Thanks Brian, see my in line comments. > The first is not valid UTF-8. I suppose it might be UTF-16: U+A3A4 or > U+A4A3 depending on little or big-endian. Or it could be some older > proprietary Asian encoding. [Ryan] How to correct this (to UTF-8), it is a English XP Pro with PRC as system locale. > > The second of these could be UTF-8. If so it would be codepoint 165, the > 'yen' symbol. Or it could be U+C2A5 in UTF-16. [Ryan] yes, it is chinese currency CNY 'yen' symbol. -- Posted via http://www.ruby-forum.com/.