> Actually dealing with shift-state dependent multi-byte encodings in a > portable way in C makes the infinite horrors of Unicode and UTF-8 > seem very attractive. This seems a quite acurate summary of the situation. -- Roberto