make invalid_value lint a bit smarter around enums#102281
make invalid_value lint a bit smarter around enums#102281bors merged 2 commits intorust-lang:masterfrom
Conversation
rustbot
added
the
T-compiler
|
r? @cjgillot (rust-highfive has picked a reviewer for you, use r? to override) |
rust-highfive
added
the
S-waiting-on-review
inquisitivecrystal
added
the
A-lints
RalfJung
force-pushed
the
invalid-enums
branch
from
3a8b164 to
c19daa4
Compare
compiler/rustc_lint/src/builtin.rs
Outdated
| let existing_variants = adt.variants().iter().filter(|v| v.fields.is_empty()).count(); | ||
| existing_variants > 1 | ||
| /// Determines whether the given type is inhabited. `None` means that we don't know. | ||
| fn ty_inhabited(ty: Ty<'_>) -> Option<bool> { |
There was a problem hiding this comment.
Why not use tcx.type_uninhabited_from?
There was a problem hiding this comment.
Good question. Unfortunately the documentation is not helpful:
Computes the set of modules from which this type is visibly uninhabited. To check whether a type is uninhabited at all (not just from a given module), you could check whether the forest is empty.
Now I still don't know if this is an underapproximation or an overapproximation -- is this "definitely uninhabited" or "maybe uninhabited"? No idea which forest it is talking about.^^ 🌲
There was a problem hiding this comment.
type_uninhabited_from means definitely uninhabited. References and arrays of unknown length are considered inhabited by this query.
The "forest" is the query's return type, I agree this is esoteric. tcx.type_uninhabited_from(param_env.and(ty)).is_empty() means that ty is possibly inhabited (opposite of definitely uninhabited).
There was a problem hiding this comment.
Looks like I can use is_ty_uninhabited_from as a nice wrapper.
There was a problem hiding this comment.
and no then I need a module to check 'from'...
RalfJung
force-pushed
the
invalid-enums
branch
from
2e5628e to
556e015
Compare
RalfJung
force-pushed
the
invalid-enums
branch
from
556e015 to
67fd09d
Compare
|
@bors r+ |
bors
added
S-waiting-on-bors
…iaskrgr Rollup of 9 pull requests Successful merges: - rust-lang#101555 (Stabilize `#![feature(mixed_integer_ops)]`) - rust-lang#102253 (rustdoc: use CSS containment to speed up render) - rust-lang#102281 (make invalid_value lint a bit smarter around enums) - rust-lang#102284 (Structured suggestion for missing `mut`/`const` in raw pointer) - rust-lang#102330 (rustdoc: remove no-op CSS `.srclink { font-weight; font-size }`) - rust-lang#102337 (Avoid LLVM-deprecated `Optional::hasValue`) - rust-lang#102356 (session: remove now-unnecessary lint `#[allow]`s) - rust-lang#102367 (rustdoc: remove redundant `#help-button` CSS) - rust-lang#102369 (Fix search result colors) Failed merges: r? `@ghost` `@rustbot` modify labels: rollup
6 participants
Fixes #102043