PowscriptLib[unicode]=$'unicode_parsing_error(utf_type msg)\n echo "ERROR: While parsing a $utf_type unicode character: $msg" >&2\n\nunicode_test_eof(utf_type string i)\n if i > ${string:length}\n unicode_parsing_error $utf_type \'Premature end of input\'\n return 1\n\nunicode_eat_delimiter(type string delim i iref)\n if ${string:slice i length ${delim:length}} isnt $delim\n unicode_parsing_error $type "required delimiter \'$delim\' not found after byte $gotten"\n return 1\n else\n iref:ref= $(math i+${delim:length})\n\n\nparse_utf8(string -- result offset start delim prefix)\n local c byte=\'\' unicode=\'\' final_value\n local state=\'get-byte\'\n declare integer i gotten needed\n i=${state:set 0}\n gotten=0\n needed=1\n if ${delim:unset?}\n prefix:unset=\'\\x\'\n delim:unset=\'\\x\'\n #\n unicode_eat_delimiter utf-8 $string $prefix $i ${i:ref} || return 1\n while gotten < needed\n unicode_test_eof utf-8 $string $i || return 1\n #\n c=${string:index i}\n switch $c\n case [0-9abcdefABCDEF]\n i+=1\n byte&=$c\n case *\n if "${byte:length}:$gotten" is \'1:0\'\n unicode="\\\\x$byte"\n gotten=1\n else\n unicode_parsing_error utf-8 "Invalid unicode $unicode\\\\x$byte"\n return 1\n if ${byte:length} is 2\n switch $gotten\n case 0\n local byte_value=0x$byte\n if byte_value < 0x80\n needed=1\n elif byte_value < 0xe0\n needed=2\n elif byte_value < 0xf0\n needed=3\n elif byte_value < 0xf8\n needed=4\n else\n unicode_parsing_error utf-8 "Invalid byte 1: $byte"\n case *\n local byte_value=0x$byte\n if byte_value < 0x80 or byte_value > 0xbf\n unicode_parsing_error utf-8 "Invalid byte $(math gotten+1): $byte"\n return 1\n state=\'get-byte\'\n gotten+=1\n unicode&="\\\\x$byte"\n byte=\'\'\n if gotten < needed\n unicode_eat_delimiter utf-8 $string $delim $i ${i:ref} || return 1\n printf -v unicode "$unicode"\n if ${result:set?}\n result:ref= $unicode\n else\n echo $unicode\n #\n if ${offset:set?}\n offset:ref= $i\n\n\nparse_utf16(string -- result start offset delim prefix)\n local c unicode1=\'\' unicode2=\'\' final_value\n local state=\'unicode1\' result\n declare integer i\n i=${start:unset 0}\n #\n if ${delim:unset?}\n prefix=${prefix:unset \'\\u\'}\n delim=${delim:unset \'\\u\'}\n #\n unicode_eat_delimiter utf-16 $string $prefix $i ${i:ref} || return 1\n while ${final_value:unset?}\n unicode_test_eof utf-16 $string $i || return 1\n #\n c=${string:index i}\n switch $state\n case unicode1\n switch $c\n case [0-9abcdefABCDEF]\n unicode1&=$c\n i+=1\n if ${unicode1:length} is 4\n local value=0x$unicode1\n if value > 0xd7ff and value < 0xdbff\n state=\'unicode2\'\n unicode_eat_delimiter utf-16 $string $delim $i ${i:ref} || return 1\n elif value > 0xdbff and value < 0xdc00\n unicode_parsing_error utf-16 "Invalid character: \\\\u$unicode1"\n else\n printf -v final_value "\\\\u$unicode1"\n case *\n if ${unicode1:length} > 0\n printf -v final_value "\\\\u$unicode1"\n else\n unicode_parsing_error utf-16 "Invalid character: \\\\u$unicode1"\n case unicode2\n switch $c\n case [0-9abcdefABCDEF]\n unicode2&=$c\n i+=1\n if ${unicode2:length} is 4\n local value=0x$unicode2\n if value > 0xdc00 and value < 0xdfff\n local high=0x$unicode1 low=$value\n final_value=$(math\n ((high - 0xd800) * 0x0400) +\n ((low - 0xdc00) + 0x10000))\n printf -v final_value \'%x\' $final_value\n printf -v final_value "\\\\U$final_value"\n else\n unicode_parsing_error utf-16 "Invalid low surrogate: \\\\u$unicode2"\n case *\n unicode_parsing_error utf-16 "Invalid low surrogate: \\\\u$unicode2"\n if ${result:set?}\n result:ref= "$final_value"\n else\n echo "$final_value"\n if ${offset:set?}\n offset:ref= $i\n\n\nparse_utf32(string -- result start offset prefix)\n local unicode="" final_value c\n declare integer i\n i=${start:unset 0}\n prefix:unset=\'\\U\'\n unicode_eat_delimiter utf-32 $string $prefix $i ${i:ref} || return 1\n #\n while ${final_value:unset?}\n unicode_test_eof utf-32 $string $i || return 1\n #\n c=${string:index i}\n switch $c\n case [0-9abcdefABCDEF]\n unicode&=$c\n i+=1\n if ${unicode:length} is 8\n printf -v final_value "\\\\U$unicode"\n case *\n if ${unicode:nonempty?}\n printf -v final_value "\\\\U$unicode"\n else\n unicode_parsing_error utf-32 "Invalid character: \\\\U"\n return 1\n if ${result:set?}\n result:ref= "$final_value"\n else\n echo "$final_value"\n if ${offset:set?}\n offset:ref= $i'

Uh oh!