ruff/crates
Carl Meyer 8248193ed9
[ty] defer inference of legacy TypeVar bound/constraints/defaults (#20598)
## Summary

This allows us to handle self-referential bounds/constraints/defaults
without panicking.

Handles more cases from https://github.com/astral-sh/ty/issues/256

This also changes the way we infer the types of legacy TypeVars. Rather
than understanding a constructor call to `typing[_extension].TypeVar`
inside of any (arbitrarily nested) expression, and having to use a
special `assigned_to` field of the semantic index to try to best-effort
figure out what name the typevar was assigned to, we instead understand
the creation of a legacy `TypeVar` only in the supported syntactic
position (RHS of a simple un-annotated assignment with one target). In
any other position, we just infer it as creating an opaque instance of
`typing.TypeVar`. (This behavior matches all other type checkers.)

So we now special-case TypeVar creation in `TypeInferenceBuilder`, as a
special case of an assignment definition, rather than deeper inside call
binding. This does mean we re-implement slightly more of
argument-parsing, but in practice this is minimal and easy to handle
correctly.

This is easier to implement if we also make the RHS of a simple (no
unpacking) one-target assignment statement no longer a standalone
expression. Which is fine to do, because simple one-target assignments
don't need to infer the RHS more than once. This is a bonus performance
(0-3% across various projects) and significant memory-usage win, since
most assignment statements are simple one-target assignment statements,
meaning we now create many fewer standalone-expression salsa
ingredients.

This change does mean that inference of manually-constructed
`TypeAliasType` instances can no longer find its Definition in
`assigned_to`, which regresses go-to-definition for these aliases. In a
future PR, `TypeAliasType` will receive the same treatment that
`TypeVar` did in this PR (moving its special-case inference into
`TypeInferenceBuilder` and supporting it only in the correct syntactic
position, and lazily inferring its value type to support recursion),
which will also fix the go-to-definition regression. (I decided a
temporary edge-case regression is better in this case than doubling the
size of this PR.)

This PR also tightens up and fixes various aspects of the validation of
`TypeVar` creation, as seen in the tests.

We still (for now) treat all typevars as instances of `typing.TypeVar`,
even if they were created using `typing_extensions.TypeVar`. This means
we'll wrongly error on e.g. `T.__default__` on Python 3.11, even if `T`
is a `typing_extensions.TypeVar` instance at runtime. We share this
wrong behavior with both mypy and pyrefly. It will be easier to fix
after we pull in https://github.com/python/typeshed/pull/14840.

There are some issues that showed up here with typevar identity and
`MarkTypeVarsInferable`; the fix here (using the new `original` field
and `is_identical_to` methods on `BoundTypeVarInstance` and
`TypeVarInstance`) is a bit kludgy, but it can go away when we eliminate
`MarkTypeVarsInferable`.

## Test Plan

Added and updated mdtests.

### Conformance suite impact

The impact here is all positive:

* We now correctly error on a legacy TypeVar with exactly one constraint
type given.
* We now correctly error on a legacy TypeVar with both an upper bound
and constraints specified.

### Ecosystem impact

Basically none; in the setuptools case we just issue slightly different
errors on an invalid TypeVar definition, due to the modified validation
code.

---------

Co-authored-by: Alex Waygood <Alex.Waygood@Gmail.com>
2025-10-09 21:08:37 +00:00
..
ruff Bump 0.14.0 (#20751) 2025-10-07 11:05:47 -07:00
ruff_annotate_snippets Display diffs for `ruff format --check` and add support for different output formats (#20443) 2025-09-30 12:00:51 -04:00
ruff_benchmark [ty] Improve assignability/subtyping between two protocol types (#20368) 2025-10-08 18:37:30 +00:00
ruff_cache Switch to Rust 2024 edition (#18129) 2025-05-16 13:25:28 +02:00
ruff_db [ty] Fix file root matching for `/` 2025-10-03 08:18:03 -04:00
ruff_dev [ty] Remove duplicate global lint registry (#20053) 2025-08-22 19:43:12 -04:00
ruff_diagnostics Fix rust feature activation (#20012) 2025-08-21 09:26:06 +02:00
ruff_formatter Display diffs for `ruff format --check` and add support for different output formats (#20443) 2025-09-30 12:00:51 -04:00
ruff_graph [ty] Remove `KnownModule::is_enum` (#19681) 2025-08-01 10:31:12 +02:00
ruff_index Update Rust toolchain to 1.88 and MSRV to 1.86 (#19011) 2025-06-28 20:24:00 +02:00
ruff_linter [`fastapi`] Fix false positives for path parameters that FastAPI doesn't recognize (`FAST003`) (#20687) 2025-10-09 16:10:21 -04:00
ruff_macros Replace two more uses of unsafe with const `Option::unwrap` (#20584) 2025-09-25 15:35:13 -04:00
ruff_memory_usage [ty] Clean up inherited generic contexts (#20647) 2025-10-03 13:55:43 -04:00
ruff_notebook Display diffs for `ruff format --check` and add support for different output formats (#20443) 2025-09-30 12:00:51 -04:00
ruff_options_metadata Update Rust toolchain to 1.89 (#19807) 2025-08-07 18:21:50 +02:00
ruff_python_ast [ty] Use 3.14 as the default version (#20759) 2025-10-08 11:38:47 +02:00
ruff_python_ast_integration_tests Disallow implicit concatenation of t-strings and other string types (#19485) 2025-07-27 12:41:03 +00:00
ruff_python_codegen Generator preferred quote style (#20434) 2025-09-18 12:57:21 +02:00
ruff_python_formatter Update default and latest Python versions for 3.14 (#20725) 2025-10-07 12:23:11 -04:00
ruff_python_importer [ruff] Add API for splicing into an existing import statement 2025-09-17 13:59:28 -04:00
ruff_python_index Track t-strings and f-strings for token-based rules and suppression comments (#20357) 2025-09-12 13:00:12 -05:00
ruff_python_literal Switch to Rust 2024 edition (#18129) 2025-05-16 13:25:28 +02:00
ruff_python_parser [syntax-errors]: future-feature-not-defined (F407) (#20554) 2025-09-25 13:52:24 -04:00
ruff_python_semantic [`ruff`] Extend FA102 with listed PEP 585-compatible APIs (#20659) 2025-10-03 09:45:32 -04:00
ruff_python_stdlib [`ruff`] Extend FA102 with listed PEP 585-compatible APIs (#20659) 2025-10-03 09:45:32 -04:00
ruff_python_trivia Handle t-string prefixes in `SimpleTokenizer` (#20578) 2025-09-25 14:33:37 -05:00
ruff_python_trivia_integration_tests Handle t-string prefixes in `SimpleTokenizer` (#20578) 2025-09-25 14:33:37 -05:00
ruff_server Use `Annotation::tags` instead of hardcoded rule matching in ruff server (#20565) 2025-09-26 09:06:26 +02:00
ruff_source_file Move diff rendering to `ruff_db` (#20006) 2025-08-21 09:47:00 -04:00
ruff_text_size [ruff] Add `TextRange::to_std_range` 2025-09-17 13:59:28 -04:00
ruff_wasm Bump 0.14.0 (#20751) 2025-10-07 11:05:47 -07:00
ruff_workspace Update default and latest Python versions for 3.14 (#20725) 2025-10-07 12:23:11 -04:00
ty [ty] Use 3.14 as the default version (#20759) 2025-10-08 11:38:47 +02:00
ty_combine [ty] Disallow std::env and io methods in most ty crates (#20046) 2025-08-22 11:13:47 -07:00
ty_completion_eval [ty] Add an evaluation for completions 2025-10-08 08:44:21 -04:00
ty_ide [ty] defer inference of legacy TypeVar bound/constraints/defaults (#20598) 2025-10-09 21:08:37 +00:00
ty_project [ty] Use 3.14 as the default version (#20759) 2025-10-08 11:38:47 +02:00
ty_python_semantic [ty] defer inference of legacy TypeVar bound/constraints/defaults (#20598) 2025-10-09 21:08:37 +00:00
ty_server [ty] Improve diagnostics for bad `@overload` definitions (#20745) 2025-10-07 21:52:57 +00:00
ty_static [ty] improve base conda distinction from child conda (#20675) 2025-10-03 13:56:06 +00:00
ty_test [ty] Disambiguate classes that live in different modules but have the same fully qualified names (#20756) 2025-10-08 18:27:40 +01:00
ty_vendored [ty] Sync vendored typeshed stubs (#20658) 2025-10-01 10:11:48 +02:00
ty_wasm [ruff,ty] Enable tracing's `log` feature 2025-10-03 08:18:03 -04:00