Migrating BCD conversion code from Java to .Net

Question

I am converting a Java class that converts BCD data to ASCII

I am converting it to .Net BCD Convertor

Following is the converted from java but it is giving wrong converted value e.g. for 123456789 it is giving 123456153153

 public static string GetStringFromBcd(int[] b)
 {
   StringBuilder buffer = new StringBuilder();
   foreach (var t in b)
   {
     if ((t & 0x0000000F) == 0x0000000F && ((t >> 4) & 0x0000000F) == 0x0000000F)
     {
         break;
     }
     buffer.Append((t & 0x0000000F) + "");
     if ((t & 0x000000F0) != 0x000000F0)
     {
         buffer.Append(((t >> 4) & 0x0000000F) + "");
     }
   }
}

What could be the problem?

EDIT: ANSWER:

I got the source program where the data has been BCD encoded. I found that nothing was wrong in that logic, then I discovered the source of the function where the data was converting from network stream to string and later converted to byte/int array. following is the code

 int bytesRead = tcpClient.Receive(message);//, 0, bytetoReadSize);
 if (bytesRead == 0)
 {
     break;
     //the client has disconnected from the server
 }

 //message has successfully been received
 data += new ASCIIEncoding().GetString(message, 0, bytesRead);

here is the problem ASCIIEncoding does not convert many encoded character and gives '?'63 instead of those character , when putting 63 in BCD conversion logic it gives 153.

To resolve this error, I Modified the last line and instead of decoding , I am simply casting the received byte to char.

 foreach (byte b in message)
 {
     data += ((char) b);
 }

Is int the correct way to go? Surely it would be uint for BCD? — KingCronus, Jul 26 '12 at 09:41
thats right, i'll make this change, anyways I checked it is not having any signed int. — Imran Rizvi, Jul 26 '12 at 10:10
Can you expand the problem your trying to solve? I've seen conversion when reading a file, but I've always processed them as byte arrays. I've never seen them processed as int or unsigned ints. — Mike, Jul 26 '12 at 14:57

score 0 · Answer 1 · edited May 23 '17 at 12:30

Here's a Similar Question that comes about it a few different ways.

Here's a Site that has an excellent detailed description of what your facing.

It shouldn't be that hard, but processing them as int's is going to be much harder.

Zoned Decimal (BCD) is pretty straight forward to convert, but you have to be careful if your taking files from a mainframe that have been converted via an ASCII transfer. It can still be converted, but the byte values change due to the ebcdic to ascii conversion during the FTP.

If your processing binary files, it's much easier to deal with.

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;

namespace TestZoned
{
  class Program
  {
    public static String zoneToString(byte[] zoneBytes)
    {
        Encoding ascii = Encoding.ASCII;
        Encoding ebcdic = Encoding.GetEncoding("IBM037");
        byte[] asciiBytes = null;
        String str = null;

        int zoneLen = zoneBytes.Length;
        int i = zoneLen - 1;
        int b1 = zoneBytes[i];
        b1 = (b1 & 0xf0) >> 4;

        switch (b1)
        {
            case 13:
            case 11:
                zoneBytes[i] = (byte)(zoneBytes[i] | 0xf0);
                asciiBytes = Encoding.Convert(ebcdic, ascii, zoneBytes);

                str = "-" + ASCIIEncoding.ASCII.GetString(asciiBytes);
                break;

            default:
                zoneBytes[i] = (byte)(zoneBytes[i] | 0xf0);
                asciiBytes = Encoding.Convert(ebcdic, ascii, zoneBytes);

                str = ASCIIEncoding.ASCII.GetString(asciiBytes);
                break;
        }

        return (str);
    }

    static void Main(string[] args)
    {
        byte[] array =  { 0xf0, 0xf0, 0xf1 }; // 001
        byte[] pos = { 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, 0xf9 }; // 123456789
        byte[] neg = { 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, 0xd9 }; // -123456789

        Console.WriteLine("Converted: {0}", zoneToString(array));
        Console.WriteLine("Converted: {0}", zoneToString(pos));
        Console.WriteLine("Converted: {0}", zoneToString(neg));
    }
}

}

be aware of ASCIIEncoding , as it will not decode all the encoded chars properly and gives '?' in place of them. So just use char casting. — Imran Rizvi, Jul 31 '12 at 09:20

Mike · Answer 2 · 2012-07-27T15:41:25.357

If you wanted to stick to something similar

    public static String GetStringFromBcd(byte[] zoneBytes)
    {
        StringBuilder buffer = new StringBuilder();

        int b1 = (zoneBytes[zoneBytes.Length - 1] & 0xf0) >> 4;
        if ( (b1 == 13) || (b1 == 11) ) buffer.Append("-");

        for (int i = 0; i < zoneBytes.Length; i++)
        {
            buffer.Append((zoneBytes[i] & 0x0f));
        }

        return buffer.ToString();
    }

Migrating BCD conversion code from Java to .Net

2 Answers2