BUG: numpy.strings.str_len fails on null object

[lithomas1]

Describe the issue:

numpy.strings.str_len should put np.nan when in that position when encountering an string array with a missing value, but instead errors with

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ValueError: The length of a null string is undefined

Reproduce the code example:

import numpy as np
a = np.array(["a", "b", None], dtype=np.dtypes.StringDType(na_object=None))
np.strings.str_len(a)

Error message:

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ValueError: The length of a null string is undefined

Python and NumPy Versions:

'2.1.0.dev0+git20240424.f8392ce'

Runtime Environment:

No response

Context for the issue:

No response

[ngoldbaum]

This is by design. What would be a sensible value for the length of a null string? Note that numpy doesn't have support for missing integer values.

Also, separately, none of the string ufuncs support None as a NaN-like NA object that passes through operations.