[ty] Fix invariant generic disjointness via materialization overlap#25583
Merged
Conversation
Typing conformance resultsNo changes detected ✅Current numbersThe percentage of diagnostics emitted that were expected errors held steady at 91.94%. The percentage of expected errors that received a diagnostic held steady at 87.09%. The number of fully passing files held steady at 92/134. |
Memory usage reportSummary
Significant changesClick to expand detailed breakdowntrio
sphinx
prefect
|
|
| Lint rule | Added | Removed | Changed |
|---|---|---|---|
type-assertion-failure |
0 | 0 | 1 |
| Total | 0 | 0 | 1 |
Raw diff:
pandas-stubs (https://github.com/pandas-dev/pandas-stubs)
- tests/test_api_types.py:372:9 error[type-assertion-failure] Type `Categorical[str]` does not match asserted type `Categorical[int]`
+ tests/test_api_types.py:372:9 error[type-assertion-failure] Type `Categorical[Unknown]` does not match asserted type `Categorical[int]`
carljm
commented
Jun 3, 2026
Comment on lines
+184
to
+187
| static_assert(is_disjoint_from(Invariant[A | Any], Invariant[B])) | ||
| static_assert(is_disjoint_from(Invariant[B], Invariant[A | Any])) | ||
| static_assert(is_disjoint_from(Invariant[Intersection[A, Any]], Invariant[B])) | ||
| static_assert(is_disjoint_from(Invariant[B], Invariant[Intersection[A, Any]])) |
Contributor
Author
There was a problem hiding this comment.
The previous "must be fully static" guard did not handle these cases -- but we know these are disjoint.
| static_assert(not is_disjoint_from(Invariant[Id[U]], Invariant[int])) | ||
|
|
||
| static_assert(not is_disjoint_from(Invariant[Id[int]], Invariant[int])) | ||
| static_assert(is_disjoint_from(Invariant[Id[int]], Invariant[str])) |
Contributor
Author
There was a problem hiding this comment.
The previous "must be fully static" guard also failed this case, because it didn't properly expand the generic typevar (expand_eagerly uses raw_value_type and throws away generics).
Comment on lines
-1294
to
-1300
| && self_alias | ||
| .specialization(db) | ||
| .generic_context(db) | ||
| .variables(db) | ||
| .any(|typevar| { | ||
| matches!(typevar.variance(db), TypeVarVariance::Invariant) | ||
| }) |
Contributor
Author
There was a problem hiding this comment.
This guard isn't needed for correctness; it's redundant with the behavior of specialization disjointness, which already handles the different variances correctly.
If this is here as a perf optimization (and you've already demonstrated that it's needed to avoid regression) then we can keep it.
Comment on lines
+1676
to
+1685
| // `Bottom[L] <: Top[R]` asks whether the materialization ranges for `L` | ||
| // and `R` have any common materialization, so this is symmetric despite | ||
| // using a directional subtyping checker. | ||
| self.as_relation_checker(TypeRelation::Subtyping) | ||
| .check_subtyping_in_invariant_position( | ||
| db, | ||
| left_type, | ||
| MaterializationKind::Bottom, | ||
| right_type, | ||
| MaterializationKind::Top, |
Contributor
Author
There was a problem hiding this comment.
This is the correct handling of gradual types here, which lets us still detect disjointness in the case of gradual types that can't possibly materialize to the same type.
carljm
commented
Jun 3, 2026
| db: &'db dyn Db, | ||
| other: Self, | ||
| constraints: &ConstraintSetBuilder<'db>, | ||
| ) -> bool { |
Contributor
Author
There was a problem hiding this comment.
All these changes are just threading through a common disjointness checker so we maintain our cycle guards when looking at cyclic MRO graphs. (It's probably clearest to look at the two commits in this PR separately: the first commit is just the semantic changes we need, the second looks more invasive but it's just threading things through to fix the stack overflow, no semantic change.)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.