Next: , Previous: , Up: Future Enhancements   [Contents][Index]

7.1 Localized Messages

It is expected that a near release of Teseq will include support for native-language translations of control label names and descriptions. Note that the acronym portions of label lines will never be translated: they represent official designations for their respective controls.

An important requirement of such translations is that they must not interfere with processing of the document; if they use a shift-based encoding, they will shift to the translation encoding only after the line prefix, which must be a plain ASCII character, and shift back to normal ASCII prior to the end of the line. Bytes with values corresponding to the ASCII control characters NUL, CR, or LF (hexadecimal ‘x00’, ‘x0D’, and ‘0x0A’) will not appear, except when they represent their ASCII functions (in no case will NULs be present in the document).

These requirements rule out the use of wide-character encodings such as UCS-2 or UTF-16, which could not be intermixed with ascii characters and which have character representations that would enclude byte values corresponding to the forbidden characters noted above, but should be compatible with UTF-8, ISO 2022-based encodings such as the ISO 8859 series, ISO 2022-JP, or EUC-JP and other encodings that meet these requirements such as GBK, Big5, or Shift JIS.