Commit graph

7755 commits

Author SHA1 Message Date
Alex Macleod
850f2f5b61 Update ui_test to 0.21.2 2023-10-08 19:35:41 +01:00
koka
68d2082d69
Fix ice 2023-10-07 01:28:06 +09:00
Philipp Krones
82c3064c47
Merge remote-tracking branch 'upstream/master' into rustup 2023-10-06 17:31:44 +02:00
bors
7217c0f3ac Auto merge of #11628 - koka831:fix/11625, r=blyxyas
Improve `redundant_locals` help message

Fixes #11625

AFAIK, `span_lint_and_help` points the beginning of spans when we pass multiple spans to the second argument, so This PR I also modified its help span and its message.

lint result of the given example in the issue will be:

```console
error: redundant redefinition of a binding `apple`
 --> src/main.rs:5:5
  |
5 |     let apple = apple;
  |     ^^^^^^^^^^^^^^^^^^
  |
help: `apple` is initially defined here
 --> src/main.rs:4:9
  |
4 |     let apple = 42;
  |         ^^^^^
  = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#redundant_locals
```

I hope that this change might help reduce user confusion, but I'd appreciate alternative suggestions:)

changelog: [`redundant_locals`]: Now points at the rebinding of the variable
2023-10-06 15:06:00 +00:00
bors
279127ce2e Auto merge of #11611 - Alexendoo:items-after-test-module-check-crate, r=blyxyas
Fix `items_after_test_module` for non root modules, add applicable suggestion

Fixes #11050
Fixes #11153

changelog: [`items_after_test_module`]: Now suggests a machine-applicable suggestion.
changelog: [`items:after_test_module`]: Also lints for non root modules
2023-10-06 14:19:45 +00:00
koka
48d2770e52
Improve redundant_locals help message 2023-10-06 22:18:11 +09:00
Alex Macleod
dcc400191e Fix items_after_test_module for non root modules, add applicable suggestion 2023-10-06 12:46:04 +00:00
Michael Goulet
c5c6d703de Point to closure return instead of output if defaulted 2023-10-04 21:09:54 +00:00
Michael Goulet
56794fa5f1 Fix clippy 2023-10-04 21:09:54 +00:00
bors
b437069f59 Auto merge of #11603 - koka831:fix/11599, r=y21
Fix: avoid changing drop order

Fixes https://github.com/rust-lang/rust-clippy/issues/11599

changelog: [`redundant_locals`] No longer lints which implements Drop trait to avoid reordering
2023-10-03 15:49:28 +00:00
koka
c7152679ef
Apply review suggestions from @y21 2023-10-04 00:13:53 +09:00
koka
1a56f90ee5
Fix: avoid changing drop order 2023-10-03 21:28:01 +09:00
koka
e465264d47
Avoid invoking ignored_unit_patterns in macro definition 2023-10-03 20:36:35 +09:00
Michael Goulet
2c525fd758 Point to full async fn for future 2023-10-03 02:25:32 +00:00
bors
81400e2db8 Auto merge of #11589 - koka831:fix/10198, r=giraffate
std_instead_of_core: avoid lint inside of proc-macro

- fixes https://github.com/rust-lang/rust-clippy/issues/10198

note: The lint for the reported `thiserror::Error` has been suppressed by [Don't lint unstable moves in std_instead_of_core](https://github.com/rust-lang/rust-clippy/pull/9545/files#diff-2cb8a24429cf9d9898de901450d640115503a10454d692dddc6a073a299fbb7eR29) because `thiserror::Error`  internally implements `std::error::Error for (derived struct)`.

changelog: [`std_intead_of_core`]: avoid linting inside proc-macro

I confirmed this change fixes the problem:
<details>
<summary>test result without the change</summary>

```console
error: used import from `std` instead of `core`
  --> tests/ui/std_instead_of_core.rs:65:14
   |
LL |     #[derive(ImplStructWithStdDisplay)]
   |              ^^^^^^^^^^^^^^^^^^^^^^^^
   |
   = note: this error originates in the derive macro `ImplStructWithStdDisplay` (in Nightly builds, run with -Z macro-backtrace for more info)
```
</details>
2023-10-03 01:26:29 +00:00
bors
08c429f241 Auto merge of #11596 - blyxyas:fix-fp-needless_pass_by_ref_mut, r=Jarcho
Move `needless_pass_by_ref_mut`: `suspicious` -> `nursery`

[Related to [this Zulip thread](https://rust-lang.zulipchat.com/#narrow/stream/257328-clippy/topic/needless_pass_by_ref_mut.20isn't.20ready.20for.20stable)]

`needless_pass_by_ref_mut` has been released with some important bugs (notably having a lot of reported false positives and an ICE). So it may not be really ready for being in stable until these problems are solved. This PR changes the lint's category from `suspicious` to `nursery`, just that.
changelog: none
2023-10-02 18:40:32 +00:00
blyxyas
07e63291ec
Modify tests to account for the new allow-by-default needless_pass_by_ref_mut 2023-10-02 20:14:43 +02:00
Alex Macleod
258b9a8562 Don't escape unicode escape braces in print_literal 2023-10-01 21:43:09 +00:00
bors
3169423ce9 Auto merge of #115670 - Zoxc:outline-panic-macro-1, r=Mark-Simulacrum
Partially outline code inside the panic! macro

This outlines code inside the panic! macro in some cases. This is split out from https://github.com/rust-lang/rust/pull/115562 to exclude changes to rustc.
2023-10-01 05:56:47 +00:00
Victor Song
e683e3eeac Don't lint manual_non_exhaustive when enum explicitly marked as non_exhaustive
There are cases where users create a unit variant for the purposes
of tracking the number of variants for an nonexhaustive enum.
We should check if an enum is explicitly marked as nonexhaustive
before reporting `manual_non_exhaustive` in these cases. Fixes #11583
2023-09-30 22:57:54 -05:00
bors
0e43a04fab Auto merge of #11587 - y21:into_iter_without_iter, r=Jarcho
new lint: `into_iter_without_iter`

Closes #9736 (part 2)

This implements the other lint that my earlier PR missed: given an `IntoIterator for &Type` impl, check that there exists an inherent `fn iter(&self)` method.

changelog: new lint: `into_iter_without_iter`

r? `@Jarcho` since you reviewed #11527 I figured it makes sense for you to review this as well?
2023-09-30 18:43:37 +00:00
y21
8eb586d154 new lint: into_iter_without_iter 2023-09-30 19:38:16 +02:00
koka
d5cc97e32c
Add macro for test which use std internally 2023-10-01 00:47:57 +09:00
bors
b00236d7f0 Auto merge of #11580 - y21:issue11579, r=Jarcho
[`manual_let_else`]: only omit block if span is from same ctxt

Fixes #11579.

The lint already had logic for omitting a block in `else` if a block is already present, however this didn't handle the case where the block is from a different expansion/syntax context. E.g.
```rs
macro_rules! panic_in_block {
  () => { { panic!() } }
}

let _ = match Some(1) {
  Some(v) => v,
  _ => panic_in_block!()
};
```
It would see this in its expanded form as `_ => { panic!() }` and think it doesn't have to include a block in its suggestion because it is already there, however that's not true if it's from a different expansion like in this case.

changelog: [`manual_let_else`]: only omit block in suggestion if the block is from the same expansion
2023-09-29 18:49:57 +00:00
y21
2d2017942a [manual_let_else]: only omit block if span is from same ctxt 2023-09-29 16:54:50 +02:00
y21
330ebbb9f9 new lint: iter_without_into_iter 2023-09-28 22:22:36 +02:00
koka
b413bf6c4e
Fix index of the remaining positional arguments 2023-09-28 17:34:02 +09:00
bors
4494b6947f Auto merge of #11569 - Alexendoo:needless-raw-string-descr, r=llogiq
Describe the type of string in raw_strings lints

changelog: none
2023-09-26 21:16:33 +00:00
Alex Macleod
ec2f62677f Add manual_hash_one lint 2023-09-26 13:49:15 +00:00
Alex Macleod
6cdff10778 Describe the type of string in raw_strings lints 2023-09-26 11:49:44 +00:00
koka
fab90003b8
Do not lint when imported item contains underscore 2023-09-26 12:58:48 +09:00
Philipp Krones
772296c50e Merge commit '7671c283a50b5d1168841f3014b14000f01dd204' into clippyup 2023-09-25 11:28:58 +02:00
Philipp Krones
81fe8dc084
Merge remote-tracking branch 'upstream/master' into rustup 2023-09-25 10:26:11 +02:00
bors
aa137a7e57 Auto merge of #11552 - jonboh:ice_threshold_0_enum_variants, r=y21
prevent ice when threshold is 0 and enum has no variants

changelog: [`enum_variant_names`]: prevent ice when threshold is 0 and enum has no variants

r? `@y21`

Fixes the same ice issue raised during review of https://github.com/rust-lang/rust-clippy/pull/11496
2023-09-24 20:58:58 +00:00
jonboh
b3e262acfd add ui-toml test 2023-09-24 18:25:20 +02:00
bors
94fc43121f Auto merge of #11560 - y21:ui-toml-tests, r=Alexendoo
Add missing tests for configuration options

I noticed that a lot of lints didn't have test(s) for their configuration. This leads to issues like #11481 where the lint just does nothing with it.

This PR adds tests for *almost*[^1] all of the lints with a configuration that didn't have a test in ui-toml.
The tests that I wrote here are usually two cases: one for where it's right above or under the limit set by the config where it shouldn't lint and another one for right above where it should.

changelog: none

[^1]: allow-one-hash-in-raw-strings is ignored by needless_raw_string_hashes
2023-09-24 15:09:33 +00:00
y21
6e80db96ea use a smaller number in test that fits in usize for 32 bit 2023-09-24 16:29:26 +02:00
y21
6ce61d543a add missing configuration tests 2023-09-24 15:19:18 +02:00
Camille GILLOT
19e160fe2f Bless clippy. 2023-09-23 13:47:30 +00:00
bors
50139e6ad2 Auto merge of #11551 - Meczka:fix-fp-needless-pass-by-ref-mut, r=xFrednet
fixed fp caused by moving &mut reference inside of a closure

changelog: [`needless_pass_by_ref mut`]: fixes false positive caused by not covering mutable references passed to a closure inside of a fuction
fixes #11545
2023-09-22 21:07:46 +00:00
pc-linux
ab51f66ec0 added tests 2023-09-22 18:04:57 +02:00
Ralf Jung
1fac304bf5 adjust how closure/generator types and rvalues are printed 2023-09-21 22:20:58 +02:00
bors
33f084ef78 Auto merge of #11544 - Alexendoo:hir-ty-to-ty, r=Jarcho
Remove most usage of `hir_ty_to_ty`

Removes the usages where there's a suitable query or the type was already available elsewhere. The remaining cases would all require more involved changes

changelog: none

r? `@Jarcho`
2023-09-21 15:25:36 +00:00
bors
4d143d7e16 Auto merge of #11518 - mojave2:issue-11420, r=Alexendoo
fix FP with needless_raw_string_hashes

changelog: Fix [`needless_raw_string_hashes`]: Continue the lint checking of raw string when `needless_raw_strings` is allowed.

fix #11420
2023-09-21 12:27:24 +00:00
bors
c55669513a Auto merge of #11522 - y21:redundant_guards_pat_lhs, r=giraffate
[`redundant_guards`]: lint if the pattern is on the left side

A tiny improvement to the `redundant_guards` lint. There's no associated issue for this, just noticed it while going through the code.
Right now it warns on `Some(x) if x == 2` and suggests `Some(2)`, but it didn't do that for `Some(x) if 2 == x` (i.e. when the local is on the right side and the pattern on the left side).

changelog: [`redundant_guards`]: also lint if the pattern is on the left side
2023-09-20 23:57:25 +00:00
Alex Macleod
01c25a8eb6 Remove most usage of hir_ty_to_ty 2023-09-20 14:05:00 +00:00
bors
ddbe110081 Auto merge of #11170 - tgross35:undocumented-unsafe-blocks-defaults, r=Centri3
Change defaults of `accept-comment-above-statement` and `accept-comment-above-attributes`

This patch sets the two configuration options for `undocumented_unsafe_blocks` to `true` by default: these are `accept-comment-above-statement` and `accept-comment-above-attributes`. Having these values `false` by default prevents what many users would consider clean code, e.g. placing the `// SAFETY:` comment above a single-line functino call, rather than directly next to the argument.

This was originally discussed in https://github.com/rust-lang/rust-clippy/issues/11162

changelog: [`undocumented_unsafe_blocks`]: set
`accept-comment-above-statement` and `accept-comment-above-attributes` to `true` by default.
2023-09-20 08:06:54 +00:00
bors
ece3878c8c Auto merge of #11492 - GuillaumeGomez:async-fn-returned-closure, r=Centri3
Fix mutaby used async function argument in closure for `needless_pass_by_ref_mut`

Fixes https://github.com/rust-lang/rust-clippy/issues/11380.

The problem was that it needed to go through closures as well in async functions to correctly find the mutable usage of async function arguments.

changelog: Correctly handle mutable usage of async function arguments in closures.

r? `@Centri3`
2023-09-20 07:48:54 +00:00
Trevor Gross
1b3e5dd0fc Change default configuration of undocumented_unsafe_blocks
This patch sets the two configuration options for
`undocumented_unsafe_blocks` to `true` by default: these are
`accept-comment-above-statement` and `accept-comment-above-attributes`.
Having these values `false` by default prevents what many users would
consider clean code, e.g. placing the `// SAFETY:` comment above a
single-line functino call, rather than directly next to the argument.

changelog: [`undocumented_unsafe_blocks`]: set
`accept-comment-above-statement` and `accept-comment-above-attributes`
to `true` by default.
2023-09-20 03:41:33 -04:00
bors
f464149b8f Auto merge of #11516 - mojave2:issue-11458, r=giraffate
fix cast_lossless with macro call

changelog: fix [`cast_lossless`] in the case when the cast operand is a macro call

fix #11458
2023-09-20 00:00:06 +00:00