In fact, it converts ascii characters into "extended ascii" characters. So the bytes count is correct. If you're interested in pack-to-unicode, I did it too: h

In fact, it converts ascii characters into "extended ascii" characters. So the bytes count is correct. If you're interested in pack-to-unicode, I did it too: https://gist.github.com/xem/7086007 https://gist.github.com/xem/7086007