> > while UTF-16 uses 2 bytes for most characters.
> since it's gone midnight and I no longer need to be helpful in this
> thread I could query the definition of most here, xFFFF not being most
> of x10FFFF by some definitions of most. (Although depending whether you
> view an unallocated unicode slot as a character, the numbers might be
> different) 

If the Unicode scalar value is less that 0xFFFF it only requires two bytes
using UTF-16 to encode but if it's greater than 0xFFFF then UTF-16
represents that value using a "surrogate pair" which is four bytes total in
length. Since most Unicode characters have a value that's less than 0xFFFF,
most characters will only require two bytes to encode.

UTF-16 can encode all characters in the 0 to 0x10FFFF range. And so can
UTF-8 and UTF-32. UCS-2, however, cannot encode characters above 0xFFFF.

Jason.

 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list