convert to named struct" assist
Fixes#15630.
Complements #13647 (same assist but missed this one), #14920 (inverse action assist).
Let `reuse` look inside git submodules
Changes `collect-license-metadata` and `generate-copyright` so they can now look at the git submodules.
Unfortunately `reuse` chokes on the LLVM submodule - it finds the word "Copyright" or the unicode copyright symbol in all kinds of places, including UTF-8 test cases. The `reuse` tool expressly won't let you ignore folders, so we let it scan everything and then strip out the LLVM sub-folder in post. Instead, we add in a hand-curated list of copyright information gleaned by reading the LLVM codebase carefully, which is stored in `.reuse/dep5` in Debian format where `reuse` can find and use it.
The `.reuse/dep5` continues to track copyright info for files in the tree that do not have SPDX metadata in them (i.e. all of them)
Use a u64 for the rmeta root position
Waffle noticed this in https://github.com/rust-lang/rust/pull/117301#discussion_r1405410174
We've upgraded the other file offsets to u64, and this one only costs 4 bytes per file. Also the way the truncation was being done before was extremely easy to miss, I sure missed it! It's not clear to me if not having this change effectively made the other upgrades from u32 to u64 ineffective, but we can have it now.
r? `@WaffleLapkin`
fix: Don't emit "missing items" diagnostic for negative impls
Negative impls can't have items, so there is no reason for this diagnostic.
LMK if I should add a test somewhere. Also LMK if that's not how we usually check multiple things in an if in r-a.
fix: Fix view mir, hir and eval function not working when cursor is inside macros
I broke the view ones completely by inverting the macro check by accident a few days ago but we don't talk about that.
detects redundant imports that can be eliminated.
for #117772 :
In order to facilitate review and modification, split the checking code and
removing redundant imports code into two PR.
feat: Prioritize import suggestions based on the expected type
Hi, this is a draft PR to solve #15384. `Adt` types work and now I have a few questions :)
1. What other types make sense in this context? Looking at [ModuleDef](05666441ba/crates/hir/src/lib.rs (L275)) I am thinking everything except Modules.
2. Is there an existing way of converting between `ModeuleDef` and `hir::Type` in the rustanalyzer code base?
3. Does this approach seem sound to you?
Ups: Upon writing this I just realised that the enum test is invalided as there are no enum variants and this no variant is passed as a function argument.
fix: resolve Self type references in delegate method assist
This PR makes the delegate method assist resolve any `Self` type references in the parameters or return type. It also works across macros such as the `uint_impl!` macro used for `saturating_mul` in the issue example.
Closes#14485
fix: Fix item tree lowering pub(self) to pub()
Prior to this, the item tree lowered `pub(self)` visibility to `pub()`
Fix#15134 - tested with a unit test and
a manual end-to-end test of building rust-analyzer from my branch and opening the reproduction repository
Before
Private functions have RawVisibility module, but were
missed because take_types returned None early. After resolve_visibility
returned None, Visibility::Public was set instead and private functions
ended up being offered in autocompletion.
Choosing such a function results in an immediate error diagnostic
about using a private function.
After
Pattern match of take_types that returns None and
query for Module-level visibility from the original_module
Fix#15134 - tested with a unit test and a manual end-to-end
test of building rust-analyzer from my branch and opening
the reproduction repository
REVIEW
Refactor to move scope_def_applicable and check function visibility
from a module
Please let me know what's the best way to add a unit tests to
nameres, which is where the root cause was
Fix panic with closure inside array len
I was working on #15947 and found out that we panic on this test:
```
fn main() {
let x = [(); &(&'static: loop { |x| {}; }) as *const _ as usize]
}
```
This PR fixes the panic. Closures in array len are still broken, but closure in const eval is not stable anyway.
feat: Allow navigation targets to be duplicated when the focus range lies in the macro definition site
Basically if a name of an item originates from the macro definition we now point to that as well as the creating macro call.
Big diff because I also made `FileId`s field private due to some debugging I had to do (having a searchable constructor makes things easier).
Add support for making lib features internal
We have the notion of an "internal" lang feature: a feature that is never intended to be stabilized, and using which can cause ICEs and other issues without that being considered a bug.
This extends that idea to lib features as well. It is an alternative to https://github.com/rust-lang/rust/pull/115623: instead of using an attribute to declare lib features internal, we simply do this based on the name. Everything ending in `_internals` or `_internal` is considered internal.
Then we rename `core_intrinsics` to `core_intrinsics_internal`, which fixes https://github.com/rust-lang/rust/issues/115597.
fix: Insert fn call parens only if the parens inserted around field name
Fixes#16014.
Sorry I missed it in previous PR. I've added a test as level to prevent regressions again.
Give any suggestions to improve the test if anything.
TokenMap -> SpanMap rewrite
Opening early so I can have an overview over the full diff more easily, still very unfinished and lots of work to be done.
The gist of what this PR does is move away from assigning IDs to tokens in arguments and expansions and instead gives the subtrees the text ranges they are sourced from (made relative to some item for incrementality). This means we now only have a single map per expension, opposed to map for expansion and arguments.
A few of the things that are not done yet (in arbitrary order):
- [x] generally clean up the current mess
- [x] proc-macros, have been completely ignored so far
- [x] syntax fixups, has been commented out for the time being needs to be rewritten on top of some marker SyntaxContextId
- [x] macro invocation syntax contexts are not properly passed around yet, so $crate hygiene does not work in all cases (but most)
- [x] builtin macros do not set spans properly, $crate basically does not work with them rn (which we use)
~~- [ ] remove all uses of dummy spans (or if that does not work, change the dummy entries for dummy spans so that tests will not silently pass due to havin a file id for the dummy file)~~
- [x] de-queryfy `macro_expand`, the sole caller of it is `parse_macro_expansion`, and both of these are lru-cached with the same limit so having it be a query is pointless
- [x] docs and more docs
- [x] fix eager macro spans and other stuff
- [x] simplify include! handling
- [x] Figure out how to undo the sudden `()` expression wrapping in expansions / alternatively prioritize getting invisible delimiters working again
- [x] Simplify InFile stuff and HirFIleId extensions
~~- [ ] span crate containing all the file ids, span stuff, ast ids. Then remove the dependency injection generics from tt and mbe~~
Fixes https://github.com/rust-lang/rust-analyzer/issues/10300
Fixes https://github.com/rust-lang/rust-analyzer/issues/15685
Implement completion for the callable fields.
Fixes#14656
PR is opened with basic changes. It could be improved by having a new `SymbolKind` for the callable fields and implementing a separate render function similar to the `render_method` for the new `SymbolKind`.
It could also be done without any changes to the `SymbolKind` of course, have the new function called based on the type of field.
I prefer the former method.
Please give any thoughts or changes you think is appropriate for this method. I could start working on that in this same PR.
chore: remove unused `PhantomData`
This PR removes an unused `PhantomData` in `FileItemTreeId`.
*Note:* I am not sure how this should be implemented, maybe as a type instead of a wrapper struct? I'd be happy to do so if needed 👍
This commit addresses the issue of excessive and unrelated errors
generated by top-level `let` statements. Now, only a single error is
produced, indicating that `let` statements are invalid at the top level.
