feat(manual_is_variant): Check for `.iter().any()` by qmwrygdxpzc · Pull Request #15989 · rust-lang/rust-clippy

qmwrygdxpzc · 2025-10-30T23:26:13Z

Now the linter checks for calling the iter or into_iter method
followed by calling the all or any method.

We have used these methods by mistake ourselves. The developer was not
familiar with the is_some_and method. He tried to use the any
method on an option. The compiler suggested using iter first. The
developer did not get to learn about the idiomatic method.

The lint search_is_some involves converting a closure that takes the
reference into one that takes the value. The code is split between the
function for the lint and a shared function. The new feature that we
are adding involves the same transformation. To do so it needs some of
the same code that is currently in the search_is_some lint. So it is
moved from the lint into the shared function.

Before this patch, line 799 in the file clippy_utils/src/sugg.rs had a
check. It made the function return none in case the closure did not
dereference the argument. However, the code for the lint
search_is_some would then fall back to using the original closure.
That achieved the same effect as the check not existing. So the check
is now removed.

changelog: [manual_is_variant]: check for calling the iter or
into_iter method followed by calling the all or any method

changelog: [manual_is_variant]: skip the check when the MSRV is before
1.70

rustbot · 2025-10-30T23:26:18Z

r? @Alexendoo

rustbot has assigned @Alexendoo.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

qmwrygdxpzc · 2025-10-30T23:37:08Z

If the input uses .iter().any() on a value that is not Copy, the current version of the patch makes the linter suggest using .as_ref().is_some_and(). The suggestion is longer than the original code. Should the linter skip it instead?

Alexendoo · 2025-11-15T15:23:27Z

+        // suggest `any(|x| ..)` instead of `any(|&x| ..)` for `find(|&x| ..).is_some()`
+        if matches!(closure_arg.pat.kind, PatKind::Ref(..)) {
+            let mut applicability = Applicability::MachineApplicable;
+            let suggestion =
+                snippet_with_applicability(cx, closure.span, "..", &mut applicability).replacen('&', "", 1);
+            return Some(DerefClosure {
+                applicability,
+                suggestion,
+            });
+        }
+        if !matches!(strip_pat_refs(closure_arg.pat).kind, PatKind::Binding(..)) {
+            return None;
+        }


I believe this should still live outside of the clippy_utils method, though it's not currently used elsewhere it's specified as being generally applicable

With this patch, both search_is_some and manual_is_variant_and use it.

Yeah for those it might be fine, but may not be applicable to other future uses

cc @Jarcho on this though since I'm not 100% sure on what deref_closure_args is doing

The deref_closure_args function changes a closure that closure that takes the reference into one that takes the value. A closure can take a reference in these forms: |x| *x, or |&x| x. Without my patch, only the search_is_some lint does so to the closure. The first form is handled in the deref_closure_args function. The second form is handled in the file for the lint. It makes sense for other lints, such as the manual_is_variant that I am modifying, to modify both forms if it needs to make the closure take the value. So I moved the logic for handling both forms into the library function.

qmwrygdxpzc · 2026-02-17T18:39:09Z

@profetia Can you please take a look? You worked on the same check recently.

The lint `search_is_some` involves converting a closure that takes the reference into one that takes the value. The code is split between the function for the lint and a shared function. The new feature that we are adding involves the same transformation. To do so it needs some of the same code that is currently in the `search_is_some` lint. So it is moved from the lint into the shared function. Before this patch, line 799 in the file `clippy_utils/src/sugg.rs` had a check. It made the function return none in case the closure did not dereference the argument. However, the code for the lint `search_is_some` would then fall back to using the original closure. That achieved the same effect as the check not existing. So the check is now removed.

Now the linter checks for calling the `iter` or `into_iter` method followed by calling the `all` or `any` method. We have used these methods by mistake ourselves. The developer was not familiar with the `is_some_and` method. He tried to use the `any` method on an option. The compiler suggested using `iter` first. The developer did not get to learn about the idiomatic method. changelog: [`manual_is_variant`]: check for calling the `iter` or `into_iter` method followed by calling the `all` or `any` method

changelog: [`manual_is_variant`]: skip the check when the MSRV is before 1.70

rustbot · 2026-03-06T17:29:25Z

This PR was rebased onto a different master commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

qmwrygdxpzc · 2026-03-06T17:32:13Z

No reviewer has said things for a long time. Does that mean we had better give up on the pull request?

qmwrygdxpzc · 2026-03-26T23:42:49Z

@Alexendoo

No reviewer has said things for a long time. Does that mean we had better give up on the pull request?

profetia · 2026-03-27T14:11:07Z

You can use r? clippy to find a new reviewer

qmwrygdxpzc · 2026-03-27T16:22:57Z

r? clippy

qmwrygdxpzc · 2026-05-07T22:24:18Z

@ada4a This patch has taken a long time. Is the reviewer busy or is there a problem with the patch?

profetia · 2026-05-15T10:13:11Z

You might use r? clippy to find a new reviewer

qmwrygdxpzc · 2026-05-15T21:49:36Z

It takes too long to get this patch reviewed.

rustbot assigned Alexendoo Oct 30, 2025

rustbot added the S-waiting-on-review Status: Awaiting review from the assignee but also interested parties label Oct 30, 2025