Skip to content

[ty] Conservative narrowing places optimization#22734

Merged
sharkdp merged 5 commits intoastral-sh:mainfrom
ericmarkmartin:synthetic-narrow-definitions
Feb 5, 2026
Merged

[ty] Conservative narrowing places optimization#22734
sharkdp merged 5 commits intoastral-sh:mainfrom
ericmarkmartin:synthetic-narrow-definitions

Conversation

@ericmarkmartin
Copy link
Contributor

@ericmarkmartin ericmarkmartin commented Jan 19, 2026

Summary

Add helpers to compute a conservative upper bound on the set of places a predicate might actually narrow, and use this to save work when computing narrowing constraints.

Test Plan

Existing tests pass

@AlexWaygood AlexWaygood added the ty Multi-file analysis & type inference label Jan 19, 2026
@ericmarkmartin ericmarkmartin force-pushed the synthetic-narrow-definitions branch from e7da68d to 2136d82 Compare January 19, 2026 20:40
@ericmarkmartin ericmarkmartin force-pushed the synthetic-narrow-definitions branch from 2136d82 to 143779d Compare January 19, 2026 20:46
@astral-sh-bot
Copy link

astral-sh-bot bot commented Jan 19, 2026

Typing conformance results

No changes detected ✅

@astral-sh-bot
Copy link

astral-sh-bot bot commented Jan 19, 2026

mypy_primer results

Changes were detected when running on open source projects
prefect (https://github.com/PrefectHQ/prefect)
+ src/integrations/prefect-docker/tests/test_containers.py:27:47: error[invalid-argument-type] Argument is incorrect: Expected `str`, found `str | bool | dict[Unknown | str, Unknown | int] | None`
+ src/integrations/prefect-docker/tests/test_containers.py:27:47: error[invalid-argument-type] Argument is incorrect: Expected `str | list[str] | None`, found `str | bool | dict[Unknown | str, Unknown | int] | None`
+ src/integrations/prefect-docker/tests/test_containers.py:27:47: error[invalid-argument-type] Argument is incorrect: Expected `str | None`, found `str | bool | dict[Unknown | str, Unknown | int] | None`
+ src/integrations/prefect-docker/tests/test_containers.py:27:47: error[invalid-argument-type] Argument is incorrect: Expected `bool | None`, found `str | bool | dict[Unknown | str, Unknown | int] | None`
+ src/integrations/prefect-docker/tests/test_containers.py:27:47: error[invalid-argument-type] Argument is incorrect: Expected `str | list[str] | None`, found `str | bool | dict[Unknown | str, Unknown | int] | None`
+ src/integrations/prefect-docker/tests/test_containers.py:27:47: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, str] | list[str] | None`, found `str | bool | dict[Unknown | str, Unknown | int] | None`
+ src/integrations/prefect-docker/tests/test_containers.py:27:47: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str | bool | dict[Unknown | str, Unknown | int] | None`
+ src/integrations/prefect-docker/tests/test_containers.py:42:47: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str`
+ src/integrations/prefect-docker/tests/test_containers.py:55:47: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str`
+ src/integrations/prefect-docker/tests/test_containers.py:68:47: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str`
+ src/integrations/prefect-docker/tests/test_containers.py:81:47: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str`
+ src/integrations/prefect-docker/tests/test_images.py:16:44: error[invalid-argument-type] Argument is incorrect: Expected `str`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:16:44: error[invalid-argument-type] Argument is incorrect: Expected `str | None`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:16:44: error[invalid-argument-type] Argument is incorrect: Expected `str | None`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:16:44: error[invalid-argument-type] Argument is incorrect: Expected `bool`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:16:44: error[invalid-argument-type] Argument is incorrect: Expected `DockerHost | None`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:16:44: error[invalid-argument-type] Argument is incorrect: Expected `DockerRegistryCredentials | None`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:16:44: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:21:16: warning[possibly-missing-attribute] Attribute `id` may be missing on object of type `Unknown | list[Unknown]`
+ src/integrations/prefect-docker/tests/test_images.py:29:47: error[invalid-argument-type] Argument is incorrect: Expected `bool`, found `str`
+ src/integrations/prefect-docker/tests/test_images.py:29:47: error[invalid-argument-type] Argument is incorrect: Expected `DockerRegistryCredentials | None`, found `str`
+ src/integrations/prefect-docker/tests/test_images.py:29:47: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str`
+ src/integrations/prefect-docker/tests/test_images.py:31:16: warning[possibly-missing-attribute] Attribute `id` may be missing on object of type `Unknown | list[Unknown]`
+ src/integrations/prefect-docker/tests/test_images.py:51:17: error[invalid-argument-type] Argument is incorrect: Expected `bool`, found `str`
+ src/integrations/prefect-docker/tests/test_images.py:51:17: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str`
+ src/integrations/prefect-docker/tests/test_images.py:53:16: warning[possibly-missing-attribute] Attribute `id` may be missing on object of type `Unknown | list[Unknown]`
+ src/integrations/prefect-docker/tests/test_images.py:64:47: error[invalid-argument-type] Argument is incorrect: Expected `str`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:64:47: error[invalid-argument-type] Argument is incorrect: Expected `str | None`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:64:47: error[invalid-argument-type] Argument is incorrect: Expected `str | None`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:64:47: error[invalid-argument-type] Argument is incorrect: Expected `bool`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:64:47: error[invalid-argument-type] Argument is incorrect: Expected `DockerRegistryCredentials | None`, found `str | bool`
+ src/integrations/prefect-docker/tests/test_images.py:64:47: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str | bool`
+ src/integrations/prefect-kubernetes/prefect_kubernetes/jobs.py:428:17: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `str`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:20:13: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `None`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:29:13: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `None`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:38:13: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `None`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:57:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:103:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:149:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:195:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:240:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:286:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_custom_objects.py:344:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_deployments.py:18:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_deployments.py:38:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_deployments.py:70:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_deployments.py:92:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_deployments.py:113:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_deployments.py:141:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_jobs.py:36:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_jobs.py:52:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_jobs.py:68:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_jobs.py:87:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_jobs.py:107:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_jobs.py:131:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_jobs.py:159:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_pods.py:29:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_pods.py:46:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_pods.py:78:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_pods.py:96:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_pods.py:115:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_pods.py:137:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
+ src/integrations/prefect-kubernetes/tests/test_pods.py:167:9: error[invalid-argument-type] Argument is incorrect: Expected `dict[str, Any]`, found `Literal["test"]`
- src/prefect/cache_policies.py:311:25: warning[possibly-missing-attribute] Attribute `__code__` may be missing on object of type `Unknown | (((...) -> Any) & ((*args: object, **kwargs: object) -> object))`
+ src/prefect/cache_policies.py:311:25: warning[possibly-missing-attribute] Attribute `__code__` may be missing on object of type `Unknown | ((...) -> Any)`
- src/prefect/input/run_input.py:672:20: error[invalid-return-type] Return type does not match returned value: expected `T@GetAutomaticInputHandler | AutomaticRunInput[T@GetAutomaticInputHandler]`, found `T@GetAutomaticInputHandler | AutomaticRunInput[T@GetAutomaticInputHandler] | Coroutine[Any, Any, T@GetAutomaticInputHandler | AutomaticRunInput[T@GetAutomaticInputHandler]]`
+ src/prefect/input/run_input.py:672:20: error[invalid-return-type] Return type does not match returned value: expected `T@GetAutomaticInputHandler | AutomaticRunInput[T@GetAutomaticInputHandler]`, found `Unknown | Coroutine[Any, Any, Unknown]`
+ src/prefect/task_engine.py:1641:28: error[invalid-await] `Unknown | R@AsyncTaskRunEngine | Coroutine[Any, Any, R@AsyncTaskRunEngine]` is not awaitable
+ src/prefect/task_engine.py:1749:47: error[invalid-argument-type] Argument to function `next` is incorrect: Expected `SupportsNext[Unknown]`, found `Unknown | R@run_generator_task_sync`
+ src/prefect/task_engine.py:1762:25: warning[possibly-missing-attribute] Attribute `throw` may be missing on object of type `Unknown | R@run_generator_task_sync`
+ src/prefect/task_engine.py:1808:48: warning[possibly-missing-attribute] Attribute `__anext__` may be missing on object of type `Unknown | R@run_generator_task_async`
+ src/prefect/task_engine.py:1820:29: warning[possibly-missing-attribute] Attribute `throw` may be missing on object of type `Unknown | R@run_generator_task_async`
- src/prefect/tasks.py:185:9: warning[possibly-missing-attribute] Attribute `__code__` may be missing on object of type `Unknown | (((...) -> Any) & ((*args: object, **kwargs: object) -> object))`
+ src/prefect/tasks.py:185:9: warning[possibly-missing-attribute] Attribute `__code__` may be missing on object of type `Unknown | ((...) -> Any)`
- Found 5393 diagnostics
+ Found 5461 diagnostics

scikit-build-core (https://github.com/scikit-build/scikit-build-core)
+ src/scikit_build_core/build/wheel.py:99:20: error[no-matching-overload] No overload of bound method `__init__` matches arguments
- Found 49 diagnostics
+ Found 50 diagnostics

pandas-stubs (https://github.com/pandas-dev/pandas-stubs)
- tests/test_pandas.py:511:18: error[assert-type-unspellable-subtype] Type `bool` does not match asserted type `TypeIs[timedelta64[timedelta & ~NaTType & ~NAType] @ np_td]`
+ tests/test_pandas.py:511:18: error[assert-type-unspellable-subtype] Type `bool` does not match asserted type `TypeIs[timedelta64[timedelta] @ np_td]`
- tests/test_pandas.py:516:22: error[assert-type-unspellable-subtype] Type `bool` does not match asserted type `TypeIs[str | bytes | date | ... omitted 9 union elements @ np_nat]`
+ tests/test_pandas.py:516:22: error[assert-type-unspellable-subtype] Type `bool` does not match asserted type `TypeIs[timedelta64[timedelta | int | None] @ np_nat]`

Memory usage changes were detected when running on open source projects
flake8 (https://github.com/pycqa/flake8)
- TOTAL MEMORY USAGE: ~66MB
+ TOTAL MEMORY USAGE: ~63MB

trio (https://github.com/python-trio/trio)
- TOTAL MEMORY USAGE: ~167MB
+ TOTAL MEMORY USAGE: ~159MB
-     struct metadata = ~11MB
+     struct metadata = ~10MB
-     memo fields = ~108MB
+     memo fields = ~103MB

sphinx (https://github.com/sphinx-doc/sphinx)
- TOTAL MEMORY USAGE: ~301MB
+ TOTAL MEMORY USAGE: ~287MB
-     memo fields = ~176MB
+     memo fields = ~167MB

prefect (https://github.com/PrefectHQ/prefect)
-     memo fields = ~424MB
+     memo fields = ~403MB

@ericmarkmartin ericmarkmartin changed the title Conservative narrowing places optimization [ty] Conservative narrowing places optimization Jan 19, 2026
@AlexWaygood
Copy link
Member

AlexWaygood commented Jan 26, 2026

Codspeed's reporting speedups of up to 4%, and this also looks like a pretty big win in terms of memory usage according to the primer report!

@AlexWaygood AlexWaygood added the performance Potential performance improvement label Jan 26, 2026
@ericmarkmartin
Copy link
Contributor Author

ericmarkmartin commented Jan 27, 2026

looks like a pretty big win in terms of memory usage according to the primer report!

Oh huh that's neat! The only cautionary note I have about this PR is that there's no particularly good mechanism in place for keeping the conservative places in line with the actually inferred places. Of course if we started narrowing a place that's not in the conservative place set then ty won't actually produce the narrowing, so that should be relatively catchable at test-time

@carljm
Copy link
Contributor

carljm commented Jan 30, 2026

@ericmarkmartin is anything lacking to put this out of draft and up for review?

@ericmarkmartin ericmarkmartin marked this pull request as ready for review February 5, 2026 04:26
@ericmarkmartin
Copy link
Contributor Author

@ericmarkmartin is anything lacking to put this out of draft and up for review?

No, not really, just still wasn't sure if we wanted this without the fix for #690

Copy link
Contributor

@sharkdp sharkdp left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is a really good optimization idea and a great implementation — thank you very much!

@codspeed-hq
Copy link

codspeed-hq bot commented Feb 5, 2026

CodSpeed Performance Report

Merging this PR will improve performance by 4.18%

Comparing ericmarkmartin:synthetic-narrow-definitions (467aef4) with main (789399c)

Summary

⚡ 1 improved benchmark
✅ 23 untouched benchmarks
⏩ 30 skipped benchmarks1

Performance Changes

Mode Benchmark BASE HEAD Efficiency
Simulation DateType 241.5 ms 231.8 ms +4.18%

Footnotes

  1. 30 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports.

@sharkdp sharkdp merged commit c8703cb into astral-sh:main Feb 5, 2026
49 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

performance Potential performance improvement ty Multi-file analysis & type inference

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants