Ruby UTF8 parser (for fun)

utf8_parser.rb

# -*- coding: utf-8 -*-
# TODO: throw exceptions if bytes.next returns an unexpected type of byte

def utf8_parse(string)
  return enum_for(:utf8_parse, string) unless block_given?
  bytes = string.bytes
  while true
    byte = bytes.next
    yield case byte 
      when 0x00..0x7F then byte
      when 0xC0..0xDF then (byte-0xC0 << 6) | (bytes.next-0x80)
      when 0xE0..0xEF then (byte-0xE0 << 12) | (bytes.next-0x80 << 6) | (bytes.next-0x80)
      else raise "Invalid byte #{byte}"
    end
  end
rescue StopIteration
end

str = "abcd¢世界"
p str.codepoints.to_a
p utf8_parse(str).to_a