fix unused_async with macros and sub async block

[hlf20010508]

changelog: [unused_async]:

• check after pro_macro_attribute
• check inside async block

fixes #13199

I've run the test, but failed with the error that I didn't understand.

error: actual output differed from expected
Execute `cargo uibless` to update `tests/ui/unused_async.stderr` to the actual output
--- tests/ui/unused_async.stderr
+++ <stderr output>
 error: unused `async` for function with no await statements
   --> tests/ui/unused_async.rs:12:5
... 46 lines skipped ...
    = help: consider removing the `async` from this function

-error: aborting due to 4 previous errors
+error: unused `async` for function with no await statements

+error: aborting due to 5 previous errors
+

[rustbot]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @Manishearth (or someone else) some time within the next two weeks.

Please see the contribution instructions for more information. Namely, in order to ensure the minimum review times lag, PR authors and assigned reviewers should ensure that the review label (S-waiting-on-review and S-waiting-on-author) stays updated, invoking these commands when appropriate:

• @rustbot author: the review is finished, PR author should check the comments and take action accordingly
• @rustbot review: the author is ready for a review, this PR will be queued again in the reviewer's queue

[hlf20010508]

I think there's one more problem.
The function can be warned if I don't use it. But once I use it in another file, the warning would be gone, I don't even need to call it.

[hlf20010508]

I'm not sure if it's is_node_func_call's problem.

[hlf20010508]

I'm not sure if it's is_node_func_call's problem.

Confirmed.

[Alexendoo]

I would say the macro detection is working as intended - #13199 (comment)

[hlf20010508]

I would say the macro detection is working as intended - #13199 (comment)

Thanks, I didn't know this before.

[hlf20010508]

I think there's one more problem. The function can be warned if I don't use it. But once I use it in another file, the warning would be gone, I don't even need to call it.

I need help. I've fixed this, but didn't pass the issue9695

mod issue9695 {
    use std::future::Future;

    async fn f() {}
    async fn f2() {}
    async fn f3() {}
    //~^ ERROR: unused `async` for function with no await statements

    fn needs_async_fn<F: Future<Output = ()>>(_: fn() -> F) {}

    fn test() {
        let x = f;
        needs_async_fn(x); // async needed in f
        needs_async_fn(f2); // async needed in f2
        f3(); // async not needed in f3
    }
}

f would be warned.

[Manishearth]

r? @Alexendoo

coming off of vacation and will be on planes for a while

[Alexendoo]

@oli-obk any idea what's happening with this diff? When I run it locally I get a more reasonable output of

--- tests/ui/unused_async.stderr
+++ <stderr output>
 error: unused `async` for function with no await statements
   --> tests/ui/unused_async.rs:12:5
... 17 lines skipped ...
 
 error: unused `async` for function with no await statements
+  --> tests/ui/unused_async.rs:43:5
+   |
+LL |     async fn f() {}
+   |     ^^^^^^^^^^^^^^^
+   |
+   = help: consider removing the `async` from this function
+
+error: unused `async` for function with no await statements
   --> tests/ui/unused_async.rs:45:5
    |
... 25 lines skipped ...
    = help: consider removing the `async` from this function
 
-error: aborting due to 4 previous errors
+error: aborting due to 5 previous errors

[oli-obk]

Weird that it's only happening in CI... the full stderr doesn't really fit the diff you posted either. If it's a recurring thing, open an issue

[Alexendoo]

Oops, I was looking at the wrong run, that one is for an earlier commit. False alarm

[bors]

☔ The latest upstream changes (presumably #13471) made this pull request unmergeable. Please resolve the merge conflicts.