Native2ascii as java code

Question

Hi all,
I need to write a function that does the work of the command native2ascii, in order to read a file containing japanese characters and write their representations \uxxxx in another one (without calling native2ascii with exec()).
I have written this function, but it is buggy since it sometimes it adds characters to the converted String.
here is the function:
public static String convertToUnicode(String entry, String encoding) {
byte[] bytes = entry.getBytes();
String temp = "";
try {
temp = new String(bytes, encoding);
} catch(Exception e) {
e.printStackTrace();
char[] chtb = temp.toCharArray();
temp = "";
for(int i = 0; i < chtb.length; i++) {
temp += "\\u" + Integer.toHexString((int)chtb);
return temp;
Regards,
Hani

Answer 1

I can see multiple possible problems there:
byte[] bytes = entry.getBytes();Conversion will happen in the default character encoding; you can get anything as result. It would be safer to use one of the 16 bit unicode encodins or get the data as an array characters.
temp = new String(bytes, encoding);I don't quite understand why you do this... if "encoding" is different from the system's default character encoding you'll most probably get nonsense back.
temp += "\\u" + Integer.toHexString((int)chtb[ i]);Integer.toHexString will not pad zeroes in front of the value so for instance the question mark will come out as "\u3f" instead of "\u003f". And by the way, you really should use StringBuffer instead of String in that loop...
here's my try:public static String convertToUnicode(String entry) {
        char[] chtb = entry.toCharArray();
        StringBuffer temp = new StringBuffer(6*chtb.length);
        int BIT_MASK = (1 << 16);
        for(int i = 0; i < chtb.length; i++) {
            temp.append("\\u");
            temp.append(Integer.toHexString(chtb[ i] | BIT_MASK).substring(1));
        return temp.toString();
}There's still some space for left for optimization...