Fixes#13793.
Interestingly enough, to actually check that the macro call has at least
two arguments, we need to use the rust lexer after getting the original
source code snippet.
changelog: Add new `useless_concat` lint
It is not enough to check if an expression type implements `Drop` to
determine whether it can have a significant side-effect.
Also, add tests for unchecked cases which were explicitly handled in the
code, such as checking for side effect in a `struct`'s fields, or in its
base expression.
Fixrust-lang/rust-clippy#14592
changelog: [`no_effect_underscore_binding`]: do not propose to remove
the assignment of an object if dropping it might indirectly have a
visible side effect
Do not attempt to fetch a snippet from expansion. Without this change,
the inside of macros could[*] be shown as the source of the problem.
[*] Due to the way the source code is processed and reparsed in this
macro, the declarative macro has to be located outside the current
source file for the bug to appear. Otherwise, the macro call itself will
be (mis)identified as a potential `struct` field definition and the lint
will not trigger.
changelog: [`empty_struct_with_brackets`]: do not lint code coming from
macro expansion
A representation attribute `#[repr(…)]` might indicate that the ordering
of the fields or the variants is dictated by the API the code is
interfacing with. Better not lint with `arbitrary_source_item_ordering`
in this case.
changelog: [`arbitrary_source_item_ordering`]: do not lint inside items
with `#[repr]` attribute
fixes#12979fixes#12951fixes#13233
Tests still need to be finished and the docs still need to be updated,
but this should otherwise ready.
changelog: Lint `declare_interior_mutable_const` and
`borrow_interior_mutable_const` more precisely
Was disabled in rust-lang/rust-clippy#10855 due to a broken suggestion.
This will only lint if a type is not proveded (e.g. `T::deref`).
changelog: none
Using the function declaration, by stripping the body, is enough to
convey the lint message.
changelog: [`unnecessary_wraps`]: do not include the whole body in the
lint span
Closesrust-lang/rust-clippy#14773
Clippy should not lint 2 blocks expression for comparison_chain.
Fixesrust-lang/rust-clippy#4725
changelog: [`comparison_chain`]: do not lint 2 blocks expression
Based on rust-lang/rust-clippy#13223
Various refactoring on `clippy_dev` before upgrading rename and
splitting `clippy_lints` into multiple crates.
Some improvements:
* The working directory is set to the root clippy directory. Running
from a subdirectory was kind of supported before sometimes. Now it just
works.
* File won't be written unless they're actually updated. Most of the
time they weren't written, but a few cases slipped through.
* Buffers are reused more for the negligible speed boost.
changelog: None
When a `non_std_lazy_statics` warning is generated about an item type
which can be replaced by a standard library one, ensure that the lint
happens on the item HIR node so that it can be expected.
changelog: [`non_std_lazy_statics`]: generate the warning onto the right
node
Fixesrust-lang/rust-clippy#14729
Note that this doesn't change anything on lints generated for the
`lazy_static::lazy_static` macro because the `expect` attribute cannot
be applied to a macro.
Added lint for catching `&[foo.clone()]` where foo is a reference and
suggests `std::slice::from_ref(foo)`.
changelog: new lint: [`cloned_ref_to_slice_refs`]
changelog: [`type_repetition_in_bounds`]: include all generic bounds in
suggestion, clearer message which now includes the repeated type name
Fixesrust-lang/rust-clippy#14744
If an expression is not going to return from the current function of
closure, it should not get linted.
This also allows `return` expression to be linted, in addition to the
final expression. Those require blockification and proper indentation.
changelog: [`return_and_then`]: only lint returning expressions
Fixesrust-lang/rust-clippy#14781
relates to rust-lang/rust-clippy#14653
changelog: [`elidable_lifetime_names`]: Fix clippy version that this
lint was introduced in.
It might be good to have some automation to check that the version is
correct.
IIUC the version should be the current nightly version at the time a PR
is merged.
If a release happens while a PR is open, the version needs to be bumped
- this is easy to forget.
Fixes https://github.com/rust-lang/rust-clippy/issues/13973.
I don't think we can make `fn_to_numeric_cast_any` to be emitted in some
special cases. Its category cannot be changed at runtime.
I think in this case, the best might be a specialized new lint so we can
target exactly what we want.
----
changelog: Add new `confusing_method_to_numeric_cast` lint
If an expression is not going to return from the current function of
closure, it should not get linted.
This also allows `return` expression to be linted, in addition to the
final expression. Those require blockification and proper indentation.
The `mir_drops_elaborated_and_const_checked` query result has been
stolen already and cannot be borrowed again. Use the `optimized_mir`
query result instead.
The problem is that `check_fn` is triggered by both function and
closure, and `visit_expr` can visit expressions in another closures
within a function or closure.
So just skip walking in a inner closure.
changelog: Fix [`unnecessary_unwrap`] emitted twice in closure which
inside in a function or another closure.
Fixes: rust-lang/rust-clippy#14763
Samuel Tardieu