Commit Graph

536 Commits

Author SHA1 Message Date
Young-Flash
3605bb38ff fix CI: generate doctest 2023-10-12 11:50:44 +08:00
Young-Flash
a7fada4650 add replace_is_some_with_if_let_some assist 2023-10-11 17:44:27 +08:00
bors
ab62c0186f Auto merge of #15696 - rmehri01:14293_tuple_return_type_to_struct, r=Veykril
feat: implement tuple return type to tuple struct assist

This PR implements the `convert_tuple_return_type_to_struct` assist, for converting the return type of a function or method from a tuple to a tuple struct. Additionally, it moves the `to_camel_case` and `char_has_case` functions from `case_conv` to `stdx` so that they can be used similar to `to_lower_snake_case`.

[tuple_return_type_to_tuple_struct.webm](https://github.com/rust-lang/rust-analyzer/assets/52933714/2803ff58-fde3-4144-9495-7c7c7e139075)

Currently, the assist puts the struct definition above the function, or above the nearest `impl` or `trait` if applicable and only rewrites literal tuples that are returned in the body of the function. Additionally, it only attempts to rewrite simple tuple pattern usages with the corresponding tuple struct pattern but does so across files and modules.

I think that this is sufficient for the majority of use cases but I could be wrong. One thing I'm still not sure how to approach is handling `Self` and generics/lifetimes in the tuple type to be extracted. I was thinking of either manually figuring out what lifetimes and generics are in scope and using them (sort of similar to the `generate_function` assist) or maybe using `ctx.sema.resolve_type` and `generic_params` on `hir::Type` but this seems to not deal with lifetimes.

Closes #14293
2023-10-09 08:14:42 +00:00
bors
695c612489 Auto merge of #15641 - alibektas:15598/fix_into_to_from, r=Veykril
fix: preceding QualifiedPathType for into_to_from assist

fixes #15598
2023-10-05 08:25:58 +00:00
Ryan Mehri
9ba8dbc902 style: clean up magic number for finding pattern usages 2023-10-04 08:04:59 -07:00
bors
7e9da40078 Auto merge of #15700 - rmehri01:15694_iterator_demorgan, r=Veykril
feat: add assist for applying De Morgan's law to `Iterator::all` and `Iterator::any`

This PR adds an assist for transforming expressions of the form `!iter.any(|x| predicate(x))` into `iter.all(|x| !predicate(x))` and vice versa.

[IteratorDeMorgans.webm](https://github.com/rust-lang/rust-analyzer/assets/52933714/aad1a299-6620-432b-9106-aafd2a7fa9f5)

Closes #15694
2023-10-04 11:08:44 +00:00
Lukas Wirth
c266387e13
Replace unwrap with expect 2023-10-04 13:06:23 +02:00
Ryan Mehri
34d3490198 feat: add assist for applying De Morgan's law to iterators 2023-10-01 21:30:10 -07:00
Ryan Mehri
146a7cc490 fix: allow more kinds of if let patterns in guarded return assist 2023-10-01 11:48:10 -07:00
Ryan Mehri
2611fbf623 implement basic version of convert_tuple_return_type_to_struct assist 2023-09-30 17:07:01 -07:00
bors
87e2c310f9 Auto merge of #15667 - rmehri01:bool_to_enum_top_level, r=Veykril
fix: make bool_to_enum assist create enum at top-level

This pr makes the `bool_to_enum` assist create the `enum` at the next closest module block or at top-level, which fixes a few tricky cases such as with an associated `const` in a trait or module:

```rust
trait Foo {
    const $0BOOL: bool;
}

impl Foo for usize {
    const BOOL: bool = true;
}

fn main() {
    if <usize as Foo>::BOOL {
        println!("foo");
    }
}
```

Which now properly produces:

```rust
#[derive(PartialEq, Eq)]
enum Bool { True, False }

trait Foo {
    const BOOL: Bool;
}

impl Foo for usize {
    const BOOL: Bool = Bool::True;
}

fn main() {
    if <usize as Foo>::BOOL == Bool::True {
        println!("foo");
    }
}
```

I also think it's a bit nicer, especially for local variables, but didn't really know to do it in the first PR :)
2023-09-29 10:20:11 +00:00
Ryan Mehri
1b3e5b2105 style: simplify node_to_insert_before 2023-09-28 10:09:13 -07:00
bors
3b1b58c225 Auto merge of #15662 - rmehri01:fix_panic_with_return_in_match, r=Veykril
fix: panic with wrapping/unwrapping result return type assists

With the `wrap_return_type_in_result` assist, the following code results in a panic (note the lack of a semicolon):

```rust
fn foo(num: i32) -> $0i32 {
    return num
}

=>

thread 'handlers::wrap_return_type_in_result::tests::wrap_return_in_tail_position' panicked at crates/syntax/src/ted.rs:137:41:
called `Option::unwrap()` on a `None` value
```

I think this is because it first walks the body expression to change any `return` expressions and then walks all tail expressions, resulting in the `return num` being changed twice since it is both a `return` and in tail position. This can also happen when a `match` is in tail position and `return` is used in a branch for example. Not really sure how big of an issue this is in practice though since this seems to be the only case that is impacted and can be reduced to just `num` instead of `return num`.

This also occurs with the `unwrap_result_return_type` assist but panics with the following instead:

```
thread 'handlers::unwrap_result_return_type::tests::wrap_return_in_tail_position' panicked at /rustc/3223b0b5e8dadda3f76c3fd1a8d6c5addc09599e/library/alloc/src/string.rs:1766:29:
assertion failed: self.is_char_boundary(n)
```
2023-09-26 14:18:33 +00:00
Ryan Mehri
73150c3f36 fix: wrap method call exprs in parens 2023-09-25 21:44:16 -07:00
Ryan Mehri
bce4be9478 fix: make bool_to_enum assist create enum at top-level 2023-09-25 21:01:54 -07:00
Milo
85ead6ec27 remove other unwraps 2023-09-25 11:48:23 +00:00
Milo
f64eecd2e2 fix one 2023-09-25 11:30:21 +00:00
Ryan Mehri
7306504b82 fix panic with wrapping/unwrapping result return type assists 2023-09-24 16:00:55 -07:00
Ali Bektas
fc258de5a3 Make QualPathTy case readable 2023-09-22 21:23:03 +02:00
bors
8139e8e072 Auto merge of #15425 - alibektas:deunwrap/convert_comment_block, r=Veykril
minor : Deunwrap convert_comment_block and desugar_doc_comment

Closes subtask 13 of #15398 . I still don't know a more idiomatic way for the for loops I added, any suggestion would make me happy.
2023-09-22 15:47:55 +00:00
bors
59bcbafc95 Auto merge of #15594 - alibektas:deunwrap/add_missing_match_arms, r=Veykril
Deunwrap add_missing_match_arms

Last subtask of #15398
2023-09-22 15:31:30 +00:00
Ali Bektas
132a6ce8fc Omit QualPathTy when possible 2023-09-22 14:04:17 +02:00
Ali Bektas
622e1a8d88 Add a test case to add_missing_match_arms
Although it doesn't panic now, further changes to how we recover from incomplete syntax
may cause this assist to panic. To mitigate this a test case has been added.
2023-09-22 13:51:19 +02:00
Ali Bektas
0a91a54794 v4 2023-09-22 13:32:20 +02:00
Ali Bektas
695a1349fa Fix doctest 2023-09-22 10:46:21 +02:00
Ali Bektas
8ad536f2d1 Make path start with a QualifiedPathType 2023-09-22 10:46:21 +02:00
bors
df75809a85 Auto merge of #15484 - rmehri01:14779_bool_to_enum_assist, r=Veykril
feat: Bool to enum assist

This adds the `bool_to_enum` assist, which converts the type of boolean local variables, fields, constants and statics to a new `enum` type, making it easier to distinguish the meaning of `true` and `false` by renaming the variants.

Closes #14779
2023-09-22 07:19:12 +00:00
bors
2ededa2f14 Auto merge of #15432 - alibektas:deunwrap/inline_call, r=Veykril
minor : Deunwrap inline call

#15398 subtask 4. There is still one instance of unwrap, which I found pretty hard to change.
2023-09-22 07:03:02 +00:00
Lukas Wirth
93562dd5bd Use parent + and_then instead of ancestors 2023-09-22 08:53:24 +02:00
Ryan Mehri
ea11846490 fix parens when inlining closure in body of function 2023-09-21 21:55:10 -07:00
Ryan Mehri
60f7473c99 fix parens when inlining closure local variables 2023-09-21 21:31:15 -07:00
Kevin Reid
cac796acb3 Give unmerge_use a label explaining what it will affect. 2023-09-16 13:29:03 -07:00
bors
9d0ccf01a1 Auto merge of #15597 - rmehri01:fix_promote_local_field_shorthand, r=HKalbasi
Field shorthand overwritten in promote local to const assist

Currently, running `promote_local_to_const` on the following:

```rust
struct Foo {
    bar: usize,
}

fn main() {
    let $0bar = 0;
    let foo = Foo { bar };
}
```

Results in:

```rust
struct Foo {
    bar: usize,
}

fn main() {
    const BAR: usize = 0;
    let foo = Foo { BAR };
}
```

But instead should be something like:

```rust
struct Foo {
    bar: usize,
}

fn main() {
    const BAR: usize = 0;
    let foo = Foo { bar: BAR };
}
```
2023-09-16 16:48:21 +00:00
Ryan Mehri
cd0a89ac4f fix: field shorthand overwritten in promote local to const assist 2023-09-11 10:59:23 -07:00
Ali Bektas
145a101fe8 Deunwrap add_missing_match_arms 2023-09-11 14:09:19 +02:00
Ali Bektas
893e19137e Make assist lazy again 2023-09-11 13:33:26 +02:00
Ali Bektas
0863024b1a Make assist lazy again 2023-09-11 13:31:42 +02:00
Ryan Mehri
25b1b3e753 feat: add support for other ADT types and destructuring patterns 2023-09-10 22:21:12 -07:00
Ali Bektas
2fdf7e4b75 v3 2023-09-10 23:15:37 +02:00
Ali Bektas
a66dbd11ed v2 2023-09-10 23:15:05 +02:00
Ali Bektas
b316bccc97 replace for loops with sth more idiomatic 2023-09-10 23:15:05 +02:00
Ali Bektas
9c6257138d Deunwrap convert_comment_block 2023-09-10 23:15:05 +02:00
Ali Bektas
0f1673c6f1 v3 2023-09-10 23:00:19 +02:00
Ali Bektas
6f7460484a v2 2023-09-10 22:45:09 +02:00
Ali Bektas
35e0d800f0 Deunwrap extract_function 2023-09-10 22:45:09 +02:00
Ali Bektas
38491fcf07 v3 2023-09-10 22:39:07 +02:00
Ali Bektas
68d24b69d4 Deunwrap inline call v2 2023-09-10 22:39:07 +02:00
Ali Bektas
5683df2965 Deunwrap inline call 2023-09-10 22:39:07 +02:00
Ryan Mehri
7ba2e130b9 fix: add checks for overwriting incorrect ancestor 2023-09-10 00:21:11 -07:00
Ryan Mehri
2e13aed3bc feat: support cross module imports 2023-09-09 11:59:59 -07:00