There is a bunch of characters above the standard ASCII (ending at 127) and the UTF-8 set of characters (first byte starting at 192).
If these are typed on a terminal, jline throws an Exception (in jline.UnixTerminal , throw new IOException("invalid UTF-8 first byte: " + firstByte)).
We have tried to set the test when to attempt to use UTF-8 to "c > 191" instead of "c > 128", but it didn't result in expected behaviour. No exception was thrown but the character wasn't represented correctly on the terminal either (we typed "§" but got "§").
jline version: 0.9.94
tested on: linux with LANG=sv_SE.UTF8
Confirmed with JRuby's console (jirb); typing ø into the console and hitting enter causes it to exit.