Stabilize const `{integer}::from_str_radix` i.e. `const_int_from_str`
This PR stabilizes the feature `const_int_from_str`.
- ACP Issue: rust-lang/libs-team#74
- Implementation PR: rust-lang/rust#99322
- Part of Tracking Issue: rust-lang/rust#59133
API Change Diff:
```diff
impl {integer} {
- pub fn from_str_radix(src: &str, radix: u32) -> Result<Self, ParseIntError>;
+ pub const fn from_str_radix(src: &str, radix: u32) -> Result<Self, ParseIntError>;
}
impl ParseIntError {
- pub fn kind(&self) -> &IntErrorKind;
+ pub const fn kind(&self) -> &IntErrorKind;
}
```
This makes it easier to parse integers at compile-time, e.g.
the example from the Tracking Issue:
```rust
env!("SOMETHING").parse::<usize>().unwrap()
```
could now be achived with
```rust
match usize::from_str_radix(env!("SOMETHING"), 10) {
Ok(val) => val,
Err(err) => panic!("Invalid value for SOMETHING environment variable."),
}
```
rather than having to depend on a library that implements or manually implement the parsing at compile-time.
---
Checklist based on [Libs Stabilization Guide - When there's const involved](https://std-dev-guide.rust-lang.org/development/stabilization.html#when-theres-const-involved)
I am treating this as a [partial stabilization](https://std-dev-guide.rust-lang.org/development/stabilization.html#partial-stabilizations) as it shares a tracking issue (and is rather small), so directly opening the partial stabilization PR for the subset (feature `const_int_from_str`) being stabilized.
- [x] ping Constant Evaluation WG
- [x] no unsafe involved
- [x] no `#[allow_internal_unstable]`
- [ ] usage of `intrinsic::const_eval_select` rust-lang/rust#124625 in `from_str_radix_assert` to change the error message between compile-time and run-time
- [ ] [rust-labg/libs-api FCP](https://github.com/rust-lang/rust/pull/124941#issuecomment-2207021921)
needless_borrows_for_generic_args: Fix for &mut
This commit fixes a bug introduced in #12706, where the behavior of the lint has been changed, to avoid suggestions that introduce a move. The motivation in the commit message is quite poor (if the detection for significant drops is not sufficient because it's not transitive, the proper fix would be to make it transitive). However, #12454, the linked issue, provides a good reason for the change — if the value being borrowed is bound to a variable, then moving it will only introduce friction into future refactorings.
Thus #12706 changes the logic so that the lint triggers if the value being borrowed is Copy, or is the result of a function call, simplifying the logic to the point where analysing "is this the only use of this value" isn't necessary.
However, said PR also introduces an undocumented carveout, where referents that themselves are mutable references are treated as Copy, to catch some cases that we do want to lint against. However, that is not sound — it's possible to consume a mutable reference by moving it.
To avoid emitting false suggestions, this PR reintroduces the referent_used_exactly_once logic and runs that check for referents that are themselves mutable references.
Thinking about the code shape of &mut x, where x: &mut T, raises the point that while removing the &mut outright won't work, the extra indirection is still undesirable, and perhaps instead we should suggest reborrowing: &mut *x. That, however, is left as possible future work.
Fixes#12856
changelog: none
Implement lint against ambiguous negative literals
This PR implements a lint against ambiguous negative literals with a literal and method calls right after it.
## `ambiguous_negative_literals`
(deny-by-default)
The `ambiguous_negative_literals` lint checks for cases that are confusing between a negative literal and a negation that's not part of the literal.
### Example
```rust,compile_fail
-1i32.abs(); // equals -1, while `(-1i32).abs()` equals 1
```
### Explanation
Method calls take precedence over unary precedence. Setting the precedence explicitly makes the code clearer and avoid potential bugs.
<details>
<summary>Old proposed lint</summary>
## `ambiguous_unary_precedence`
(deny-by-default)
The `ambiguous_unary_precedence` lint checks for use the negative unary operator with a literal and method calls.
### Example
```rust
-1i32.abs(); // equals -1, while `(-1i32).abs()` equals 1
```
### Explanation
Unary operations take precedence on binary operations and method calls take precedence over unary precedence. Setting the precedence explicitly makes the code clearer and avoid potential bugs.
</details>
-----
Note: This is a strip down version of https://github.com/rust-lang/rust/pull/117161, without the binary op precedence.
Fixes https://github.com/rust-lang/rust/issues/117155
`@rustbot` labels +I-lang-nominated
cc `@scottmcm`
r? compiler
Fix running compile-test under cargo nextest
`ui_test` itself has `cargo nextest` support which we now use - https://github.com/oli-obk/ui_test/pull/161
It prints `ui_test` as its test name whereas we printed `compile_test`, this ended up being treated as a test name filter causing all the tests to be filtered out
changelog: none
Misc refactorings part 5
`toplevel_ref_arg` gets a small fix so it can be allowed on function arguments. Otherwise just some rearrangements.
changelog: none
add lint for recreation of an entire struct
This lint makes Clippy warn about situations where an owned struct is
essentially recreated by moving all its fields into a new instance of
the struct. The lint is not machine-applicable because the source
struct may have been partially moved.
This lint originated in something I spotted during peer review. While
working on their branch a colleague ended up with a commit where a
function returned a struct that 1:1 replicated one of its owned inputs
from its members. Initially I suspected they hadn’t run their code
through Clippy but AFAICS there is no lint for this situation yet.
changelog: new lint: [`redundant_owned_struct_recreation`]
### New lint checklist
- \[+] Followed [lint naming conventions][lint_naming]
- \[+] Added passing UI tests (including committed `.stderr` file)
- \[+] `cargo test` passes locally
- \[+] Executed `cargo dev update_lints`
- \[+] Added lint documentation
- \[+] Run `cargo dev fmt`
[`pathbuf_init_then_push`]: Checks for calls to `push` immediately a…
changelog: [`pathbuf_init_then_push`]: new lint: Checks for calls to `push` immediately after creating a new `PathBuf`
Just a mirror of VEC_INIT_THEN_PUSH
Lint `zero_repeat_side_effects` only if array length is a literal zero
changelog: [`zero_repeat_side_effects` ] Lint only if array length is a literal zero
Fixes#13110 .
r? y21
Show progress while running dogfood test
Outputs the regular cargo progress in colour when running the dogfood test, helpful to see because it can take a long time to run
changelog: none
* Construct lint passes by taking `Conf` by reference.
* Use `HashSet` configs in less places
* Move some `check_crate` code into the pass constructor when possible.
This is a very large commit since a lot needs to be changed in order to
make the tests pass. The salient changes are:
- `ConstArgKind` gets a new `Path` variant, and all const params are now
represented using it. Non-param paths still use `ConstArgKind::Anon`
to prevent this change from getting too large, but they will soon use
the `Path` variant too.
- `ConstArg` gets a distinct `hir_id` field and its own variant in
`hir::Node`. This affected many parts of the compiler that expected
the parent of an `AnonConst` to be the containing context (e.g., an
array repeat expression). They have been changed to check the
"grandparent" where necessary.
- Some `ast::AnonConst`s now have their `DefId`s created in
rustc_ast_lowering rather than `DefCollector`. This is because in some
cases they will end up becoming a `ConstArgKind::Path` instead, which
has no `DefId`. We have to solve this in a hacky way where we guess
whether the `AnonConst` could end up as a path const since we can't
know for sure until after name resolution (`N` could refer to a free
const or a nullary struct). If it has no chance as being a const
param, then we create a `DefId` in `DefCollector` -- otherwise we
decide during ast_lowering. This will have to be updated once all path
consts use `ConstArgKind::Path`.
- We explicitly use `ConstArgHasType` for array lengths, rather than
implicitly relying on anon const type feeding -- this is due to the
addition of `ConstArgKind::Path`.
- Some tests have their outputs changed, but the changes are for the
most part minor (including removing duplicate or almost-duplicate
errors). One test now ICEs, but it is for an incomplete, unstable
feature and is now tracked at #127009.
Add more doc-valid-idents
* "AccessKit" is a commonly used accessibility toolkit used in Rust GUIs.
* "CoreFoundation", "CoreGraphics", "CoreText" are frameworks on Apple OSes.
* "Direct2D", "Direct3D", "DirectWrite" are frameworks on Windows
* "PostScript" is a programming language and is mentioned when talking about text and vector graphics.
* "OpenAL" is an audio framework / API.
* "OpenType" is a font format (TrueType is already mentioned).
* "WebRTC", "WebSocket", "WebTransport" are web networking technologies.
* "NetBSD" and "OpenBSD" are like the already included FreeBSD.
changelog: [`doc_markdown`]: Add AccessKit, CoreFoundation, CoreGraphics, CoreText, Direct2D, Direct3D, DirectWrite, PostScript, OpenAL, OpenType, WebRTC, WebSocket, WebTransport, NetBSD, and OpenBSD to `doc-valid-idents`.
* "AccessKit" is a commonly used accessibility toolkit used in Rust GUIs.
* "CoreFoundation", "CoreGraphics", "CoreText" are frameworks on Apple OSes.
* "Direct2D", "Direct3D", "DirectWrite" are frameworks on Windows
* "PostScript" is a programming language and is mentioned when talking about
text and vector graphics.
* "OpenAL" is an audio framework / API.
* "OpenType" is a font format (TrueType is already mentioned).
* "WebRTC", "WebSocket", "WebTransport" are web networking technologies.
* "NetBSD" and "OpenBSD" are like the already included FreeBSD.
Fix and rename `overflow_check_conditional`
fixes#2457
Other changes:
* Limit the lint to unsigned types.
* Actually check if the operands are the same rather than using only the first part of the path.
* Allow the repeated expression to be anything as long as there are no side effects.
changelog: Rename `overflow_check_conditional` to `panicking_overflow_check` and move to `correctness`
Fix guidance of [`float_cmp`] and [`float_cmp_const`] to not incorrectly recommend `f__::EPSILON` as the error margin.
Using `f32::EPSILON` or `f64::EPSILON` as the floating-point equality comparison error margin is incorrect, yet `float_cmp` has until now recommended this be done. This change fixes the given guidance (both in docs and compiler hints) to not reference these unsuitable constants.
Instead, the guidance now clarifies that the scenarios in which an absolute error margin is usable, provides a sample implementation for using a user-defined absolute error margin (as an absolute error margin can only be used-defined and may be different for different comparisons) and references the floating point guide for a reference implementation of relative error based equality comparison for cases where absolute error margins cannot be identified.
changelog: [`float_cmp`] Fix guidance to not incorrectly recommend `f__::EPSILON` as the error margin.
changelog: [`float_cmp_const`] Fix guidance to not incorrectly recommend `f__::EPSILON` as the error margin.
Fixes#6816
This lint makes Clippy warn about situations where an owned
struct is essentially recreated by moving all its fields into a
new instance of the struct. Until now this lint only triggered
for structs recreated from a base struct.
NB: The new functionality too will cause false positives for the
situation where a non-copy struct consisting of all copy members
is touched again in subsequent code.
Using `f32::EPSILON` or `f64::EPSILON` as the floating-point equality comparison error margin is incorrect, yet `float_cmp` has until now recommended this be done. This change fixes the given guidance (both in docs and compiler hints) to not reference these unsuitable constants.
Instead, the guidance now clarifies that the scenarios in which an absolute error margin is usable, provides a reference implementation of using a user-defined absolute error margin (as an absolute error margin can only be used-defined and may be different for different comparisons) and references the floating point guide for a reference implementation of relative error based equaltiy comparison for when absolute error margin cannot be used.
changelog: Fix guidance of [`float_cmp`] and [`float_cmp_const`] to not incorrectly recommend `f64::EPSILON` as the error margin.
Fixes#6816
`significant_drop_in_scrutinee`: Trigger lint also for scrutinees in `while let` and `if let`
This lint should also work for `if let` and `while let`, so this PR makes it actually work.
For `while let`, I can't think of any reason why this lint shouldn't be enabled. The only problem is that the lint suggests moving the significant drop above the `while let`, which is clearly invalid in the case of `while let`. I don't know if this is fixable, but this PR simply disables the wrong suggestions.
For `if let`, it seems that another lint called `if_let_mutex` has some overlapping functionality. But `significant_drop_in_scrutinee` is a bit stricter, as it will trigger even if the `else` branch does not try to lock the same mutex.
changelog: [`significant_drop_in_scrutinee`]: Trigger lint also for scrutinees in `while let` and `if let`.
r? `@blyxyas` (the third PR as promised in https://github.com/rust-lang/rust-clippy/pull/12740#issuecomment-2094876350, thanks for your review!)
[`missing_const_for_fn`]: fix FP when arg ty is impl trait alias ty
closes: #13009
---
changelog: [`missing_const_for_fn`]: fix FP when arg ty is impl trait alias ty
Refactor `disallowed_methods` and narrow span
Using the span of the call site just produces noisy diagnostics for long calls. Especially multi-line calls.
changelog: none
Rename thread_local_initializer_can_be_made_const to missing_const_for_thread_local
Close#12934
As discussed at #12934 name `thread_local_initializer_can_be_made_const` sounds against convention for other lints which describe the issue/wrong code but not suggestion and it is quite long. The new name take example from existing lint `missing_const_for_fn`
changelog: `thread_local_initializer_can_be_made_const` : Rename to [`missing_const_for_thread_local`]
feat: add cfg_not_test lint
<!--
- \[x] Followed [lint naming conventions][lint_naming]
- \[x] Added passing UI tests (including committed `.stderr` file)
- \[x] `cargo test` passes locally
- \[x] Executed `cargo dev update_lints`
- \[ ] Added lint documentation
- \[x] Run `cargo dev fmt`
[lint_naming]: https://rust-lang.github.io/rfcs/0344-conventions-galore.html#lints
-->
Fixes#11234
changelog: new lint: [`cfg_not_test`]
I don't know whether to lint only the `attr` or also the item associated to it. I guess this would mean putting the check in another place than `check_attribute` but I can't find a way to get the associated item to the attribute.
Also, I'm not sure how to document this lint, I feel like my explications are bad.
Remove `is_in_test_module_or_function`
Uses are replaced with `is_in_test` for consistency with other lints and to simplify the implementation of the lints. This means the module name is no longer checked, but that was a horrible hack from a time when late passes couldn't see `#[cfg(..)]` attributes.
changelog: none
[`unnecessary_to_owned`]: catch `to_owned` on byte slice to create temporary `&str`
Closes#11648
Detects the pattern `&String::from_utf8(bytes.to_vec()).unwrap()` and suggests `core::str::from_utf8(bytes).unwrap()`, which avoids the unnecessary intermediate allocation.
I decided to put this in the existing `unnecessary_to_owned` lint (rather than creating a new lint) for a few reasons:
- we get to use some of its logic (for example, recognizing any of the functions in the `to_owned` family, e.g. `to_vec`)
- the actual inefficient operation that can be avoided here is the call to `.to_vec()`, so this is in a way similar to the other cases caught by `unnecessary_to_owned`, just through a bunch of type conversions
- we can make this more "generic" later and catch other cases, so imo it's best not to tie this lint specifically to the `String` type
changelog: [`unnecessary_to_owned`]: catch `&String::from_utf8(bytes.to_vec()).unwrap()` and suggest `core::str::from_utf8(bytes).unwrap()`
[`missing_const_for_fn`]: fix suggestions for fn with abi that requires `const_extern_fn` feature
closes: #13008
---
changelog: [`missing_const_for_fn`]: fix suggestions for fn with abi that requires `const_extern_fn` feature.
This patch adds a new lint that checks for potentially harder to read
byte char slices: `&[b'a', b'b']` and suggests to replace them with the
easier to read `b"ab"` form.
Signed-Off-By: Marcel Müller <m.mueller@ifm.com>
Co-authored-by: Matthias Beyer <matthias.beyer@ifm.com>
Use iterator to skip validation
Signed-off-by: Marcel Müller <m.mueller@ifm.com>
Suggested-by: Alex Macleod <alex@macleod.io>
Convert quote escapes to proper form
Signed-off-by: Marcel Müller <m.mueller@ifm.com>
Add more convertable test cases
Signed-off-by: Marcel Müller <m.mueller@ifm.com>
Add new lint `hashset_insert_after_contains`
This PR closes https://github.com/rust-lang/rust-clippy/issues/11103.
This is my first PR creating a new lint (and the second attempt of creating this PR, the first one I was not able to continue because of personal reasons). Thanks for the patience :)
The idea of the lint is to find insert in hashmanps inside if staments that are checking if the hashmap contains the same value that is being inserted. This is not necessary since you could simply call the insert and check for the bool returned if you still need the if statement.
changelog: new lint: [hashset_insert_after_contains]
