Canonical way of testing for str dtype?

Question

What is the recommended way of checking for str (U...) dtype? I suppose I could do dtype.char == "U" but that feels a bit hackish.

Some dtypes can be directly compared to their Python equivalent, for example

np.dtype("f8") == float
# True

and str can be used to create a U... array, for example

 np.arange(4).astype(str)
 # array(['0', '1', '2', '3'], dtype='<U21')

But

 np.arange(4).astype(str).dtype == str
 # False

:-(

Answer 1

I think np.issubdtype() should be what you are looking for:

>>> s = np.arange(4).astype(str)
>>> s
array(['0', '1', '2', '3'], 
      dtype='<U24')
>>> np.issubdtype(s.dtype, str)
True