The binary data type has as its domain sequences of bits, each bit being represented by 0 or 1. Syntax: 172 binary_type = BINARY [ width_spec ] . 318 width_spec = '(' width ')' [ FIXED ] . 317 width = numeric_expression . A binary data type may be defined as either fixed or varying width (number of bits). If it is not specifically defined as fixed width (by using the fixed reserved word in the definition) the binary data type has varying width. The domain of a fixed width binary data type is the set of all bit sequences of exactly the width specified in the type definition. The domain of a varying width binary data type is the set of all bit sequences of width less than or equal to the maximum width specified in the type definition. If no width is specified, the domain is the set of all bit sequences, with no constraint on the width of these sequences. Subbinaries and individual bits may be addressed using subscripts as described in 12.3. Python mapping: BINARY is mapped to the 'str' type. A check is performed to validate it is a binary string representing a number.