Closed
Description
At runtime, when the first argument is of dtype[str_] and StringDType dtype, the second argument can (also) be a native str, a (nested) sequence of str, or any other np.str_ array-like (including NDArray[str_]).
>>> np.char.index(np.array(["where's Wally?"], dtype=np.str_), "Wally")
array([8])
>>> np.char.index(np.array(["where's Wally?"], dtype=np.dtypes.StringDType()), "Wally")
array([8])
>>> np.char.index(np.array(["where's Wally?"], dtype=np.dtypes.StringDType()), np.str_("Wally"))
array([8])But this doesn't appear to be the case when the dtypes "flipped", i.e.
>>> np.char.index(np.array(["where's Wally?"], dtype=np.str_), np.array("Wally", dtype=np.dtypes.StringDType()))
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "~/.pyenv/versions/3.12.6/lib/python3.12/site-packages/numpy/_core/strings.py", line 327, in index
return _index_ufunc(a, sub, start, end)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
numpy._core._exceptions._UFuncNoLoopError: ufunc 'index' did not contain a loop with signature matching types (<class 'numpy.dtypes.StrDType'>, <class 'numpy.dtypes.StringDType'>, <class 'numpy.dtypes._PyLongDType'>, <class 'numpy.dtypes._PyLongDType'>) -> NoneOriginally posted by @jorenham in #27470 (comment)
Metadata
Metadata
Assignees
Labels
No labels