From: Antoine Leca (Antoine10646@leca-marti.org)
Date: Tue Jan 18 2005 - 12:39:41 CST
> Of course this loses the fact that UTF-8 data will never contain
> 0xFE or 0xFF (and so UTF-16 with a BOM will never be confused with
> UTF-8, a fact that is important to XML parsers for one
> application).
It cannot be confusing since the 8-bit extended UTF will have something in
the range 0x80-0xBF as second byte, none of them could be expected as part
of UTF-16 or even UTF-32.
Anyway, UTF-8 is not defined this way, so using such a strange stream WILL
be confusing, even if not as first character.
Antoine
This archive was generated by hypermail 2.1.5 : Tue Jan 18 2005 - 12:44:41 CST