BUG: infer_dtype with decimal/complex #37176
Conversation
| @@ -2546,8 +2565,6 @@ def fast_multiget(dict mapping, ndarray keys, default=np.nan): | |||
| # kludge, for Series | |||
| return np.empty(0, dtype='f8') | |||
|
|
|||
| keys = getattr(keys, 'values', keys) | |||
|
|
|||
pandas/_libs/lib.pyx
Outdated
| @@ -1414,10 +1414,12 @@ def infer_dtype(value: object, skipna: bool = True) -> str: | |||
| return "time" | |||
|
|
|||
| elif is_decimal(val): | |||
| return "decimal" | |||
| if all(is_decimal(x) for x in values): | |||
There was a problem hiding this comment.
can you create is_decimal_array to have the same style
There was a problem hiding this comment.
i think you actually want to use the validator e.g as a raw NaN would not work here
There was a problem hiding this comment.
can you expand on this? do you expect something other than Decimal objects to work here?
There was a problem hiding this comment.
NaNs are valid as well here; but more important is that every validator has the same pattern. a is_* function e.g. is_decimal_array, which should call here.
There was a problem hiding this comment.
updated to use the is_* pattern.
since decimal has Decimal("nan"), i wouldnt expect to accept np.nan here
There was a problem hiding this comment.
since decimal has Decimal("nan"), i wouldnt expect to accept np.nan here
yeah ok, though i think we could actually accept np.nan as well and infer a decimal if that is the case (as we do this for all other dtypes). can you open an issue.
|
thanks |
black pandasgit diff upstream/master -u -- "*.py" | flake8 --diffDoes this need a whatsnew?