Winsock Math Operations client \ server

I have an assignment that takes existing client and server code, that is fully able to make a TCP connection with arguments and share data and add to it.

We're asked to add some code to it so that the following happens:
1. Client asks for two floating point numbers
2. Client asks for which Math operation to use (+,-,/,*)..

Then sends those variables to the server for processing.
The server in turn, needs to send back the result to the client.

I've got the client prompting the user for the numbers and operator and storing the input in 3 variables, but from there, I'm lost. While the remaining code continues on to make a connection with the server, I've no idea how to actually tell the server to execute the operation.

Can anyone point out a code example of this type of thing that I might be able to study?

Here's a nice example http://msdn.microsoft.com/en-us/libr...(v=vs.85).aspx

Thanks.
While that has all of the nuts and bolts to establish the connection, I'm really after seeing an example that actually passes a function back and forth.
Again, I'm trying to collect input from the client side, pass it on to the server for processing and have the result be given back to the client.

Any help there would be greatly appreciated.

One of your previous excercises (client/server that had the server return the time to the client) contains all that you need to establish the connection - the existing client/server code. What you need to do is to change what is passed between the client/server to enable you to undertake what is required. In your previous code on the client side

Code:

---

sprintf(szBuf, "From the Client [%s]", szSvr);
nRet = sendto(theSocket,                    // Socket
                  szBuf,                        // Data buffer
                  (int)strlen(szBuf),                // Length of data
                  0,                                // Flags
                  (LPSOCKADDR)&saServer,        // Server address
                  sizeof(struct sockaddr));    // Length of address

---

So instead of passing the message 'From the client...' you pass a string containing what you want your server to do eg. "+,1.23,3.45". In your server code

Code:

---

nRet = recvfrom(theSocket,                // Bound socket
                szBuf,                        // Receive buffer
                sizeof(szBuf),                // Size of buffer in bytes
                0,                        // Flags
                (LPSOCKADDR)&saClient,        // Buffer to receive client address
                &nLen);                        // Length of client address buffer

---

szbuf would now contain "+,1.23,3.45". The server code would parse this, calculate the result and then pass it back to the client the same way as it passed back the time. The client would then display that result instead of displaying the time. Easy!:D

As always 2kaud, Thanks in advance for your insight. It's greatly valued.

As you referenced, I am using the very same code as the last exercise, which accepts command arguments and establishes a connection.
So on the client end, I took a crack at adding some code to prompt the end-user.
I'm able to retrieve all the info from the user and stuff it in 3 variables from the client side, see code:

HTML Code:

---

float f1;
        float f2;
        char mFunc;
       
        // Prompt user for two floating point numbers and what math operation.
        cout << "Enter the first decimal number." << endl;
        cin >> f1;
        cout << "Enter the second decimal number." << endl;
        cin >> f2;
        cout << "Which mathmatical operation would you like to use? (+, -, *, /)" << endl;
        cin >> mFunc;
        cout << "You've entered " << f1 <<" " << mFunc << " "<< f2 << " to be calculated. Sending to server for processing, One moment please." << endl;

---

but if I'm reading you correctly, I'd need to send that in a string to the server in this portion of code:

HTML Code:

---

sprintf(szBuf, "From the Client [%s]", szSvr);

        nRet = send(theSocket,                                       
                                  szBuf,                               
                                  (int)strlen(szBuf),               
                                  0);

---

Have I missed the boat completely, or am I on the right track?

Yes, you're on the right track. Just change the sprintf statement in the client.

Code:

---

sprintf(szBuf, "%c,%f,%f", mFunc,f1,f2);

---

In the server following the recvfrom statement you will need to parse szBuf to extract the operation and the two numbers (each separated by comma). Then perform the operation and finally send the result as a string back to the client (instead of sending the time).

Okay, so doing so passes the variables to the server and the server acknowledges what was sent.

Here's where this really gets tricky for me.
In an effort to fully understand the operation side of it (to actually "do the math"), I created a full program that just handles the cin \ cout and function of doing the math operation.
The program works perfectly, but obviously is not a client \ server program. However, the function that I created to do so should work in the actual project.

The function called mathOp and I added it to the server side
Here it is:

HTML Code:

---

int mathOp()
{
       
                if (mFunc == '+')
                {
                        answer = f1 + f2;
                }
            if (mFunc == '-')
                {
                        answer = f1 - f2;
                }
                if (mFunc == '*')
                {
                        answer = f1 * f2;
                }
                if (mFunc == '/')
                {
                        answer = f1 / f2;
                }
            cout << "The solution to your math query is: " << answer << endl;
               

return 0;
}

---

I declare it as a function prototype at the head of the program and call it out at the same place that my teacher's old time statement was.
The timestamp code:

HTML Code:

---

GetTime(szBuf, szSvr);

---

My call:

HTML Code:

---

mathOp();

---

Note that he puts the szBuf and szSvr in as arguments. When I try to do it with my mathOp, it tells me that the function can't hold arguments.
With the additions that I've made, the server does compile and load, but I know that I have NOT parsed the szBuf. I've searched all night trying to find how to do it, but am having no luck. Can you elaborate on "parsing." I know it means taking "parts" of the string and putting them on their own lines (or other). I don't know what code makes that happen though.

If you don't have a sufficient grasp of c++ basics to be able to code a simple function that takes arguments, then I suggest you look at

http://www.learncpp.com/cpp-tutorial...and-arguments/

One of the ways to extract numbers as a double from a string is to use strtod(..). So if

Code:

---

sprintf(szBuf, "%c,%f,%f", mFunc, f1, f2);

---

is used to create a string, then

Code:

---

double        f1,
                f2;

char        *s = &szBuf[2],
        *e,
        mFunc;

        mfunc = *szBuf;
        f1 = strtod(s, &e);
        f2 = strtod(e + 1, &s);

---

can be used to parse it back into operator and the two numbers. These can then be passed as parameters to your mathop function (which should return a float as your f1, f2 numbers are of type float).

You're correct. I haven't got enough knowledge yet of the basics to have performed this yet.
Been looking all night

This link is a great help. Thanks again.

While the link you provided helps with building functions with arguments, or parameters, my real problem is putting it all in context to the stuff that's already there.

I see and appreciate what you've done here, taking the string that entered the server and parsing it with this code:

HTML Code:

---

double        f1,
                f2;

char        *s = &szBuf[2],
        *e,
        mFunc;

        mfunc = *szBuf;
        f1 = strtod(s, &e);
        f2 = strtod(e + 1, &s);

---

However, how is it I incorporate these variables into my mathOp function to actually perform the calculation?

HTML Code:

---

double mathOp()
{

                        if (mFunc == '+')
                        {
                                answer = f1 + f2;
                        }
                        if (mFunc == '-')
                        {
                                answer = f1 - f2;
                        }
                        if (mFunc == '*')
                        {
                                answer = f1 * f2;
                        }
                        if (mFunc == '/')
                        {
                                answer = f1 / f2;
                        }
               

        return 0;
}

---

Quote:

---

Originally Posted by tmcfadden

However, how is it I incorporate these variables into my mathOp function to actually perform the calculation?

---

Your approach to this problem is backwards, to be honest with you. You should start from the "inside-out", not from the "outside-in".

The first thing you should have done is write a function that takes two double arguments, an operation, and returns the answer. It doesn't matter where the arguments came from, whether it be from a server, the keyboard, an input file, or hard-coded values from a sample main() program.

Code:

---

#include <iostream>

double mathOp(double arg1, double arg2, char functionToDo)
{
    switch (functionToDo)
    {
        case '*':
            return arg1 * arg2;
    }
    return 0; 
//...
}

int main()
{
  std::cout << "The answer to 3 * 6 is " << mathOp(3, 6, '*');
}

---

The function above doesn't care where the numbers or operation came from. Then you test the daylights out of that function, replacing 3, 6, and the '*' with other numbers, operations, etc. to make sure it works correctly.

Once you have a simple main() program working, then and only then do you now consider "client / server", parsing input, etc. The function is in place, and now it's just a matter of taking the data from wherever it comes from, and supplying the data to the function.

Regards,

Paul McKenzie