[red-knot] Improve type inference for except handlers where a tuple of exception classes is caught #13646
Conversation
This file contains 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
AlexWaygood
commented
Oct 6, 2024
Comment on lines
+6197
to
+6203
| // TODO: these `__class_getitem__` diagnostics are all false positives: | ||
| // (`builtins.type` is unique at runtime | ||
| // as it can be subscripted even though it has no `__class_getitem__` method) | ||
| "Cannot subscript object of type `Literal[type]` with no `__class_getitem__` method", | ||
| "Cannot subscript object of type `Literal[type]` with no `__class_getitem__` method", | ||
| "Cannot subscript object of type `Literal[type]` with no `__class_getitem__` method", | ||
| "Cannot subscript object of type `Literal[type]` with no `__class_getitem__` method", |
There was a problem hiding this comment.
Cc. @charliermarsh -- this is a little edge case in the logic you added in #13579. I think the special-casing that allows e.g. type[int] at runtime is buried deep in the interpreter somewhere. According to all the normal rules, it shouldn't be subscriptable... but it is!
Python 3.12.4 (main, Jun 12 2024, 09:54:41) [Clang 15.0.0 (clang-1500.3.9.4)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> type[int]
type[int]
>>> type.__getitem__
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
AttributeError: type object 'type' has no attribute '__getitem__'. Did you mean: '__getstate__'?
>>> type.__class_getitem__
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
AttributeError: type object 'type' has no attribute '__class_getitem__'
>>> type.__dict__.keys()
dict_keys(['__new__', '__repr__', '__call__', '__getattribute__', '__setattr__', '__delattr__', '__init__', '__or__', '__ror__', 'mro', '__subclasses__', '__prepare__', '__instancecheck__', '__subclasscheck__', '__dir__', '__sizeof__', '__basicsize__', '__itemsize__', '__flags__', '__weakrefoffset__', '__base__', '__dictoffset__', '__name__', '__qualname__', '__bases__', '__mro__', '__module__', '__abstractmethods__', '__dict__', '__doc__', '__text_signature__', '__annotations__', '__type_params__'])
There was a problem hiding this comment.
Yeah this is literally hardcoded directly into PyObject_GetItem as a special case 🫣
There was a problem hiding this comment.
🙃
I'm working on a PR now to add the special case to our semantic model...
|
carljm
approved these changes
Oct 7, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Summary
There's still a bunch of stuff that is valid in an exception handler but that we can't do type inference for until we have generalised support for
type[T]. But this improves our type inference for some common cases involving exception tuples. It seems like a nice little short-term winTest plan
I fixed the TODO in the existing test
except_handler_multiple_exceptions, and added a new test with some new TODOs to illustrate where our type inference still falls short.