CODEWITHSUNDEEP
javabase64encoder
user1525699
user1525699

Reputation: 19

BASE64Encoder results differ from windows to Unix

The following program output differs on Windows vs. Unix systems. Could you please explain to me why this is happening, and how this behavior can be made consistent across Unix and Windows.

Because i am using Unix for the encoding of a file and it would not be possible to decode it on Windows systems.

Any help in this regard would be appreciated. Thanks in advance.

import sun.misc.BASE64Encoder;
import sun.misc.CharacterEncoder;

public class TestEncode {

public static void main(String[] args) {        

        byte signature[] = "[35, 44, -31, 18, 78, 84, -113, 1, 27, 36, -79, -60, 75, -14, -80, -99, 65, 11, -45, -54, 23, -100, 74, -54, -26, -77, 33, -40, 104, 90, -33, 32, -123, -76, -27, -118, -25, -97, -85, 22, -64, 102, -7, 119, -65, 35, -114, 31, -83, 73, -57, 63, -7, 47, -31, 48, 28, -109, 54, -90, -24, -21, -102, 59, 82, -14, -52, -77, -22, -25, -15, -81, 70, 52, -42, 93, 76, -51, 96, 87, 29, -37, -40, -71, -121, 44, -44, 74, 23, -76, 29, 108, -56, 48, 46, -26, -73, -53, 90, 53, 25, -96, 115, -79, 93, -128, -46, -119, -30, 22, -107, -27, 6, -120, 2, 19, -72, -5, 30, -54, -34, 26, -22, -44, 93, 40, 84, -125]".getBytes();

    byte encodedSignature[] = null;

      CharacterEncoder encoder;
      encoder = new BASE64Encoder();
      encodedSignature = encoder.encode(signature).getBytes();


      System.out.println(encodedSignature.length);

}

}

Upvotes: 1

Views: 5423

Answers (2)

mprivat
mprivat

Reputation: 21902

You might be using a different charset on each machine. Try this to find out:

System.out.println("Default Charset=" + Charset.defaultCharset());

I suspect your problem when calling the getBytes() method. By default it uses the platform's default charset. If you want to guarantee that it's using the same one, specify it in the getBytes() method by calling getBytes("UTF-8");

Upvotes: 3

David Kroukamp
David Kroukamp

Reputation: 36423

Can you give this a try and what are your results for windows vs linux:

import com.sun.org.apache.xml.internal.security.exceptions.Base64DecodingException;
import com.sun.org.apache.xml.internal.security.utils.Base64;
import java.io.IOException;

public class Base64Test {

    public static void main(String args[]) throws IOException, Base64DecodingException {
        String orig = "original String before base64 encoding in Java";

        //encoding  byte array into base 64
        String encoded = Base64.encode(orig.getBytes("UTF-8"));//make sure the bytes are utf-8 and not platform default     

        System.out.println("Original String: " + orig );
        System.out.println("Base64 Encoded String : " + new String(encoded,"UTF-8"));//ensure string is utf-8

        //decoding byte array into base64
        byte[] decoded = Base64.decode(encoded);      
        System.out.println("Base 64 Decoded  String : " + new String(decoded,"UTF-8"));//ensure string is utf-8

    }
}

Mine is (Windows only):

Original String: original String before base64 encoding in Java

Base64 Encoded String : b3JpZ2luYWwgU3RyaW5nIGJlZm9yZSBiYXNlNjQgZW5jb2RpbmcgaW4gSmF2YQ==

Base64 Decoded String : original String before base64 encoding in Java

The problems that you might be facing are:

  • The character encoding used by default on Windows is different that the one on Linux. Especailly because getBytes() is involved maybe set it to your own default like getBytes("UTF-8") (in my example i tried to do this)
  • It could be caused by the line separator which defaults to ("\r\n") and ("\n") respectively.

But I cant be certain

Upvotes: 1

Related Questions