Extended latin-1 character set is not encoded while it should #23
Description
Here it is specifically set that any character in the extended set (see bottom) is not encoded.
These should actually be encoded, since otherwise Java would convert them and escape them with \u00C2, resulting in double the amount of characters than initially intended.
Personally I tested this on the server.properties files of Minecraft servers, but I'm sure it's reproducible in other applications too.