2.2. Scalar Data Types

Sourcery VSIPL++ defines default types for scalar floating-point and integer data, in both real and complex forms:

typedef float scalar_f;
typedef int scalar_i;
typedef std::complex<scalar_f> cscalar_f;
typedef std::complex<scalar_i> cscalar_i;

Sourcery VSIPL++ also supports double-precision scalar and complex data types.

In addition, Sourcery VSIPL++ defines a number of integer types for indices, strides, and other related values.

dimension_type
Unsigned integer for dimension numbers (0, 1, or 2).
index_type
Unsigned integer for view indices.
index_difference_type
Signed integer for index differences.
stride_type
Signed integer for strides in index ranges.
length_type
Unsigned integer for lengths of index ranges.