Add new UNSTABLE_INTRINSICS_WITH_STABLE_WRAPPER lint
#12226
Conversation
rustbot
added
the
S-waiting-on-review
|
Interestingly enough, we cannot differentiate between the intrinsics item and the "reexported" one (like for |
GuillaumeGomez
force-pushed
the
UNSTABLE_INTRINSICS_WITH_STABLE_WRAPPER
branch
5 times, most recently
from
98e64e3
to
5808e25
Compare
|
Took me a while to realize that the error was coming from the doc example... ^^' PR is now ready! |
GuillaumeGomez
force-pushed
the
UNSTABLE_INTRINSICS_WITH_STABLE_WRAPPER
branch
from
5808e25
to
f241339
Compare
| fn_name: &str, | ||
| segments: &[PathSegment<'_>], | ||
| ) { | ||
| static FNS: OnceLock<Vec<DefId>> = OnceLock::new(); |
There was a problem hiding this comment.
I don't think having a Vec will be a problem here, we don't have that many intrinsics anyway.
There was a problem hiding this comment.
I guess. A benchmark is always handy, I just think it's cleaner at least and more greppable
| Res::Def(DefKind::Fn, fn_def_id) => Some(fn_def_id), | ||
| _ => None, | ||
| }) | ||
| .filter(|fn_def_id| !get_doc(cx.tcx, *fn_def_id).contains("does not have a stable counterpart")) |
There was a problem hiding this comment.
Some intrinsics don't have this. For example is_val_statically_known or const_eval_select
There was a problem hiding this comment.
It's probably best to at least check for "the stabilized version of this intrinsic is" after this too
|
☔ The latest upstream changes (presumably #12306) made this pull request unmergeable. Please resolve the merge conflicts. |
| @@ -1,5 +1,5 @@ | |||
| error: transmute from a reference to a pointer | |||
There was a problem hiding this comment.
I think this lint should be allowed in other tests.
|
Hey @GuillaumeGomez , this is a ping from triage. There hasn't been any activity for some time, perhaps it was forgotten? @rustbot author |
rustbot
added
S-waiting-on-author
|
I'll close it because it should likely be a rustc lint directly. |
Fixes #2997.
To prevent having to keep the list of intrinsics with stable counterpart up-to-date on clippy side, I instead check if the documentation contains the sentence
"does not have a stable counterpart". To prevent running this check more than once, I store all the non-matching intrinsicsDefIdinto astatic.r? @llogiq
changelog: Add new
UNSTABLE_INTRINSICS_WITH_STABLE_WRAPPERlint