Home/ Questions/Q 4029164
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-05-20T11:18:39+00:00

I sit with a weird problem. I use Apache’s commons-codec (ver 1.4). The following

  • 0

I sit with a weird problem. I use Apache’s commons-codec (ver 1.4). The following code snippet work correctly on my PC (java version 1.6, Glassfish ver 2.1) in a standalone app and in an EJB, but on my server in an EJB it doesn’t work correctly (but in a standalone app it works correctly)

...
org.apache.commons.codec.binary.Base64 b64 = new org.apache.commons.codec.binary.Base64();
byte[] bytes = b64.decode(makeSignedBytes(strB64.getBytes("UTF-8")));
...
private byte[] makeSignedBytes(byte[] ubytes)
{
    byte[] sbytes = new byte[ubytes.length];
    for (int i = 0; i < ubytes.length; i++)
    {
        sbytes[i] = (byte)(0x000000FF & ((int) ubytes[i]));
    }
    return sbytes;
}

The input string is: 

4-sDHXi_2Tu2a8k8NPs1FBT3t7UvN7CksUV6gfSE_Ks0aiCPbdeGM8qLdC58b2_hFH7lEp8m9cyPYQOTo4E0t66ZYP8n8tRhT87c8iD34pCd80qvP9vIXsNsodRaGzK5

The output byte array should look like this (I’ve hex printed it):

|E3|EB|03|1D|78|BF|D9|3B|B6|6B|C9|3C|34|FB|35|14|14|F7|B7|B5|2F|37|B0|A4|B1|45|7A|81|F4|84|FC|AB|34|6A|20|8F|6D|D7|86|33|CA|8B|74|2E|7C|6F|6F|E1|14|7E|E5|12|9F|26|F5|CC|8F|61|03|93|A3|81|34|B7|AE|99|60|FF|27|F2|D4|61|4F|CE|DC|F2|20|F7|E2|90|9D|F3|4A|AF|3F|DB|C8|5E|C3|6C|A1|D4|5A|1B|32|B9|

96 bytes long, when the server get’s it wrong it’s only 93 bytes and looks like this:

|E2|C0|C7|5E|2D|93|BB|66|BC|93|C3|4F|B3|51|41|4F|7B|7B|52|F3|7B|0A|4B|14|57|A8|1F|48|42|AC|D1|A8|82|3D|B7|5E|18|CF|2A|2D|D0|B9|F1|BD|A1|14|7E|E5|12|9F|26|F5|CC|8F|61|03|93|A3|81|34|B7|AE|99|60|FF|27|F2|D4|61|4F|CE|DC|F2|20|F7|E2|90|9D|F3|4A|AF|3F|DB|C8|5E|C3|6C|A1|D4|5A|1B|32|B9|

I have no idea why it works on my pc and not on the server 🙁

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    The wrong result is caused by replacing the second character in the input string (U+002D HYPHEN-MINUS) with U+2010 HYPHEN and replacing underscores with spaces. Perhaps it’s a result of passing input string through some “smart” text editor. So, actually it looks like you pass a wrong input string in.

    Other flaws in your code:

    • makeSignedBytes() method makes no sense and isn’t needed
    • strB64.getBytes("UTF-8") is semantically wrong, it should be strB64.getBytes("ASCII")
    • 0