* add utf8 support to tokenizer * wrap utf8 functions in string table using a 'u' prefix * document new utf8 functions