Currently on Python 3.x:

>>> import ctypes
>>> memoryview(ctypes.c_long()).format
'<l'

This is invalid on 64-bit platforms: the above means 32-bit little-endian float. The '<' endian specification turns on the "standard size" mode (similarly as for the struct module), which makes type character have a platform-independent meaning.

Unfortunately, the struct module format syntax does *not* allow specifying native-size non-native-endian items. So just replacing '<' by '^' cannot be in general be done.

Suggested fix attached. It adds a converter function that maps the platform-dependent ctypes codes to struct module standard-size codes; handling c_long and c_bool specially.

    ***

After this patch (and the one in http://bugs.python.org/issue10744 ):

>>> import numpy as np
>>> from ctypes import *
>>> class Point(Structure):
...     _fields_ = [("x", c_long), ("y", c_long)]
... 
>>> class StructWithArrays(Structure):
...     _fields_ = [("x", c_long * 3 * 2), ("y", Point * 4)]
... 
>>> x = StructWithArrays()
>>> y = np.asarray(x)
>>> y.dtype
dtype([('x', '<i8', (2, 3)), ('y', [('x', '<i8'), ('y', '<i8')], (4,))])
>>> y['x'] = [[1,2,3],[4,5,6]]
>>> y['y']['x'] = np.arange(4) + 10
>>> y['y']['y'] = np.arange(4) + 20
>>> x.x[0][0]
1
>>> x.x[0][1]
2
>>> x.x[0][2]
3
>>> x.y[0].x
10
>>> x.y[1].x
11
>>> x.y[0].y
20
>>> x.y[1].y
21