Python 3.14: bump fractions stubs #14215
Conversation
This comment has been minimized.
This comment has been minimized.
|
Thank you! I'll just add that a PR #14070 has recently arrived, but the author decided to close it, so I think it will be possible to accept your PR |
| if sys.version_info >= (3, 14): | ||
| assert_type(pow(Fraction(3, 4), 2, None), Fraction) # pyright: ignore[reportAssertTypeFailure] | ||
| assert_type(pow(Fraction(3, 4), 2, "Non-none value"), Any) # type: ignore[misc] | ||
| assert_type(pow(Fraction(3, 4), 2, True), Any) # type: ignore[misc] |
There was a problem hiding this comment.
Why does pyright report an error for the first test? And why doesn't it for the second? There should be at least a comment explaining why pyright doesn't work correctly.
The second and third tests basically test the same. It would also make more sense to use a "more expected" type like an integer. That a string would cause problems is kind of a given. Also, the assert_type in tests that we expect to fail doesn't make much sense.
We should also keep the old test for Python < 3.14:
| if sys.version_info >= (3, 14): | |
| assert_type(pow(Fraction(3, 4), 2, None), Fraction) # pyright: ignore[reportAssertTypeFailure] | |
| assert_type(pow(Fraction(3, 4), 2, "Non-none value"), Any) # type: ignore[misc] | |
| assert_type(pow(Fraction(3, 4), 2, True), Any) # type: ignore[misc] | |
| if sys.version_info >= (3, 14): | |
| assert_type(pow(Fraction(3, 4), 2, None), Fraction) # pyright: ignore[reportAssertTypeFailure] | |
| pow(Fraction(3, 4), 2, 1) # type: ignore[misc] | |
| else: | |
| pow(Fraction(), 5, 8) # type: ignore |
There was a problem hiding this comment.
The second and third tests basically test the same. It would also make more sense to use a "more expected" type like an integer. That a string would cause problems is kind of a given. Also, the assert_type in tests that we expect to fail doesn't make much sense.
We should also keep the old test for Python < 3.14:
Got it, thx! Regarding the pyright-specific ignore - it was already described it this test-file, its due to the more complex inferred type, rather than Fraction, so I just moved that comment below.
|
According to mypy_primer, this change has no effect on the checked open source code. 🤖🎉 |
Bump stubs for
fractions.Fractions(py314). According to this comment, it will be relevant just to putmoduloas None for__pow__and__rpow__.