API: NumericIndex([1 2, 3]).dtype should be int64 on 32-bit systems #49815
Conversation
topper-123
commented
Nov 21, 2022
topper-123
commented
- closes API: Should NumericIndex([1 2, 3]).dtype be int64 or int32 on 32-bit systems? #49813
- Tests added and passed if fixing a bug or adding a new feature
- All code checks passed.
| "String dtype not supported, you may need " | ||
| "to explicitly cast to a numeric type" | ||
| ) | ||
|
|
There was a problem hiding this comment.
This was only used in NumericIndex.
| if dtype is None and data.dtype.kind == "f": | ||
| if cls is UInt64Index and (data >= 0).all(): | ||
| # https://github.com/numpy/numpy/issues/19146 | ||
| data = np.asarray(orig, dtype=np.uint64) | ||
|
|
||
| if issubclass(data.dtype.type, str): | ||
| cls._string_data_error(data) | ||
|
|
There was a problem hiding this comment.
Removing this simplifies error handling by not special casing string data.
| @@ -198,7 +201,8 @@ def _ensure_dtype(cls, dtype: Dtype | None) -> np.dtype | None: | |||
| return cls._default_dtype | |||
|
|
|||
| dtype = pandas_dtype(dtype) | |||
| assert isinstance(dtype, np.dtype) | |||
| if not isinstance(dtype, np.dtype): | |||
| raise TypeError(f"{dtype} not a numpy type") | |||
There was a problem hiding this comment.
I strengthen this check here, though it will not be important in the end when everything is moved into Index
mroeschke
added
Dtype Conversions
| self._index_cls(data) | ||
|
|
||
| # shouldn't | ||
| data = ["0", "1", "2"] | ||
| with pytest.raises(TypeError, match=msg): |
There was a problem hiding this comment.
probably requires a whatsnew note
There was a problem hiding this comment.
or could just keep the string_error_whatever to keep this tightly focused and avoid the API change
There was a problem hiding this comment.
This error can in pandas <2.0 only be hit by instantiating Int64Index et al. directly, because Index will give a object dtype to this. Int64Index will be removed in pandas 2.0, so IMO a whatsnew notes should not be added?
Doesn't hit this either it we combine string data with numeric dtype:
>>> import pandas as pd
>>> pd.Index(["1", "2"], dtype="int64")
ValueError: Trying to coerce float values to integers # different code pathThere was a problem hiding this comment.
so IMO a whatsnew notes should not be added
OK, but on the off-chance that the Int64Index removal doesn't happen in time, pls be sure to add a note.
|
thanks @topper-123 |
…andas-dev#49815) * API: NumericIndex([1 2, 3]).dtype should be int64 on 32-bit systems * cleanups * cleanups II
