Tex Texin scripsit: > However, that leaves open the question whether only the Unicode > transform signatures are acceptable or other signatures are also > allowed. So if a vendor defines a code page, and defines a signature > (perhaps mapping BOM/ZWNSP specifically to some code point or byte > string) does that then become acceptable?
IMHO yes. XML documents are not *required* to be in one of the character sets that can be automatically detected by the methods of Appendix F. You can encode your documents in (hypothetical) JOECODE, which uses leading 00 as a signature (ignored by the XML parser) and then A=01, B=02, C=03, and so on. Autodetection will not work here, but it is perfectly conformant to have a processor that understands only UTF-8, UTF-16, and JOECODE. Of course some encodings, such as US-BSCII, which looks just like US-ASCII except that A=0x42, B=0x41, a=0x62, b=0x61 will cause problems for anybody. :-) I am a member of, but not speaking for, the XML Core WG. -- John Cowan [EMAIL PROTECTED] www.ccil.org/~cowan www.reutershealth.com "The competent programmer is fully aware of the strictly limited size of his own skull; therefore he approaches the programming task in full humility, and among other things he avoids clever tricks like the plague." --Edsger Dijkstra
