PATH Mac OS 8 and 9 Developer Documentation > Text Encoding Conversion Manager
Programming With the Text Encoding Conversion Manager

---

Code-Switching Schemes for Multiple Character Sets

Code-switching schemes generally use a sequence of 7-bit values, so they are suitable for mail. ISO 2022 specifies a general code-switching scheme. In its general 7-bit form, it uses

• escape sequences to specify the character sets currently assigned to G0-G3 and C0-C1

• certain C0 and C1 controls to switch the current character set to be any of G0-G3 (using the character sets previously assigned to G0-G3)

• other C1 controls for a temporary character set switch that applies only to the next character

However, ISO 2022 it is rarely used in this form on the Internet. Instead, for certain languages there are one or more predefined combinations of character sets and protocols for use with ISO 2022: for example, ISO-2022-JP (Japanese), ISO-2022-KR (Korean), and ISO-2022-CN (simplified Chinese). Each of these specifies the character sets to be used, the escape sequences or controls used to switch among them, and necessary defaults and reset behavior (such as initial state and the end-of-line reset).

Another common code-switching scheme is HZ, used for Chinese mail and news. This uses ~} and ~{ for switching between ASCII and GB 2312.

The EBCDIC Host encodings used on IBM mainframes for CJK text are a special case and use a sequence of 8-bit values. These encodings combine a single-byte EBCDIC character set and a double-byte IBM character set with graphic characters in the range 0x41-0xFE. The EBCDIC control character Shift Out (SO, 0x0E) is used to switch to the double-byte character set, and the control character Shift In (SI, 0x0F) is used to switch to the single-byte character set.