npctypes.types module

npctypes.types.ctype(a_type)

Takes a numpy.dtype or any type that can be converted to a numpy.dtype and returns its equivalent ctype.

Parameters:a_type (type) – the type to find an equivalent ctype to. Returns:the ctype equivalent to the dtype provided. Return type:(ctype)

Examples

>>> ctype(float)
<class 'ctypes.c_double'>

>>> ctype(numpy.float64)
<class 'ctypes.c_double'>

>>> ctype(numpy.float32)
<class 'ctypes.c_float'>

>>> ctype(numpy.dtype(numpy.float32))
<class 'ctypes.c_float'>

>>> ctype(int)
<class 'ctypes.c_long'>

npctypes.types.get_ndpointer_type(a)

Takes a numpy.ndarray and gets a pointer type for that array.

Parameters:a (ndarray) – the ndarray to get the pointer type for. Returns:the pointer type associated with this array. Return type:(PyCSimpleType)

Examples

>>> a = numpy.zeros((3, 4), dtype=float)
>>> a_ptr = get_ndpointer_type(a)

>>> a_ptr
<class 'numpy.ctypeslib.ndpointer_<f8_2d_3x4_C_CONTIGUOUS_ALIGNED_WRITEABLE_OWNDATA'>

>>> a_ptr._dtype_
dtype('float64')
>>> a_ptr._ndim_
2
>>> tuple(int(s) for s in a_ptr._shape_)
(3, 4)
>>> a_ptr._flags_
1285
>>> numpy.ctypeslib.flagsobj(a_ptr._flags_)
  C_CONTIGUOUS : True
  F_CONTIGUOUS : True
  OWNDATA : True
  WRITEABLE : False
  ALIGNED : True
  WRITEBACKIFCOPY : False
  UPDATEIFCOPY : False

npctypes.types.tinfo(a_type)

Takes a numpy.dtype or any type that can be converted to a numpy.dtype and returns its info.

Parameters:a_type (type) – the type to find info for. Returns:info about the type. Return type:(np.core.getlimits.info)

Examples

>>> tinfo(float)
finfo(resolution=1e-15, min=-1.7976931348623157e+308, max=1.7976931348623157e+308, dtype=float64)

>>> tinfo(numpy.float64)
finfo(resolution=1e-15, min=-1.7976931348623157e+308, max=1.7976931348623157e+308, dtype=float64)

>>> tinfo(numpy.float32)
finfo(resolution=1e-06, min=-3.4028235e+38, max=3.4028235e+38, dtype=float32)

>>> tinfo(complex)
finfo(resolution=1e-15, min=-1.7976931348623157e+308, max=1.7976931348623157e+308, dtype=float64)

>>> tinfo(numpy.int32)
iinfo(min=-2147483648, max=2147483647, dtype=int32)