`typecast`

Reinterprets the bytes of a value as another numeric type, without converting the values.

Category: Language Fundamentals > Data Types

Syntax

Y = typecast(X, type)

Returns the bytes of X read as numbers of class type.

Description

Unlike double(x) or uint8(x), which convert values, typecast keeps the memory bits intact and only changes how they are read. For example, the single-precision number 1 has the bit pattern 0x3F800000, so typecast(single(1), 'uint32') is 1065353216.

The total number of bytes must fit evenly in the new type: eight uint8 bytes make one double, but seven do not, and an error is raised. The result follows the byte order of the machine, which is little endian on common hardware.

Input arguments

X: Numeric or logical vector.
type: Target class: 'uint8', 'int16', 'single', 'double', and so on.

Output arguments

Y: Vector with the same bytes as X. The number of elements changes according to the sizes of the two types.

Example

Reading the same bytes as other types

typecast(single(1), 'uint32')
typecast(uint8([1 0 0 0]), 'uint32')
typecast(uint8([0 0 128 63]), 'single')
typecast(int8([-1 -1]), 'uint16')

Output

ans = uint32
   1065353216
ans = uint32
   1
ans = single
   1
ans = uint16
   65535

typecast