To your updated question: at the moment, all you can do is write some kind of user code that processes reading / writing files at the byte level and does the necessary conversions.
If you are referring to how you can use different ruby character encodings with version 1.9, I point you to
Ruby 1.9
M17n