While trying out ruby's utf-8 features, i came across a problem. I input a string with only special chars ("ş"
to be more precise). After executing the code below, the text cursor and the input field got corrupted.

the code:
a = "ş"
puts a.slice(0,1)

Because the special chars stored with 2 bytes, slicing a one byte part doesn't return a correct value. I believe this
is caused by an unhandled exception.

I attached the problems screenshot also.

fxri was running on a windows xp and it was bundled with ruby-186-26 self-extraction package.