Need help with Charset class issue ... :-)

[Dominic Steng�rd]

Hi JavaGurus!

I am trying to convert two characters in the Big5 character set, 0xA2CF and 0xa55f that correspond to the characters A and 北, to the equivalent characters but in the Unicode character set.

The code is:

// Big5 to Unicode
char[] ch1 = {0xA2CF,0xa55f}; // Big5 hexcode for A and 北
String userData = new String(ch1);
byte[] byteArray = userData.getBytes("Big5");
ByteBuffer bBuffer = ByteBuffer.wrap(byteArray);
Charset charset = Charset.forName("UTF-16");
CharsetDecoder decoder = charset.newDecoder();
CharBuffer cBuffer = decoder.decode(bBuffer);
String UTF8UserData = cBuffer.toString();
System.out.println(UTF8UserData);
System.out.println("Encoding version: " + charset.displayName());

The output in the DOS prompt is:

    
?
Encoding version: UTF-16

Not the two characters A and 北, encoded in Unicode, that I had expected. There is not even two ?? but one ? ... meaning the data is being interpreted as one character.

What am I doing wrong?
Any help is hugely appreciated.
Thanks in advance! :-)

Best regards

[Paul Clapham]

No, you're on the wrong track entirely.

char[] ch1 = {0xA2CF,0xa55f};
// This creates two Unicode characters with the specified code points.
// Note that saying that they are Big5 characters doesn't make them that.
String userData = new String(ch1);
// This does nothing but copy the data. It is pointless.
byte[] byteArray = userData.getBytes("Big5");
// This says "Convert these Unicode characters to their Big5 encoding"; the
// exact opposite of what you wanted to do.

Okay, enough of that. The main point is that there's no such thing as a "Big5 character" in Java. All characters in Java are Unicode characters. It is possible to encode Java characters as a stream (or array) of bytes using Big5, and it's possible to decode those bytes back into Unicode characters using Big5 as well. Perhaps something like this:

byte[] bytes = {0xA2, 0xCF, 0xa5, 0x5f};
String chars = new String(bytes, "Big5");