# File lib/itex_stringsupport.rb, line 107
   def is_utf8?
     #expand NCRs to utf-8
     text = self.check_ncrs.as_bytes
     
     # You might think this is faster, but it isn't
     #pieces = self.split(/&#[xX]([a-fA-F0-9]+);/)
     #1.step(pieces.length-1, 2) {|i| pieces[i] = [pieces[i].hex].pack('U*')}
     #pieces = pieces.join.split(/&#(\d+);/)
     #1.step(pieces.length-1, 2) {|i| pieces[i] = [pieces[i].to_i].pack('U*')}
     #text = pieces.join
          
     #ensure the resulting string of bytes is valid utf-8
     text =~ UTF8_REGEX
   end