#!/usr/bin/perl use strict; use warnings; binmode(STDIN); my %found; while() { while(/((?:(?:[\xF0-\xF4][\x90-\xBF]|[\xE0-\xEF])[\x80-\xBF]|[\xC2-\xDF])[\x80-\xBF])/g) { print "UTF-8 CHAR: $1\n" unless( $found{$1}++ ); } }