Objective #1: Learn how to build structured programs that are divided into functions.

• Every C++ program must have a main function. However, it is wise to develop an algorithm for any given project that divides large tasks into smaller ones. Each of the smaller tasks (let's say, subtasks) can be coded as C++ functions that go together with the main function to make up a structured program.
• Using functions makes it easier to code programs.
• Using functions also makes it easier to debug large programs.
• Using functions makes it easier to maintain and upgrade a program after the first version has been finished.
• Since functions are meant to perform single and well-defined tasks, they may be reused in other programs or, even, by other programmers
• Functions are given valid identifier names, just like variables. However, a function name must be followed by parentheses. C++ programmers generally add functions to the bottom of a C++ program, that is after the main function. While this is not a requirement of the compiler, it makes your code easier to read and follow by fellow programmers. If the functions are listed after the main function, then function prototypes must be included at the top of the program, above the main function. The function prototypes "warn" the compiler about the eventual use of the prototyped function and allocate memory. An error will definitely result if you call a function from within the main function that has not been prototyped.

  Example: The following program includes a function prototype which "declares" to the program the existence of the function printMyName which is detailed below the main function. The main function "calls" the function printMyName and sends the argument userInput which has a value of 3. printMyName receives the value and stores it in the parameter numOfTimes. The function then prints out the string "Mr. Minich" 3 times using a for loop.

  

Objective #2: Understand what is meant by the phrase "scope of variables."

• Variables can be global or local in nature. Global variables are ones that are declared outside and above the main function. They can be used in any function throughout the program. It is not wise to use global variables any more than you have to. One small error (for example, miscalculating a value) could lead to multiple, hard-to-find errors in large programs. Local variables are ones that are declared inside of a function, including main. They cannot be used or referred to in other functions.
• In order to strengthen your program and functions' autonomy and encapsulation, you should avoid the use of global variables and try to use local variables instead.
• The scope of a variable is the area in which it can be legally referenced. A global variable's scope is the whole program. A local variable's scope is the function in which it is declared. The scope of a global variable is considered to be wider than the more narrow scope of a local variable.
• In the example program above (with the function printMyName), the variable userInput is a local variable which belongs to the main function. The variable i is a local variable to the function printMyName. (numOfTimes isn't really a variable. It is a parameter technically. See Objective #3 below for more details on parameters.)

Objective #3: Understand how data is passed to functions.

• Data is passed to functions as arguments and parameters. When a function is "called" by the main function (or another function), one or more arguments are passed to the function. On the receiving end, the function accepts these arguments (technically, as parameters). The variable names of the arguments (from the "calling" function) do not have to be the same as the names of the parameters (in the "called" function.) But, the datatypes of the arguments and the parameters must match exactly. The list of parameters in the order that they are specified in the function prototype and function header (see below) is called the function's "parameter list".
• The values that are sent by the calling function to the called function are known as arguments or actual parameters. On the other end, in the called function, those values are stored in formal parameters. The number of actual parameters (or arguments) must match the number of formal parameters or a compiler error will result (unless default values were included). Also, the order of the actual parameters with regard to data type must exactly match the order of the formal parameters with regard to data type or a compiler error will result.
• Within the function definition itself, we distinguish between its first line and the rest of the function. The first line is called the function header and will look very similar to the function's prototype except for a semicolon at the end. The rest of the function is enclosed in braces and is called the function body. Function definitions are commonly placed below the main function in C++. Function prototypes, by the way, are commonly placed before the main function.
• It is good style to include pre- and postconditions with functions that you write though this is not required by our class Coding Standards. Preconditions are conditions that must be true for the function to work properly. Postconditions are conditions that are guaranteed to be true after the function has executed.
• Local variables may be declared and initialized within the body of a function. However, these variables cannot be accessed anywhere else in the C++ program outside of that function since they have local scope.
• Many functions need to "return" a value with a return statement. We type
  
  return 0;
  
  as the last statement in the main functions of all the programs that we write in this C++ course. This returns a message (of "0") to the operating system indicating that our C++ program ran successfully. Often, though, you want your function to return a computed value to the calling function. This value is used in the exact statement that the function was called. For example,

  fahrenheit = celsiusToFahrenheit(celsius);

  sends the argument, celsius, to the function, celsiusToFahrenheit, where it is used to determine the equivalent Fahrenheit temperature. Then, the final value is returned from within the function to the main function where it is assigned to the fahrenheit variable. The keyword void is used as the return type if a function does not have a return statement or need to return a value to the calling function.

• If a function does not need any values in order to operate and perform its function, it will have an empty parameter list such as in:
  
  void displayTitleOfProgram( )
  {
       cout << "Minich's Sample C++ Program" << endl;
  }// end of displayTitleOfProgram
  
  where no arguments needed to be sent to the function for it to fulfill its purpose.

• There are mainly two ways to pass data as arguments to functions in C++. You can pass data by value or by reference. We will not be studying how to pass by reference in this course but it is explained below anyway.
  
  • passing by value is the preferred method. You simply use a variable name, an actual numeric literal, or an expression in the parentheses of the call statement. This method is the safest because it does not actually change the value of the argument in the calling function.
    
  • passing by reference is to be used when you want a single function to actually and permanently change the values of one or more variables. This method is dangerous but sometimes more efficient since it allows you to affect and permanently change the values of arguments in your calling function. You must use an ampersand (&) before the parameter names in the function definition (the first line of the actual called function) to pass variable by reference.

Objective #4: Learn how to use the library functions that are included with the compiler.

• There are many functions available to C++ programmers which were written by other programmers. You can use the #include compiler directive at the top of your program to take advantage of these functions. The source code for those functions does not need to be included within your program. In fact, you do not even have to know exactly how they work. You do have to know how many arguments to send to the functions and what datatypes to use for those functions. For example, there are many mathematical library functions that can be used in your program if you "include" the cmath header file at the top of your program. Similarly, the ctype header file may be useful as well.
• There are many functions available to C++ programmers which were written by other programmers. These functions are known as library functions (and sometimes called predefined functions.) You can use the #include compiler directive at the top of your program to take advantage of these functions. The source code for those functions does not need to be included within your program.
  
• There are many library functions in C++. You should be familiar with the library functions sqrt, pow, abs, ceil, floor , and rand. You must add #include <cmath> to the top of your program if you use the functions sqrt, pow, ceil, & floor. You should add #include <cstdlib> to the top of your program if you use the functions abs and rand.
  
  
  • The sqrt function can be used to compute a square root. The following statement displays the square root of 13:
    
    cout << sqrt(13) << endl;
    
    
  • The pow function can be used to calculate a number to a power. The statement
    
    cout << pow(3, 2) << endl;
    
    displays the value 9.
    
    
  • The abs function is used to calculate the absolute value of an integer. The statement
    
    cout << abs(-9) << endl;
    
    displays the positive value 9.
    
    
  • The ceil function returns the next integer greater than it's parameter. The statement
    
    cout << ceil(9.2) << endl;
    
    displays the value 10. While
    
    cout << ceil(-3.9) << endl;
    
    displays the value -3.
    
    
  • The floor function returns the next smallest integer than it's parameter. The statement
    
    cout << floor(9.9) << endl;
    
    displays the value 9. While
    
    cout << floor(-3.9) << endl;
    
    displays the value -4.
    
    
  • The rand function is used to generate a random value as in
    
    cout << "A random number is " << rand() << endl;
    
    
• The rand, srand, and randomize library functions are useful for generating pseudorandom numbers. The rand function produces a random number between or equal to 0 and RAND_MAX, where RAND_MAX is defined in the cstdlib header file for your particular compiler and platform. The srand function sets a seed value for rand so that rand will not always produce the same random numbers every time the C++ program is executed. Our compiler, Visual C++, allows you to use the statement
  
  randomize( );
  
  instead of
  
  srand(time(NULL));
  
  But you must include stdlib and time library functions, in order to use randomize in this way.
  
  
• You may have to manipulate the pseudorandom numbers generated by rand within a program using a technique called scaling. For example, the expression
  
  int(rand( ) / RAND_MAX * 100)
  
  will generate a random integer between 0 and 100. The general expression
  
  low + rand( ) % (high + 1 - low)
  
  will genate a random integer between low and high.