Don't suggest impl Trait in path position
#113310
Conversation
|
r? @TaKO8Ki (rustbot has picked a reviewer for you, use r? to override) |
rustbot
added
S-waiting-on-review
|
Some changes occurred in need_type_info.rs cc @lcnr |
| let ty_var_is_impl_trait = move |ty_vid| { | ||
| let mut infcx_inner = self.infcx.inner.borrow_mut(); | ||
| let ty_vars = infcx_inner.type_variables(); | ||
| let var_origin = ty_vars.var_origin(ty_vid); | ||
| if let TypeVariableOriginKind::TypeParameterDefinition(name, _) = var_origin.kind | ||
| && name.as_str().starts_with("impl ") | ||
| { | ||
| true | ||
| } else { | ||
| false | ||
| } | ||
| }; |
There was a problem hiding this comment.
Is there a less hacky way to do this check?
There was a problem hiding this comment.
I think it's better to change fn fmt_printer to not use the name ofTypeVariableOriginKind::TypeParameterDefinition if the DefId is from an argument position impl trait instead.
There was a problem hiding this comment.
I think reaching 11 indentation levels might be justification to refactor some of this to a function anyway 🙂
There was a problem hiding this comment.
I think it's better to change
fn fmt_printerto not use the name ofTypeVariableOriginKind::TypeParameterDefinitionif theDefIdis from an argument position impl trait instead.
How do you check if a DefId is from an arg position impl Trait?
There was a problem hiding this comment.
tcx.generics_of(tcx.parent(def_id)), look at the https://doc.rust-lang.org/nightly/nightly-rustc/rustc_middle/ty/generics/struct.GenericParamDef.html for whether the type param is synthetic.
I am not 10% sure whether the parent is necessary 🤷 try it without, if it ices, add tcx.parent
There was a problem hiding this comment.
Thanks, I think that works, but I'm not sure if it's 100% correct.
jieyouxu
force-pushed
the
dont-suggest-impl-trait-in-paths
branch
from
78874c7 to
aaf1a13
Compare
…iler-errors `TypeParameterDefinition` always require a `DefId` the `None` case never actually reaches diagnostics so it feels better for diagnostics to be able to rely on the `DefId` being there, cc rust-lang#113310
…iler-errors `TypeParameterDefinition` always require a `DefId` the `None` case never actually reaches diagnostics so it feels better for diagnostics to be able to rely on the `DefId` being there, cc rust-lang#113310
jieyouxu
force-pushed
the
dont-suggest-impl-trait-in-paths
branch
from
aaf1a13 to
fafc689
Compare
compiler/rustc_infer/src/infer/error_reporting/need_type_info.rs
Outdated
Show resolved
Hide resolved
jieyouxu
force-pushed
the
dont-suggest-impl-trait-in-paths
branch
from
fafc689 to
e6b9a55
Compare
| if let Some(idx) = generics.param_def_id_to_index(infcx.tcx, def_id) | ||
| && let generic_param_def = generics.param_at(idx as usize, infcx.tcx) |
There was a problem hiding this comment.
please unwrap instead of using if let. Not getting Some here indicates a bug
jieyouxu
force-pushed
the
dont-suggest-impl-trait-in-paths
branch
from
e6b9a55 to
b5208b3
Compare
|
@bors r+ rollup |
bors
added
S-waiting-on-bors
…iaskrgr Rollup of 4 pull requests Successful merges: - rust-lang#112717 (Implement a few more rvalue translation to smir) - rust-lang#113310 (Don't suggest `impl Trait` in path position) - rust-lang#113497 (Support explicit 32-bit MIPS ABI for the synthetic object) - rust-lang#113560 (Lint against misplaced where-clauses on associated types in traits) r? `@ghost` `@rustbot` modify labels: rollup
Fixes #113264.