-
Notifications
You must be signed in to change notification settings - Fork 13.4k
Shorten Span of unused macro lints #90761
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
Conversation
69bfa8e
to
3f9b7d2
Compare
#![allow(dead_code)] | ||
#![allow(dead_code, unused_macros)] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This one is very insteresting.
rust/src/test/ui/proc-macro/auxiliary/issue-39889.rs
Lines 10 to 17 in 493ea77
pub fn f(_input: TokenStream) -> TokenStream { | |
let rules = r#" | |
macro_rules! id { | |
($($tt:tt)*) => { $($tt)* }; | |
} | |
"#; | |
rules.parse().unwrap() | |
} |
Any idea why the unused macro wasn't linted before?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good question, no idea. My best guess is that now we have a span for id
, but we didn't for the whole macro. I think changing the test is reasonable. The issue is about the macro expansion machinery, so it's reasonable to allow
it.
#![allow(dead_code)] | ||
#![allow(dead_code, unused_macros)] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good question, no idea. My best guess is that now we have a span for id
, but we didn't for the whole macro. I think changing the test is reasonable. The issue is about the macro expansion machinery, so it's reasonable to allow
it.
I have one stylistic nitpick, I'm r+ing, but if you want to fix it gets picked up, go ahead. @bors r+ |
📌 Commit 3f9b7d2113e50e062a7464ad861d933290e5fda0 has been approved by |
This comment has been minimized.
This comment has been minimized.
8ea10b7
to
9f6ca74
Compare
The span has been recuded to the actual ident, instead of linting the *whole* macro.
@bors r=estebank |
📌 Commit 9f6ca74 has been approved by |
Shorten Span of unused macro lints The span has been reduced to the actual ident of the macro, instead of linting the *whole* macro. Closes rust-lang#90745 r? `@estebank`
⌛ Testing commit 9f6ca74 with merge 5926b64a90e7d28dacce5b89ec935eb73656db23... |
💔 Test failed - checks-actions |
The job Click to see the possible cause of the failure (guessed by this bot)
|
That looks spurious, nothing I can do?! |
@bors retry |
⌛ Testing commit 9f6ca74 with merge 6cc0c20eda28e8c057542cb97e0043e066fcef74... |
💔 Test failed - checks-actions |
The job Click to see the possible cause of the failure (guessed by this bot)
|
@bors retry |
Shorten Span of unused macro lints The span has been reduced to the actual ident of the macro, instead of linting the *whole* macro. Closes rust-lang#90745 r? `@estebank`
…askrgr Rollup of 7 pull requests Successful merges: - rust-lang#90589 (rustc_llvm: update PassWrapper for recent LLVM) - rust-lang#90644 (Extend the const swap feature) - rust-lang#90704 (Unix ExitStatus comments and a tiny docs fix) - rust-lang#90761 (Shorten Span of unused macro lints) - rust-lang#90795 (Add more comments to explain the code to generate the search index) - rust-lang#90798 (Document `unreachable!` custom panic message) - rust-lang#90826 (rustc_feature: Convert `BuiltinAttribute` from tuple to a struct) Failed merges: r? `@ghost` `@rustbot` modify labels: rollup
…span, r=cjgillot Reduce span of let else irrefutable_let_patterns warning Huge spans aren't good for IDE users as they underline constructs that are possibly multiline. Similar PR to rust-lang#90761 which did the same for the `unused_macros` lint.
Shrink `missing_{safety,errors,panics}_doc` spans Shrink the spans of `clippy::missing_*_doc` to match `missing_docs` and don't cover the entire item anymore. This helps readability in IDEs by not coloring the entire screen yellow, similar to rust-lang/rust#103749 and rust-lang/rust#90761. before:  after:  changelog: [`missing_safety_doc`], [`missing_error_doc`], [`missing_panics_doc`]: These lints no longer span the entire item.
The span has been reduced to the actual ident of the macro, instead of linting the
whole macro.
Closes #90745
r? @estebank