bors[bot] [Wed, 6 Jan 2021 16:41:22 +0000 (16:41 +0000)]
Merge #7181
7181: Document project_model::PackageData and project_model::TargetData r=arnaudgolfouse a=arnaudgolfouse
This PR adds some documentation for the `project_model` crate.
Some of the field descriptions were taken directly from their `cargo_metadata` counterpart :
- `PackageData` -> `cargo_metadata::Package`
- `TargetData` -> `cargo_metadata::Target`
bors[bot] [Tue, 5 Jan 2021 14:46:47 +0000 (14:46 +0000)]
Merge #7171
7171: Emit diagnostics for unresolved item-level macros r=jonas-schievink a=jonas-schievink
We have been emitting macro *expansion* errors for macro calls in item position, as well as expansion and resolution errors for macro calls in bodies. This fills in the last gap, which is *resolution* errors for macro calls in item position (ie. those that name resolution cares about).
bors r+
Co-authored-by: Jonas Schievink <jonasschievink@gmail.com>
bors[bot] [Tue, 5 Jan 2021 12:04:35 +0000 (12:04 +0000)]
Merge #7140
7140: Store trait associated items in fst r=matklad a=SomeoneToIgnore
Store imported traits' associated function/methods and constants into `ImportMap.fst` and pefrorm the imports search on them.
This is a first step towards trait autoimport during completion functionality, the way I see it, after this PR, only a few major things are left to be done:
* store all traits' assoc items into fst, not only the ones in scope, as we do now. Any code pointers on how to do this are welcome 😄
* adjust a few modules in completions crate (`dot.rs`, `qualified_path.rs` at least) to query the import map, reusing the `import_assets` logic heavily
==
With the current import and autoimport implementations, it looks like for a single query, we're either interested in either associated items lookup or in all other `fst` contents lookup, but never both simultaneously.
I would rather not split `fst` in two but add another `Query` parameter to separate those, but let me know if you have any ideas.
bors[bot] [Tue, 5 Jan 2021 10:53:24 +0000 (10:53 +0000)]
Merge #7162
7162: Introduce queries to avoid problems when performing completion for enums with many variants r=matklad a=danielframpton
This change introduces two new queries to compute:
1) attributes for all variants of an enum, and
2) attributes for all fields of a variant.
The purpose of this change is to avoid the current n^2 behavior when rendering completion for variants (which prevents completion for enums with large numbers of variants).
Co-authored-by: Daniel Frampton <Daniel.Frampton@microsoft.com>
bors[bot] [Mon, 4 Jan 2021 20:36:13 +0000 (20:36 +0000)]
Merge #7147
7147: ssr: Allow replacing expressions with statements r=davidlattimore a=MarijnS95
Depends on #6587
Until that is merged, the diff is https://github.com/MarijnS95/rust-analyzer/compare/stmt..replace-expr-with-stmt
---
Now that statements can be matched and replaced (#6587) some usecases require expressions to be replaced with statements as well. This happens when something that can ambiguously be an expression or statement like `if` and loop blocks appear in the last position of a block, as trailing expression. In this case a replacement pattern of the form `if foo(){$a();}==>>$a();` will only substitute `if` blocks in the list of statements but not if they (implicitly) end up in the trailing expression, where they are not wrapped by an EXPR_STMT (but the pattern and template are, as parsing only succeeds for the `stmt ==>> stmt` case).
Instead of adding two rules that match an expression - and emit duplicate matching errors - allow the template for expressions to be a statement if it fails to parse as an expression.
---
Another gross change that does not seem to break any tests currently, but perhaps a safeguard should be added to only allow this kind of replacement in blocks by "pushing" the replacement template to the statement list and clearing the trailing expression?
Arnaud [Mon, 4 Jan 2021 17:58:42 +0000 (18:58 +0100)]
Make it possible to retrieve `hir::Function`'s return type
This is done by adding a `ret_type` method to `hir::Function`.
I followed `assoc_fn_params` convention by creating a new `RetType` type,
that contains the actual return type accessible via a `ty` method.
bors[bot] [Mon, 4 Jan 2021 15:23:53 +0000 (15:23 +0000)]
Merge #7154
7154: Show goto type actions for Const and TypeParams r=matklad a=Veykril
Shows goto type actions for type parameters:
![Code_6hn3rowu9M](https://user-images.githubusercontent.com/3757771/103547890-42aaeb00-4ea5-11eb-8ac7-f166869af5f8.png)
Shows goto type actions for const parameters:
![Code_8UFCcbZL3z](https://user-images.githubusercontent.com/3757771/103547891-43438180-4ea5-11eb-91e8-50681e4d831e.png)
Also shows implementations for `Self`:
![Code_eQj1pWfser](https://user-images.githubusercontent.com/3757771/103547892-43438180-4ea5-11eb-9122-461f2e0fdd01.png)
bors[bot] [Mon, 4 Jan 2021 11:14:40 +0000 (11:14 +0000)]
Merge #6587
6587: SSR: Support statement matching and replacing r=davidlattimore a=MarijnS95
For #3186
Hi!
This is a smaller initial patchset that came up while working on support for statement lists (and my first time working on RA :grin:). It has me stuck on trailing semicolons for which I hope to receive some feedback. Matching (and replacing) `let` bindings with a trailing semicolon works fine, but trying to omit these (to make patterns more ergonomic) turns out more complex than expected.
The "optional trailing semicolon solution" implemented in this PR is ugly because `Matcher::attempt_match_token` should only consume a trailing `;` when parsing `let` bindings to prevent other code from breaking. That at the same time has a nasty side-effect of `;` ending up in the matched code: any replacements on that should include the trailing semicolon as well even if it was not in the pattern. A better example is in the tests:
The end result to achieve is (I guess) allowing replacement of let bindings without trailing semicolon like `let x = $a ==>> let x = 1` (but including them on both sides is still fine), and should make replacement in a macro call (where `foo!(let a = 2;)` for a `$x:stmt` is invalid syntax) possible as well. That should allow to enable/fix these tests:
Marijn Suijten [Sun, 3 Jan 2021 16:45:49 +0000 (17:45 +0100)]
ssr: Allow replacing expressions with statements
Now that statements can be matched and replaced (#6587) some usecases
require expressions to be replaced with statements as well. This happens
when something that can ambiguously be an expression or statement like
`if` and loop blocks appear in the last position of a block, as trailing
expression. In this case a replacement pattern of the form `if
foo(){$a();}==>>$a();` will only substitute `if` blocks in the list of
statements but not if they (implicitly) end up in the trailing
expression, where they are not wrapped by an EXPR_STMT (but the pattern
and template are, as parsing only succeeds for the `stmt ==>> stmt`
case).
Instead of adding two rules that match an expression - and emit
duplicate matching errors - allow the template for expressions to be a
statement if it fails to parse as an expression.
bors[bot] [Sun, 3 Jan 2021 18:22:07 +0000 (18:22 +0000)]
Merge #7113
7113: Manual updates r=matklad a=tekul
Add some details on how to build the manual and some clarification on how to deal with "proc macro2 warnings.
For context, this arose from [this question](https://users.rust-lang.org/t/how-to-disable-rust-analyzer-proc-macro-warnings-in-neovim/53150) on users.rust-lang.org.