2629: Remove imports from hir r=matklad a=matklad
We only used them to avoid self-confirming completions (`use self::foo`), but that can be handled more locally.
bors r+
Co-authored-by: Aleksey Kladov <aleksey.kladov@gmail.com>
2628: Add macro 2.0 support in parser r=matklad a=edwin0cheng
This PR added a new syntax kind : `MACRO_DEF` and a keyword `MACRO_KW`
there are two syntax for declarative macro 2.0 :
1. Normal : `macro m { ($i:ident) => {} }` , which handle similar to legacy one.
2. Call like: `macro m($i:ident) {}`, it produces a single token tree which have two child token trees : `($i:ident)` and `{}`
Co-authored-by: Edwin Cheng <edwin0cheng@gmail.com>
2623: Add support macros in impl blocks r=matklad a=edwin0cheng
This PR add support for macros in impl blocks, which reuse `Expander` for macro expansion.
see also: #2459
Co-authored-by: Edwin Cheng <edwin0cheng@gmail.com>
2615: Fix wrong path parsing for macro call in pattern position r=edwin0cheng a=edwin0cheng
The parser incorrectly insert a `PathPat` inside `MacroCall` syntax node when parsing inside a pattern position, for example :
```rust
let foo!() = 0;
```
become:
```
MACRO_CALL@[60; 66)
PATH_PAT@[60; 63) <------------- It should not exist
PATH@[60; 63)
PATH_SEGMENT@[60; 63)
NAME_REF@[60; 63)
IDENT@[60; 63) "foo"
EXCL@[63; 64) "!"
TOKEN_TREE@[64; 66)
L_PAREN@[64; 65) "("
R_PAREN@[65; 66) ")"
```
This PR fix this bug and add some test to make sure goto-defintion works for macro inside pattern.
Co-authored-by: Edwin Cheng <edwin0cheng@gmail.com>
2614: Clippy cleanup r=matklad a=kjeremy
Just a few tweaks from the latest clippy. There are a lot more but we should probably tweak our settings.
Co-authored-by: kjeremy <kjeremy@gmail.com>
2592: Add std::ops::Index support for infering r=edwin0cheng a=edwin0cheng
see also #2534
Seem like this can't fix#2534 for this case:
```rust
fn foo3(bar: [usize; 2]) {
let baz = bar[1]; // <--- baz is still unknown ?
println!("{}", baz);
}
```
Co-authored-by: Edwin Cheng <edwin0cheng@gmail.com>
