Improving the Program

In our bottom-up approach to developing this program, now that we have a running, useful version of the code ( stock2.c), we can beginto think about adding features to the program to meet the spec given in our task.

The next part of the task we might add is to handle the data for how much we paid for the stock and compute the profit or loss we have on this portfolio. To begin this change, we go back to the algorithm and add the necessary steps:

        Get the number of shares
        Get the selling price
        Get the purchase price
        Print the data values
        Compute the portfolio cost
        Compute the portfolio value
        Compute the profit
        Print the result

We could now translate these new steps into C and modify the stock2 program. We need to declare additional variables for the purchase price, stock cost, and profit. We already know how to compute a value (or cost) given the number of shares and price.

But wait...

This computation is exactly the same for the portfolio cost and the portfolio value. Just the data values (variables) we use are different. This is an excellent example of where we might consider writing a general function which can be given the data and can be used to do both computations.

I call this function stock_price(). We can write an algorithm for this function:

    GIVEN:  the number of shares of stock and a price as
                  whole, numerator, and denominator
    RETURN: the extended price of this stock

	Compute the stock price
	return the value

and then code it in C:

float stock_price(int shares, int whole, int num, int denom)
/* Given: the number of shares, and stock price in whole, numerator
                and denominator.
   Return: the cost/value of this amount of stock.
*/
{  float value;

        /*  Compute the stock price  */
        value = shares * (whole + (float) num / (float) denom);

        /*  Return the value         */
        return value;
}

Things to note about this code:

The header of this function has more information than the header we had for main(). Besides its name (stock_price) it has:

a type identifier before the name: This declares the type of data this function returns.
a list of parameters: These declare the number and types of information to be passed to the function. They look like variable declarations - and they are. These are just like variable declarations in the body of the function (they tell the compiler to allocate cells to store this information and associate names with the cells), except, these variables will get values when the function is called.

The function has another local variable, value.

The assignment statement to compute the extended price uses ONLY the variable names local to the function.

The value to be returned from the function is done with a return statement.

We can call this function in main() with a statement like:

        portfolio_value =  stock_price(shares,sell_whole,
                        sell_numer,sell_denomin);

We pass information to the function in the argument list. When the function is called, the value of each of the argument expressions is COPIED to the corresponding parameter list cell -

the first argument to the first parameter
the second argument to the second parameter
the third argument to the third parameter
the fourth argument to the fourth parameter

The code for the entire program is in stock3.c.

How does this program run?

What went wrong?

to Overview.