C++ Basics (Part 1)

Structure of a C++ Program

Sequence of statements, typically grouped into functions
- function: a subprogram. a section of a program performing a specific task
- Every function body is defined inside a block (see below)
For a C++ executable, exactly one function called main()
Can consist of multiple files and typically use libraries
Statement -- smallest complete executable unit of a program
- Declaration statement
- Execution statement
- Compound statement -- any set of statements enclosed in set braces { } (often called a block)
- Simple C++ statments end with a semi-colon. (A block does not typically need a semi-colon after it, except in special circumstances).
Library
- usually pre-compiled code available to the programmer to perform common tasks
- Compilers come with many libraries. Some are standard for all compilers, and some may be system specific
- Two parts
  - Interface: header file, which contains names and declarations of items available for use
  - Implementation: pre-compiled definitions, or implementation code. In a separate file, location known to compiler
- Use the #include directive to make a library part of a program (satisfies declare-before-use rule)

Building and Running a C++ Program

Starts with source code, like the first sample program

Pre-processing
- The #include directive is an example of a pre-processor directive (anything starting with #).
- #include <iostream> tells the preprocessor to copy the standard I/O stream library header file into the program
Compiling
- Syntax checking, translation of source code into object code (i.e. machine language). Not yet an executable program.
Linking
- Puts together any object code files that make up a program, as well as attaching pre-compiled library implementation code (like the standard I/O library implementation, in this example)
- End result is a final target -- like an executable program
Run it!

Variables

Variables are used to store data. Every C++ variable has a:

Name -- chosen by the programmer (aka identifier)
Type -- specified in the declaration of the variable
Size -- determined by the type
Value -- the data stored in the variable's memory location
Address -- the starting location in memory where the value is stored

When a program is compiled, a symbol table is created, mapping each variable name to its type, size, and address

Identifiers

Identifiers are the names for things (variables, functions, etc) in the language. Some identifiers are built-in, and others can be created by the programmer.

User-defined identifiers can consist of letters, digits, and underscores
Must start with a non-digit
Identifiers are case sensitive (count and Count are different variables)
Reserved words (keywords) cannot be used as identifiers
- A list of reserved keywords in C++ can be found on page 43 in the textbook

Style-conventions (for indentifiers)

Don't re-use common identifiers from standard libraries (like cout, cin)
Start names with a letter, not an underscore. System identifiers and symbols in preprocessor directives often start with the underscore.

Pick meaningful identifiers -- self-documenting

  numStudents, firstName                // good
  a, ns, fn                             // bad

a couple common conventions for multiple word identifiers
1. numberOfMathStudents
2. number_of_math_students

Declaring Variables

Declare Before Use: Variables must be declared before they can be used in any other statements

Declaration format: typeName variableName1, variableName2, ...;

 int numStudents;       // variable of type integer
 double weight;         // variable of type double
 char letter;           // variable of type character

 // Examples of multiple variables of the same type in single
 //  declaration statements

 int test1, test2, finalExam;
 double average, gpa;

Atomic Data Types

Atomic data types are the built-in types defined by the C++ language

bool - has two possible values, true or false
integer types
- char - 1 byte on most systems
  - typically used for representing characters
  - stored with an integer code underneath (ASCII on most computers today)
- short - (usually at least 2 bytes)
- int - (4 bytes on most systems)
- long - (usually 4 or more bytes)
- The integer types have regular and unsigned versions
  - The keyword unsigned used as a type by itself is a shortcut for unsigned int
floating point types - for storage of decimal numbers (i.e. a fractional part after the decimal)
- float
- double
- long double

Sizes

Sizes for these types are implementation dependent
numerical types above listed in smallest to largest order
can use built-in sizeof() function to find out sizes of types or variables for a system
For most computers today: char typically 1 byte; int typically 4 bytes

This program will print the number of bytes used by each type - sizes may vary from system to system

Initializing Variables

To declare a variable is to tell the compiler it exists, and to reserve memory for it
To initialize a variable is to load a value into it for the first time
If a variable has not been initialized, it contains whatever bits are already in memory at the variable's location (i.e. a garbage value)

One common way to initialize variables is with an assignment statement. Examples:

 int numStudents;
 double weight;
 char letter;

 numStudents = 10;
 weight = 160.35;
 letter = 'A';

Variables of built-in types can be declared and initialized on the same line, as well

 int numStudents = 10;
 double weight = 160.35;
 char letter = 'A';

 int test1 = 96, test2 = 83, finalExam = 91;
 double x = 1.2, y = 2.4, z = 12.9;

An alternate form of initializing and declaring at once:

 int numStudents(10);		// these are equivalent to the ones above
 double weight(160.35);
 char letter('A');

 int test1(96), test2(83), finalExam(91);
 double x(1.2), y(2.4), z(12.9);

More Declaration Examples, using various types

Constant Variables

(Woohoo! An oxymoron!)

A variable can be declared to be constant. This means it cannot change once it's declared and initialized
Use the keyword const

MUST declare and initialize on the same line

 const int SIZE = 10;
 const double PI = 3.1415;

 // this one is illegal, because it's not
 //  initialized on the same line
 const int LIMIT;			// BAD!!!
 LIMIT = 20;

A common convention is to name constants with all-caps (not required)

Symbolic Constants (an alternative)

A symbolic constant is created with a preprocessor directive, #define. (This directive is also used to create macros).

Examples:

  #define PI 3.14159
  #define DOLLAR '$'
  #define MAXSTUDENTS 100

The preprocessor replaces all occurrences of the symbol in code with the value following it. (like find/replace in MS Word). This happens before the actual compilation stage begins

Questions to consider

What's the difference between these?

  #define SIZE 10
  const int SIZE = 10;

Why use a constant variable (or symbolic constant)? i.e. what's the advantage?
When is use of a constant a good idea? When is it not worthwhile?

Other Basic Code Elements

Literals

Literals are also constants. They are literal values written in code.

integer literal -- an actual integer number written in code (4, -10, 18)
- If an integer literal is written with a leading 0, it's interpreted as an octal value (base 8).
- If an integer literal is written with a leading 0x, it's interpreted as a hexadecimal value (base 16)
- Example:
```
   int x = 26;			// integer value 26
   int y = 032;			// octal 32 = decimal value 26
   int z = 0x1A;		// hex 1A = decimal value 26
 
```
floating point literal -- an actual decimal number written in code (4.5, -12.9, 5.0)
- Note: these are interpreted as type double by standard C++ compilers
- Can also be written in exponential (scientific) notation: (3.12e5, 1.23e-10)
character literal -- a character in single quotes: ('F', 'a', '\n')
string literal -- a string in double quotes: ("Hello", "Bye", "Wow!\n")

Escape Sequences

String and character literals can contain special escape sequences
They represent single characters that cannot be represented with a single character from the keyboard in your code
The backslash \ is the indicator of an escape sequence. The backslash and the next character are together considered ONE item (one char)
Some common escape sequences are listed in the table below

Escape
Sequence Meaning

\n newline

\t tab

\r carriage return

\a alert sound

\" double quote

\' single quote

\\ backslash

Escape Sequence	Meaning
`\n`	newline
`\t`	tab
`\r`	carriage return
`\a`	alert sound
`\"`	double quote
`\'`	single quote
`\\`	backslash

Comments

Comments are for documenting programs. They are ignored by the compiler.

Block style (like C)

    /* This is a comment.
       It can span multiple lines  */

Line comments -- use the double-slash //

    int x;    // This is a comment 
    x = 3;    // This is a comment

Input and Output with streams in C++:

In C++ we use do I/O with "stream objects", which are tied to various input/output devices.

Some pre-defined stream objects

These stream objects are defined in the iostream library

cout -- standard output stream
- Of class type ostream (to be discussed later)
- Usually defaults to the monitor
cin -- standard input stream
- Of class type istream (to be discussed later)
- Usually defaults to the keyboard
cerr -- standard error stream
- Of class type ostream
- Usually defaults to the monitor, but allows error messages to be directed elsewhere (like a log file) than normal output

To use these streams, we need to include the iostream library into our programs.

 #include <iostream>
 using namespace std;

The using statement tells the compiler that all uses of these names (cout, cin, etc) will come from the "standard" namespace.

Using Input and Output Streams

output streams are frequently used with the insertion operator <<

Format: outputStreamDestination << itemToBePrinted
The right side of the insertion operator can be a variable, a constant, a value, or the result of a computation or operation

Examples:

 cout << "Hello World";		// print a string literal
 cout << 'a';			// print a character literal
 cout << numStudents;		// print contents of a variable
 cout << x + y - z;		// print result of a computation
 cerr << "Error occurred";	// string literal printed to standard error
 //We won't utilize cerr in this course. It's less common than cout esp. in intro programming, but here for completeness.

When printing multiple items, the insertion operator can be "cascaded". Place another operator after an output item to insert a new output item:
```
 cout << "Average = " << avg << '\n';
 cout << var1 << '\t' << var2 << '\t' << var3;
 
```

input streams are frequently used with the extraction operator >>
- Format: inputStreamSource >> locationToStoreData
- The right side of the extraction operator MUST be a memory location. For now, this means a single variable!
- By default, all built-in versions of the extraction operator will ignore any leading "white-space" characters (spaces, tabs, newlines, etc)
- Examples:
```
 int numStudents;
 cin >> numStudents;	// read an integer

 double weight;
 cin >> weight;		// read a double

 cin >> '\n';		// ILLEGAL.  Right side must be a variable
 cin >> x + y;		// ILLEGAL.  x + y is a computation, not a variable
 
```
- The extraction operator can be cascaded, as well:
```
 int x, y;
 double a;
 
 cin >> x >> y >> a;	// read two integers and a double from input
 
```

Some special formatting for decimal numbers

NOTE: You'll most likely need to #include the library iomanip to use these techniques.

By default, decimal (floating-point) numbers will usually print out only as far as needed, up to a certain preset number of decimal places (before rounding the printed result)
```
  double x = 4.5, y = 12.666666666666, z = 5.0;

  cout << x;		// will likely print 4.5
  cout << y;		// will likely print 12.6667
  cout << z;		// will likely print 5
```

A special "magic formula" for controlling how many decimal places are printed:

  cout.setf(ios::fixed);	// specifies fixed point notation
  cout.setf(ios::showpoint);	// so that decimal point will always be shown
  cout.precision(2);		// sets floating point types to print to
				//  2 decimal places (or use your desired number)

Any cout statements following these will output floating-point values in the usual notation, to 2 decimal places.

  double x = 4.5, y = 12.666666666666, z = 5.0;

  cout << x;		// prints 4.50
  cout << y;		// prints 12.67
  cout << z;		// prints 5.00

Here's an alternate way to set the "fixed" and "showpoint" flags

  cout << fixed;        // uses the "fixed" stream manipulator
  cout << showpoint;    // uses the "showpoint" stream manipulator
  cout << setprecision(3);    // uses the set precision stream manipulator (you'll need the iomanip library for this)
				//The above sets precision of the value to 3 numbers. You can change this value based on what you need.

These statements use what are called stream manipulators, which are symbols defined in the iostream library as shortcuts for setting those particular formatting flags

An alternative to fixed point notation is scientific notation:

  cout.setf(ios::scientific);	// float types formatted in exponential notation

Some simple program examples

Hello World
Hello World 2
Hello World 3 - this is how it would be written if the using statement was left out
A sample program with cout and cin
Another version of the above sample program
Example illustrating I/O and formatting of floating-point values in fixed point notation
Example illustrating formatting of floating-point values in scientific notation