How to structure custom binary protocol TCP request

Question

I'm working on a code challenge which, in short, provides a server that encodes/decodes the data sent in the request. It is a TCP/IP server with the following custom binary request/response structures:

I need to create a desktop client that will send requests and receive responses, and display data accordingly. I'm having trouble understanding how the request would be formed, and how I would send the request to the TCP/IP server (using C#). Could someone provide perhaps an example of what this might look like?

after googling "C# byte array" and "C# send bytes TCP" where did you get stuck? — Keith Nicholas
– Keith Nicholas, Commented Nov 16, 2017 at 1:58
I'm stuck at understanding how the request should look. If I had an example request I might be able to figure it out. e.g. if I was trying to use "test string" as the data, what would that look like in request form? — Chris V.
– Chris V., Commented Nov 16, 2017 at 2:06

Keith Nicholas · Accepted Answer · 2017-11-16 03:16:26Z

2

not sure about your checksum, but constructing a packet would look something like

        byte[] CreatePacket(string s, Operation op)
        {
            var packet = new List<byte>();
            packet.AddRange(BitConverter.GetBytes((UInt16) 0x1092));
            packet.Add(1);
            var data = Encoding.ASCII.GetBytes(s);
            packet.AddRange(BitConverter.GetBytes((UInt32) (9 + data.Length)));
            packet.Add((byte) (op == Operation.Encode ? 1 : 2));
            packet.AddRange(data);
            packet.Add(CheckSum(packet));
            return packet.ToArray();
        }

given

    enum Operation
    {
        Encode,
        Decode,
    }

and

server.Send(CreatePacket("test string", Operation.Encode))

Also with the checksum spec...

        byte CheckSum(List<byte> packet)
        {
            return (byte)(packet.Select((b, i) => (value: b, index: i))
                .Sum(o => (o.value & (1 << (o.index % 8))) > 0 ? 1 :0) % 256);
        }

edited Nov 16, 2017 at 3:16

answered Nov 16, 2017 at 2:38

Keith Nicholas

44.4k15 gold badges101 silver badges168 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

Chris V. Over a year ago

FWIW, The checksum is calculated by looping through each byte and evaluating a bit of the byte in a specific position depending on the byte being evaluated. This checksum is specifically - For the 1st byte in the Request / Response, evaluate the bit in the 1st position. If the bit being evaluated is 1 then increment the checksum by 1. If the bit being evaluated is 0 do nothing. - For the 2nd byte in the Request / Response, evaluate the bit in the 2nd position. If the bit being evaluated is 1 then increment the checksum by 1. If the bit being evaluated is 0 do nothing.

Chris V. Over a year ago

also, does your solution take into consideration the fact that the length aspect expects little endian UInt32?

Keith Nicholas Over a year ago

by default it is little endian. But it can change on other platforms ( I'm not sure there is a platfrom that is different) you can see what it is on your platform with something like Console.WriteLine(BitConverter.IsLittleEndian);

Keith Nicholas Over a year ago

also, not sure about the checksum definition, I made some assumptions so may or may not work,

Collectives™ on Stack Overflow

How to structure custom binary protocol TCP request

1 Answer 1

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related