Name: boT120536 Date: 11/10/2000


Solaris VM (build Solaris_JDK_1.2.2_05, native threads, sunwjit)

The JISAutoDetect converter can't recognized Japanese-EUC strings correctly. My
fix is for 1.2.2, but JDK 1.3 has same bug, too.

This cause of misrecognition is that a current mask table of 1st byte has wrong
Shift-JIS code range data.
TEST CASE:
import java.io.IOException;

public class jadtest {
    public static void main(String[] args) throws IOException {
byte euc[] = {
(byte)0xc0, (byte)0xbe,
(byte)0xbb, (byte)0xb3,
};
System.out.println(new String(euc, "JISAutoDetect"));
    }
}
(Review ID: 107930)
======================================================================