TYP: Incorrect type hint from histogram2d #30525
Description
Describe the issue:
Get a 2-D array as output of histogram2d, then call .tolist() on it. The resulting type is "list[float]" rather than "list[list[float]]" as it should be.
With the example code, running it (without mypy/pyright) shows the run-time python types:
Numpy Version: 2.4.0
type(a)=<class 'numpy.ndarray'>
type(a[0])=<class 'numpy.ndarray'>
type(a[0][0])=<class 'numpy.float64'>
type(al)=<class 'list'>
type(al[0])=<class 'list'>
type(al[0][0])=<class 'float'>
However mypy shows a revealed type of al as a one-dimensional list of floats:
tests/minimal_numpy.py:18: note: Revealed type is "builtins.list[builtins.float]"
Likewise, pyright shows:
tests/minimal_numpy.py:18:17 - information: Type of "al" is "list[float]"
The issue is not seen with Numpy 2.3.4, where the relevant types are "Any":
tests/minimal_numpy.py:12:17 - information: Type of "a" is "ndarray[tuple[Any, ...], dtype[float64]]"
tests/minimal_numpy.py:13:17 - information: Type of "a[0]" is "Any"
tests/minimal_numpy.py:20:17 - information: Type of "al" is "Any"
tests/minimal_numpy.py:21:17 - information: Type of "al[0]" is "Any"
Thanks for your work on NumPy!
Reproduce the code example:
import numpy as np
import typing
a, _, _ = np.histogram2d([1, 2], [3, 4])
print(f"Numpy Version: {np.__version__}")
if typing.TYPE_CHECKING:
reveal_type(a)
reveal_type(a[0])
else:
print(f"{type(a)=}")
print(f"{type(a[0])=}")
print(f"{type(a[0][0])=}")
al = a.tolist()
if typing.TYPE_CHECKING:
reveal_type(al)
reveal_type(al[0])
else:
print(f"{type(al)=}")
print(f"{type(al[0])=}")
print(f"{type(al[0][0])=}")Error message:
Python and NumPy Versions:
2.4.0
3.13.7 (main, Sep 2 2025, 14:05:52) [Clang 20.1.4 ]
Type-checker version and settings:
mypy 1.19.0 (compiled: yes)
pyright 1.1.407
uv run --with pyright,numpy python -m pyright tests/minimal_numpy.py
Additional typing packages.
No response