TYP: check_untyped_defs pandas.core.nanops #34689
Conversation
| @@ -130,7 +130,7 @@ def f( | |||
|
|
|||
| return result | |||
|
|
|||
| return f | |||
| return cast(F, f) | |||
There was a problem hiding this comment.
What is the cast required here for? Wouldn't the return annotation already provide this info?
There was a problem hiding this comment.
This is how we have been typing all our decorators to preserve the signature. Its recommended best practice. https://mypy.readthedocs.io/en/stable/generics.html#declaring-decorators
if we don't cast, we get a mypy error pandas\core\nanops.py:133: error: Incompatible return value type (got "Callable[..., Any]", expected "F")
pandas/core/nanops.py
Outdated
| values: np.ndarray, | ||
| axis: Optional[int] = None, | ||
| skipna: bool = True, | ||
| mask: np.ndarray = None, |
There was a problem hiding this comment.
Does this need to be typed as Optional[np.ndarray], since the default is None?
There was a problem hiding this comment.
yep. good catch. np.ndarray evaluates to Any, so None is compatible. will update.
pandas/core/nanops.py
Outdated
| @@ -557,7 +562,8 @@ def nanmean(values, axis=None, skipna=True, mask=None): | |||
| count = _get_counts(values.shape, mask, axis, dtype=dtype_count) | |||
| the_sum = _ensure_numeric(values.sum(axis, dtype=dtype_sum)) | |||
|
|
|||
| if axis is not None and getattr(the_sum, "ndim", False): | |||
| if not is_scalar(count): | |||
There was a problem hiding this comment.
Those two if statements don't seem equivalent on first sight?
(they might be in practice though)
There was a problem hiding this comment.
IIUC, if count is a scalar, ct_mask = count == 0 would evaluate to a bool and then ct_mask.any() would raise an AttributeError.
pandas\core\nanops.py:571: error: Item "bool" of "Union[bool, Any]" has no attribute "any"
|
@WillAyd @jorisvandenbossche updated + green |
|
Lgtm @jorisvandenbossche merge if nothing outstanding |
|
thanks @simonjayhawkins |
pandas\core\nanops.py:565: error: Item "bool" of "Union[bool, Any]" has no attribute "any"