3.3.3.2 Numeric character codes

The following variations for numeric character codes are allowed. The examples all assume the character set is ASCII.

• Octal numerals preceded by a zero are taken to be an octal number. For example, 0113 also means decimal 75. If a would-be character code starts with a zero but contains any characters other than the digits `0' through `7', it is invalid.

• Hexadecimal "digits" preceded by `0x' or `0X' are taken to be a hexadecimal number. Case is irrelevant. For example, 0x4b, 0X4b, 0x4B, and 0X4B all mean decimal 75. As with octal, a would-be character code starting with `0x' and containing any characters other than `0'--`9', `a'--`f', and `A'--`F' is invalid.

• A decimal number (consisting of more than one numeral) is itself. For example, 75 means the character code decimal 75. As before, a would-be character code starting with `1'--`9' and containing any characters other than `0'--`9' is invalid.

• A single digit, or a single character not in the encoding vector as a name, is taken to represent its value in the underlying character set. For example, K means the character code decimal 75, and 0 (the numeral zero) means the character code decimal 48 (if the machine uses ASCII).

• If the string being parsed as a character code starts with a digit, the appropriate one of the previous cases is applied. If it starts with any other character, the string is first looked up as a name.

Character codes must be between zero and 255 (decimal), inclusive.