Move autoderef to rustc_hir_analysis
#106170
Conversation
|
r? @davidtwco (rustbot has picked a reviewer for you, use r? to override) |
rustbot
added
A-translation
|
cc @davidtwco, @compiler-errors, @JohnTitor, @estebank, @TaKO8Ki |
|
r? @lcnr might have some thoughts about this -- leaving this PR for them when they return from holiday, though this is a very low-pri pr. |
There was a problem hiding this comment.
r=me with one nit, does this conflict with #106171 (comment)?
| @@ -690,6 +690,7 @@ impl<'tcx> InferCtxt<'tcx> { | |||
| typeck_results: None, | |||
| fallback_has_occurred: false, | |||
| normalize_fn_sig: Box::new(|fn_sig| fn_sig), | |||
| autoderef_steps: Box::new(|ty| vec![(ty, vec![])]), | |||
There was a problem hiding this comment.
my hope was that we only rely on autoderef_steps in errors during typeck. I guess we also autoderef when checking method receivers?
If so, don't mind this.
There was a problem hiding this comment.
So right now the current users of autoderef_steps are obligation cause codes that only come from typeck, but there's nothing stopping someone adding a check that uses this during other trait error reporting code, for example, so I'd rather not ICE here.
I'll rename it to autoderef_steps_for_diagnostic so people know it's best effort...
There was a problem hiding this comment.
r=me after adding debug_assert
compiler-errors
force-pushed
the
autoderef-to-analysis
branch
from
daadade to
c8334ce
Compare
|
Addressed nit @bors r=lcnr |
bors
added
S-waiting-on-bors
…sis, r=lcnr Move autoderef to `rustc_hir_analysis` Not sure if this is a change we actually want, but autoderef really is only (functionally) used by `rustc_hir_analysis` and `rustc_hir_typeck`, so it probably should live there. Instead, implement a separate autoderef helper in `TypeErrCtxt` for the one use-case that goes against the ordering of the crate graph..
…mpiler-errors Rollup of 8 pull requests Successful merges: - rust-lang#103236 (doc: rewrite doc for signed int::{carrying_add,borrowing_sub}) - rust-lang#103800 (Stabilize `::{core,std}::pin::pin!`) - rust-lang#106097 (Migrate mir_build diagnostics 2 of 3) - rust-lang#106170 (Move autoderef to `rustc_hir_analysis`) - rust-lang#106323 (Stabilize f16c_target_feature) - rust-lang#106360 (Tweak E0277 `&`-removal suggestions) - rust-lang#106524 (Label `struct/enum constructor` instead of `fn item`, mention that it should be called on type mismatch) - rust-lang#106739 (Remove `<dyn AstConv<'tcx>>::fun(c, ...)` calls in favour of `c.astconv().fun(...)`) Failed merges: r? `@ghost` `@rustbot` modify labels: rollup
Not sure if this is a change we actually want, but autoderef really is only (functionally) used by
rustc_hir_analysisandrustc_hir_typeck, so it probably should live there.Instead, implement a separate autoderef helper in
TypeErrCtxtfor the one use-case that goes against the ordering of the crate graph..