Name: nt126004 Date: 04/23/2003


FULL PRODUCT VERSION :
java version "1.4.1_02"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.1_02-b06)
Java HotSpot(TM) Client VM (build 1.4.1_02-b06, mixed mode)


FULL OS VERSION :
Microsoft Windows 2000 [Version 5.00.2195]

A DESCRIPTION OF THE PROBLEM :
The ByteToChar (and CharToByte) converters map shift-jis character 0x8160 (known as the "wave dash" or "full width tilde") to unicode U+301C. However, the Unicode spec (http://www.unicode.org/charts/PDF/U3000.pdf) states that the mapping should be to unicode U+FF5E. Referring to character U+301C it says:

   This character was encoded to match JIS C-6226 1978 1-33 "wave dash".
   Subsequent revisions of the JIS standard and industry practice have
   settled on JIS 1-33 as being the fullwidth tilde character.

Further, the Windows-31J (aka MS932) converter and deconverter map 0x8160 to U+FF5E.


STEPS TO FOLLOW TO REPRODUCE THE PROBLEM :
run test program below

EXPECTED VERSUS ACTUAL BEHAVIOR :
SJIS?F ?` (81 60) ?? 0xff5e(?`) (fullwidth tilde)
MS932?F ?` (81 60) ?? 0xff5e(?`) (fullwidth tilde)
SJIS?F ?` (81 60) ?? 0x301c(?) (wave dash)
MS932?F ?` (81 60) ?? 0xff5e(?`) (fullwidth tilde)

REPRODUCIBILITY :
This bug can be reproduced always.

---------- BEGIN SOURCE ----------
/*
 * bake.java
 * Java??MS932??SJIS???????R?[?h
 * Created on 2003/04/10, 18:37
 */

import java.io.*;

/**
 * bake - Java??MS932??SJIS???????R?[?h
 * @author wkeese
 */
public class bake {
    
    /** Creates a new instance of bake */
    public bake() {
    }
    
    static String toHexString(byte[] ary) {
        String out = "";
        for(int i=0; i < ary.length; i++) {
            if ( i > 0 ) out = out + " ";
            out = out + Integer.toString( (int) ((char)ary[i]) % 0xff, 16);
        }
        return out;
    }
    
    public static void doit(String desc, byte[] array, String encoding) {
        InputStream is = new ByteArrayInputStream(array);
        
        InputStreamReader isr;
        try {
            isr = new InputStreamReader(is, encoding);
        } catch ( UnsupportedEncodingException uce ) {
            System.out.println("Unsupported coding exception for " + encoding);
            return;
        }
        
        int myChar;
        try {
            myChar = isr.read();
        } catch (IOException ioe) {
            System.out.println("IOException");
            return;
        }
        
        
        System.out.print(isr.getEncoding() + "?F " + desc + " (" +
        toHexString(array) + ") ?? 0x" + Integer.toString( myChar, 16) + "(" + (char)myChar + ") ");
        
        switch(myChar)
        {
            case 0x301c: System.out.println("(wave dash)"); break;
            case 0xff5e: System.out.println("(fullwidth tilde)"); break;
            case 0xfffd: System.out.println("(unknown character)"); break;
            case 0x2460: System.out.println("(circled digit 1)"); break;
        }
    }
    
    /**
     * @param args the command line arguments
     */
    public static void main(String[] args) {
        
        
        System.out.println("java.version = " + System.getProperties().getProperty("java.version"));
        System.out.println("java.vm.version = " + System.getProperties().getProperty("java.vm.version"));
        
         
        // ?f?t?@???g???G???R?f?B???O
        byte[] dummy = { 0x01 };
        InputStream is = new ByteArrayInputStream(dummy);
        InputStreamReader isr2 = new InputStreamReader(is);
        System.out.println("?f?t?@???g???G???R?f?B???O?F" + isr2.getEncoding());
        
        // What is Shift_JIS an alias for?
        try {
            isr2 = new InputStreamReader(is, "SJIS");
            System.out.println("Alias: SJIS ??" + isr2.getEncoding());
         } catch ( UnsupportedEncodingException uce ) {
            System.out.println("Unsupported coding exception SJIS");
        }
        try {
            isr2 = new InputStreamReader(is, "shift_jis");
            System.out.println("Alias: shift_jis ??" + isr2.getEncoding());
        } catch ( UnsupportedEncodingException uce ) {
            System.out.println("Unsupported coding exception for shift_jis ");
        }
       
      // Windows codes for the wave dash character (?`) and maru-ichi ?@
        byte[] wave= { (byte)0x81, 0x60 };
        byte [] maruichi = { (byte)0x87, 0x40 };
        byte[] eucwave= { (byte)0xa1, (byte)0xc1 };
        byte [] eucmaruichi = { (byte)0xad, (byte)0xa1 };
        byte [] iso2022maruichi = { 0x1b, 0x24, 0x42, 0x2d, 0x21, 0x1b, 0x28, 0x42 };
        doit("?`", wave, "SJIS");
        doit("?`",wave, "MS932");
        doit("?`",eucwave, "EUC-JP");
        doit("?@",maruichi, "SJIS");
        doit("?@",maruichi, "MS932");
        doit("?@",eucmaruichi, "EUC-JP");
        //doit("?@",iso2022maruichi, "ISO-2022-JP");
    }
    
}

---------- END SOURCE ----------
(Review ID: 184107)
======================================================================