RE: Multiple Character Encodings in XSLT/XML

> If this is not supported, why not? - Any theoretical reason?

You would need a protocol that indicates where one encoding ends and the
next one starts. This needs to be below the XML syntax layer, because the
character decoding is generally operating on byte streams in a way that is
independent of the syntactic context (and it would be very inefficient to do
otherwise). Such protocols exist, for example ISO 2022, but they are not
widely used, and with good reason: UTF-8 does the same thing much better.
For example, with ISO 2022 you only know what encoding is used for a
subsequence of the file if you have read forwards to that point from the
beginning of the file.

http://en.wikipedia.org/wiki/ISO_2022

Michael Kay
http://www.saxonica.com/