Use diagnostic or language items instead of paths

[Jarcho]

I think that gets everything except ones used in a list of paths to check.

changelog: none

[rust-highfive]

r? @phansch

(rust-highfive has picked a reviewer for you, use r? to override)

[phansch]

Minus the comments from @camsteffen, this looks good to me, thanks!

[phansch]

woops, misclicked 😅

[phansch]

@bors r+ thanks!

[bors]

📌 Commit 11375b5 has been approved by phansch

[bors]

⌛ Testing commit 11375b5 with merge f01d655...

[bors]

💔 Test failed - checks-action_test

[phansch]

@bors r+

[bors]

📌 Commit e4ffff9 has been approved by phansch

[bors]

⌛ Testing commit e4ffff9 with merge 5945e85...

[bors]

☀️ Test successful - checks-action_dev_test, checks-action_remark_test, checks-action_test
Approved by: phansch
Pushing 5945e85 to master...

[matthiaskrgr]

Is this expected to have a direct effect on lint results?
I'm asking because the change causes ~160 expl_impl_clone_on_copy warnings to appear in the lintcheck logs.

Most of them seem to be caused by macro usage

target/lintcheck/sources/libc-0.2.81/src/macros.rs:84:9

#[allow(unused_macros)]
macro_rules! s {
    ($($(#[$attr:meta])* pub $t:ident $i:ident { $($field:tt)* })*) => ($(
        s!(it: $(#[$attr])* pub $t $i { $($field)* });
    )*);
    (it: $(#[$attr:meta])* pub union $i:ident { $($field:tt)* }) => (
        compile_error!("unions cannot derive extra traits, use s_no_extra_traits instead");
    );
    (it: $(#[$attr:meta])* pub struct $i:ident { $($field:tt)* }) => (
        __item! {
            #[repr(C)]
            #[cfg_attr(feature = "extra_traits", derive(Debug, Eq, Hash, PartialEq))]
            #[allow(deprecated)]
            $(#[$attr])*
            pub struct $i { $($field)* }
        }
        #[allow(deprecated)]
        impl ::Copy for $i {}
        #[allow(deprecated)]
        impl ::Clone for $i { // <- HERE
            fn clone(&self) -> $i { *self }
        }
    );
}

warning: you are implementing `Clone` explicitly on a `Copy` type
    --> src/macros.rs:84:9
     |
84   | /         impl ::Clone for $i {
85   | |             fn clone(&self) -> $i { *self }
86   | |         }
     | |_________^
     |
    ::: src/unix/linux_like/linux/mod.rs:3502:1
     |
3502 |   expand_align!();
     |   ---------------- in this macro invocation
     |
note: consider deriving `Clone` or removing `Copy`
    --> src/macros.rs:84:9
     |
84   | /         impl ::Clone for $i {
85   | |             fn clone(&self) -> $i { *self }
86   | |         }
     | |_________^
     |
    ::: src/unix/linux_like/linux/mod.rs:3502:1
     |
3502 |   expand_align!();
     |   ---------------- in this macro invocation
     = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#expl_impl_clone_on_copy
     = note: this warning originates in a macro (in Nightly builds, run with -Z macro-backtrace for more info)

warning: you are implementing `Clone` explicitly on a `Copy` type
    --> src/macros.rs:120:9
     |
120  | /         impl ::Clone for $i {
121  | |             fn clone(&self) -> $i { *self }
122  | |         }
     | |_________^
     |
    ::: src/unix/linux_like/linux/mod.rs:3502:1
     |
3502 |   expand_align!();
     |   ---------------- in this macro invocation
     |
note: consider deriving `Clone` or removing `Copy`
    --> src/macros.rs:120:9
     |
120  | /         impl ::Clone for $i {
121  | |             fn clone(&self) -> $i { *self }
122  | |         }
     | |_________^
     |
    ::: src/unix/linux_like/linux/mod.rs:3502:1
     |
3502 |   expand_align!();
     |   ---------------- in this macro invocation
     = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#expl_impl_clone_on_copy
     = note: this warning originates in a macro (in Nightly builds, run with -Z macro-backtrace for more info)

warning: you are implementing `Clone` explicitly on a `Copy` type
    --> src/macros.rs:120:9
     |
120  | /         impl ::Clone for $i {
121  | |             fn clone(&self) -> $i { *self }
122  | |         }
     | |_________^
     |
    ::: src/unix/linux_like/linux/mod.rs:3502:1
     |
3502 |   expand_align!();
     |   ---------------- in this macro invocation
     |
note: consider deriving `Clone` or removing `Copy`
    --> src/macros.rs:120:9
     |
120  | /         impl ::Clone for $i {
121  | |             fn clone(&self) -> $i { *self }
122  | |         }
     | |_________^
     |
    ::: src/unix/linux_like/linux/mod.rs:3502:1
     |
3502 |   expand_align!();
     |   ---------------- in this macro invocation
     = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#expl_impl_clone_on_copy
     = note: this warning originates in a macro (in Nightly builds, run with -Z macro-backtrace for more info)

warning: you are implementing `Clone` explicitly on a `Copy` type
    --> src/macros.rs:120:9
     |
120  | /         impl ::Clone for $i {
121  | |             fn clone(&self) -> $i { *self }
122  | |         }
     | |_________^
     |
    ::: src/unix/linux_like/linux/mod.rs:3502:1
     |
3502 |   expand_align!();
     |   ---------------- in this macro invocation
     |
note: consider deriving `Clone` or removing `Copy`
    --> src/macros.rs:120:9
     |
120  | /         impl ::Clone for $i {
121  | |             fn clone(&self) -> $i { *self }
122  | |         }
     | |_________^
     |
    ::: src/unix/linux_like/linux/mod.rs:3502:1
     |
3502 |   expand_align!();
     |   ---------------- in this macro invocation
     = help: for further information visit https://rust-lang.github.io/rust-clippy/master/index.html#expl_impl_clone_on_copy
     = note: this warning originates in a macro (in Nightly builds, run with -Z macro-backtrace for more info)

[flip1995]

Maybe comparing def_ids isn't equivalent to the match_path method?

[Jarcho]

The lang item might not be defined at the point. I didn't think that would actually happen though. I'll have a pr in a bit.

[Jarcho]

Turns out this did fix expl_impl_clone_on_copy, which is causing all those warnings. #6868 (comment)

[flip1995]

Yeah, I agree. Triggering in the above example makes sense to me. I also think triggering in local macros is fine.