newint changes behavior of int on numpy.string_ conversion

[ssolari]

newint blows up code that works with 'int' in python 2 on numpy strings. python 3 int works as expected. When both python 2 and python 3 int do the same thing, newint should probably maintain the default expected behavior.

Python 2.7.11 |Anaconda 2.5.0 (x86_64)| (default, Dec  6 2015, 18:57:58) 
[GCC 4.2.1 (Apple Inc. build 5577)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
Anaconda is brought to you by Continuum Analytics.
Please check out: http://continuum.io/thanks and https://anaconda.org
>>> import numpy as np
>>> from builtins import int as newint
>>> int(np.string_('1'))
1
>>> newint(np.string_('1'))
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "~/anaconda/envs/py27/lib/python2.7/site-packages/future/types/newint.py", line 57, in __new__
    val = x.__int__()
TypeError: don't know how to convert scalar number to int

[ssolari]

and even different inconsistent behavior with numpy that will cause bugs later without causing the same immediate TypeError.

>>> np.array(['1']).astype(int)
array([1])
>>> np.array(['1']).astype(newint)
array(['1'], dtype=object)

[nharada1]

I'm also seeing this issue even from conversions not from strings. For example:

>>> a = np.array([1.2])
>>> a
array([ 1.2])
>>> a.astype(int)
array([1.2], dtype=object)

[nharada1]

I'm able to work around this by casting to np.int64