This page contains an archived post to the Java Answers Forum made prior to February 25, 2002. If you wish to participate in discussions, please visit the new Artima Forums.

Message:

SHIFT_JIS to EUC_JP conversion does not work for some Kana characters.

Posted by Sanjay Agnani on February 02, 2001 at 1:16 AM

Hi,

I am trying to convert from SHIFT_JIS to EUC_JP in a Servlet on a Windows 2000 PC. The conversion code I am using is as follows :

try {
temp = request.getParameter("PrivateArea");
privateArea = new String(temp.getBytes"8859_1"),
"JISAutoDetect");
System.out.println( "PrivateArea :" +
�@�@�@�@ privateArea + ":");
printBytes(privateArea.getBytes(),
�@�@ "PrivateArea");
System.out.println();

eucString = new String(privateArea.getBytes("EUC_JP"));
System.out.println("eucString :" +
�@�@ �@�@�@�@�@�@�@�@�@�@�@eucString +�@":");
System.out.println("eucString.getBytes().length :"
�@�@ �@�@+ eucString.getBytes().length +
�@�@�@�@�@�@�@�@�@�@�@�@�@�@":");
printBytes(eucString.getBytes(), "EUC");
System.out.println();
encodeString = URLEncoder.encode(eucString);
System.out.println("encodeString :" +
�@�@ �@encodeString + ":");
System.out.println("encodeString.getBytes).length�@:" +�@
�@�@ �@�@�@�@ encodeString.getBytes().length +
�@�@�@�@�@�@�@�@�@�@�@�@�@ ":");
} catch(java.io.UnsupportedEncodingException ex) {
System.err.println(ex);
}

******************************************

I am able to convert most characters like
�����������@or �A�C�E�G�I. The output is as
follows :
PrivateArea :�A�C�E�G�I:
PrivateArea[0] = 0x83
PrivateArea[1] = 0x41
PrivateArea[2] = 0x83
PrivateArea[3] = 0x43
PrivateArea[4] = 0x83
PrivateArea[5] = 0x45
PrivateArea[6] = 0x83
PrivateArea[7] = 0x47
PrivateArea[8] = 0x83
PrivateArea[9] = 0x49

eucString :����������:
eucString.getBytes().length :10:
EUC[0] = 0xa5
EUC[1] = 0xa2
EUC[2] = 0xa5
EUC[3] = 0xa4
EUC[4] = 0xa5
EUC[5] = 0xa6
EUC[6] = 0xa5
EUC[7] = 0xa8
EUC[8] = 0xa5
EUC[9] = 0xaa

encodeString :%A5%A2%A5%A4%A5%A6%A5%A8%A5%AA:
encodeString.getBytes().length :30:

*******************************************************
PrivateArea:����������:
PrivateArea[0] = 0x82
PrivateArea[1] = 0xa0
PrivateArea[2] = 0x82
PrivateArea[3] = 0xa2
PrivateArea[4] = 0x82
PrivateArea[5] = 0xa4
PrivateArea[6] = 0x82
PrivateArea[7] = 0xa6
PrivateArea[8] = 0x82
PrivateArea[9] = 0xa8

eucString :����������:
eucString.getBytes().length :10:
EUC[0] = 0xa4
EUC[1] = 0xa2
EUC[2] = 0xa4
EUC[3] = 0xa4
EUC[4] = 0xa4
EUC[5] = 0xa6
EUC[6] = 0xa4
EUC[7] = 0xa8
EUC[8] = 0xa4
EUC[9] = 0xaa

encodeString :%A4%A2%A4%A4%A4%A6%A4%A8%A4%AA:
encodeString.getBytes().length :30:
*******************************************************

But I am not able to convert some characters like :
������ or ����������.
The output of my Servlet is as follows :
PrivateArea :����������:
PrivateArea[0] = 0x83
PrivateArea[1] = 0x89
PrivateArea[2] = 0x83
PrivateArea[3] = 0x8a
PrivateArea[4] = 0x83
PrivateArea[5] = 0x8b
PrivateArea[6] = 0x83
PrivateArea[7] = 0x8c
PrivateArea[8] = 0x83
PrivateArea[9] = 0x8d

eucString ::
eucString.getBytes().length :0:

encodeString ::
encodeString.getBytes().length :0:

*******************************************************
PrivateArea:������:
PrivateArea[0] = 0x82
PrivateArea[1] = 0xe7
PrivateArea[2] = 0x82
PrivateArea[3] = 0xe8
PrivateArea[4] = 0x82
PrivateArea[5] = 0xe9
PrivateArea[6] = 0x82
PrivateArea[7] = 0xea
PrivateArea[8] = 0x82
PrivateArea[9] = 0xeb

eucString ::
eucString.getBytes().length :0:

encodeString ::
encodeString.getBytes().length :0:

*******************************************************

What I would like to know is :
1. Am I doing something wrong in my code ?
2. If answer to Q. 1 is No, then is this a bug in Sun's JDK 1.3 for Windows and Solaris (I tried this on Solaris also and I got the same results)?
3. Is there any solution or work-around for this problem that you'll know of ?

Your opinion and help is highly appreciated.

Thank you.

Sanjay.