Remove unnecessary impl sorting in queries and metadata
Removes unnecessary impl sorting because queries already return their keys in HIR definition order: https://github.com/rust-lang/rust/issues/120371#issuecomment-1926422838
r? `@cjgillot` or `@lcnr` -- unless I totally misunderstood what was being asked for here? 😆fixes#120371
[`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
Clean up a few minor refs in `format!` macro, as it has a performance cost. Apparently the compiler is unable to inline `format!("{}", &variable)`, and does a run-time double-reference instead (format macro already does one level referencing).
Inlining format args prevents accidental `&` misuse.
Lintcheck: Update crates and expand CI testset to 200 crates
This PR adds a new `ci_crates.toml` to lintcheck for our CI. The 200 crates take about 14 minutes, which is slightly more than the 10 I aimed for but still reasonable. The testset is constructed from:
* 5 crates that compile to binaries
* 4 crates that have been mentioned in ICE issues
* 1 crates "random" crates from `lintcheck_crates.toml`
* 190 crates from the top 200 crates from crates.io
During testing, I noticed a few panics in lintcheck. I've fixed them where possible, or at least improved the error message.
The new test set generates 500+ MB of json lints, which are compressed to a ~24mb artifact.
---
This PR also updates our `lintcheck_crates.toml`. I mainly updated the versions, removed some very outdated crates, and added some new ones. I targeted 25 crates as those are pretty fast to lint and a good precursor for our CI.
---
Optional TODO:
* It's likely that some crates are compiled several times. We could potentially safe some time, by using `--recursive` in our CI.
This is something I want to investigate, but it shouldn't be a blocker for this PR.
---
r? `@Alexendoo`
changelog: none
Represent type-level consts with new-and-improved `hir::ConstArg`
### Summary
This is a step toward `min_generic_const_exprs`. We now represent all const
generic arguments using an enum that differentiates between const *paths*
(temporarily just bare const params) and arbitrary anon consts that may perform
computations. This will enable us to cleanly implement the `min_generic_const_args`
plan of allowing the use of generics in paths used as const args, while
disallowing their use in arbitrary anon consts. Here is a summary of the salient
aspects of this change:
- Add `current_def_id_parent` to `LoweringContext`
This is needed to track anon const parents properly once we implement
`ConstArgKind::Path` (which requires moving anon const def-creation
outside of `DefCollector`).
- Create `hir::ConstArgKind` enum with `Path` and `Anon` variants. Use it in the
existing `hir::ConstArg` struct, replacing the previous `hir::AnonConst` field.
- Use `ConstArg` for all instances of const args. Specifically, use it instead
of `AnonConst` for assoc item constraints, array lengths, and const param
defaults.
- 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 https://github.com/rust-lang/rust/issues/127009.
### Followup items post-merge
- Use `ConstArgKind::Path` for all const paths, not just const params.
- Fix (no github dont close this issue) #127009
- If a path in generic args doesn't resolve as a type, try to resolve as a const
instead (do this in rustc_resolve). Then remove the special-casing from
`rustc_ast_lowering`, so that all params will automatically be lowered as
`ConstArgKind::Path`.
- (?) Consider making `const_evaluatable_unchecked` a hard error, or at least
trying it in crater
r? `@BoxyUwU`
Lintcheck: Add URL to lint emission place in diff
This PR adds links to the emission code in our lintcheck CI. When reviewing changes, I would like to be able to see the bigger context, which isn't always included in the lint message. This PR adds a nice link to the lintcheck diff that allows for simple investigation of the code in question.
At first, I wanted to simply use the doc.rs links and call it a day, but then I figured out that some crates might have their source files remapped. Cargo was the crate that made me notice this. As a response, I made the link configurable. (See https://github.com/rust-lang/docs.rs/issues/2551 for a detailed explanation and possible solution to remove this workaround in the future.)
It's probably easiest to review the individual commits. The last one is just a dummy to showcase the change.
[🖼️ rendered 🖼️](https://github.com/rust-lang/rust-clippy/actions/runs/9960834924?pr=13104)
---
r? `@Alexendoo`
changelog: none
---
That's it, I hope that everyone who's reading this has a beautiful day :D
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
Create lint passes using `Conf`
This slightly reduces the amount of code changes needed to add a config to a lint and makes things makes things more consistent between passes. A dependence on the config being a static reference is also added. This would only ever be a problem if multiple crates end up compiled in a single process.
Other changes include:
* Removing useless `#[derive(..)]`s
* Removing `#[must_use]` on lint pass constructors.
* Unified the parsing of the `DisallowedPath` struct in lint passes.
* Update `disallowed_types` and `await_holding_invalid` messages to be consistent with other disallowed lints.
* Remove the `(from clippy.toml)` message. I plan on having all the configured lints point to point to a span in `clippy.toml` which will be more useful.
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.
fix [`excessive_precision`] suggestions on floats written in scientific notation
fixes#12954
changelog: fix [`excessive_precision`] suggestions on float literal written in scientific notation
Bump ui_test version
r? `@alexendoo`
the rustfix diff is caused by https://github.com/oli-obk/ui_test/pull/244
This should solve the issues around missing summaries at the end
changelog: none
