Fredrik Lundh wrote: > character references refer to code points in the Unicode code > space, so you just convert the bytes you get after converting > to UTF-8. "so you cannot just", of course. </F>