# 1

Chapter 3 - Functions
Outline 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 3.13 3.14 3.15 Introduction Program Components in C++ Math Library Functions Functions Function Definitions Function Prototypes Header Files Random Number Generation Example: A Game of Chance and Introducing enum Storage Classes Scope Rules Recursion Example Using Recursion: The Fibonacci Series Recursion vs. Iteration Functions with Empty Parameter Lists

2

Chapter 3 - Functions
Outline 3.16 3.17 3.18 3.19 3.20 3.21 Inline Functions References and Reference Parameters Default Arguments Unary Scope Resolution Operator Function Overloading Function Templates

3

3.1  Divide and conquer

Introduction

± Construct a program from smaller pieces or components ± Each piece more manageable than the original program

All rights reserved.4 3. classes ± Prepackaged: from the standard library  Functions invoked by function call ± Function name and information (arguments) it needs  Function definitions ± Only written once ± Hidden from other functions  2003 Prentice Hall.2 Program Components in C++  Modules: functions and classes  Programs use new and ³prepackaged´ modules ± New: programmer-defined functions. . Inc.

.5 3. All rights reserved..e.2 Program Components in C++  Boss to worker analogy ± A boss (the calling function or caller) asks a worker (the called function) to perform a task and return (i.  2003 Prentice Hall. report back) the results when the task is done. Inc.

Inc. All rights reserved. argument2. ± sqrt (square root) function The preceding statement would print 30 ± All functions in math library return a double  2003 Prentice Hall.6 3. «). . or ± functionName(argument1.  Example cout << sqrt( 900.3 Math Library Functions  Perform common mathematical calculations ± Include the header file <cmath>  Functions called by writing ± functionName (argument).0 ).

Inc.6x ).  2003 Prentice Hall.3 Math Library Functions  Function arguments can be ± Constants  sqrt( 4 ). All rights reserved. ± Expressions  sqrt( sqrt( x ) ) . .  sqrt( 3 . ± Variables  sqrt( x ).7 3.

0 ) is 7. y ) remainder of x/y as a floatingfmod( 13. M e tho d ceil( x ) . y ) x raised to power y (xy) pow( 2.0 ) is 30.71828 exp( 2.0 fabs( -8.0 ) is 1.0 (x in radians) exp( x ) exponential function ex exp( 1. 3.0 fmod( x.657.389056 ) is 2. 7 ) is 128 pow( 9.0 ) is 0.0 ) is 2.992 point number log( x ) natural logarithm of x (base e) log( 2.0 pow( x. .8 De sc rip tio n Exa m p le rounds x to the smallest integer ceil( 9.718282 ) is 1. All rights reserved.0 ) is 3.0 tan( x ) trigonometric tangent of x tan( 0.1 ) is 5. 2.  2003 Prentice Hall.76 ) is 8.8 ) is -9.0 log( 7.0 ) is 0 (x in radians) sqrt( x ) square root of x sqrt( 900.0 cos( x ) trigonometric cosine of x cos( 0.0 log10( x ) logarithm of x (base 10) log10( 10.0 ) is 2.0 ) is 0 (x in radians) Fig .2 ) is 9.38906 fabs( x ) absolute value of x fabs( 5.5 ) is 3 sin( x ) trigonometric sine of x sin( 0.8 ) is -10.2 M a th lib ra ry func tio ns.0 not greater than x floor( -9. Inc.0 sqrt( 9.1 fabs( 0.0 ) is 1.2 ) is 10.333 ) is 1.0 log10( 100.76 floor( x ) rounds x to the largest integer floor( 9.0 not less than x ceil( -9.

. Inc. All rights reserved.9 3.4  Functions ± Modularize a program ± Software reusability Functions  Call function multiple times  Local variables ± Known only in the function in which they are defined ± All variables declared in function definitions are local variables  Parameters ± Local variables passed to function when called ± Provide outside information  2003 Prentice Hall.

All rights reserved. passes back result  2003 Prentice Hall. .5 Function Definitions  Function prototype ± Tells compiler argument type and return type of function ± int square( int ). Inc.10 3. ± Parentheses an operator used to call function  Pass argument x  Function gets its own copy of arguments ± After finished.  Function takes an int and returns an int ± Explained in more detail later  Calling/invoking a function ± square(x).

All rights reserved.5 Function Definitions  Format for function definition return-value-type function-name( parameter-list ) { declarations and statements } ± Parameter list  Comma separated list of arguments ± Data type needed for each argument  If no arguments. . Inc. use void or leave blank ± Return-value-type  Data type of result returned (use void if nothing returned)  2003 Prentice Hall.11 3.

use return. ± Function ends when reaches right brace  Control goes to caller  Functions cannot be defined inside other functions  Next: program examples  2003 Prentice Hall.5 Function Definitions  Example function int square( int y ) { return y * y. and control goes to function¶s caller  If no data to return. All rights reserved. Inc.12 3. . }  return keyword ± Returns data.

All rights reserved. . return 0.cpp (1 of 2) Function prototype: specifies data types of arguments and return values. // 13 Outline fig03_03. Inc. it returns the result. and returns function prototype an int. square expects and int. int square( int ). When // square of x each time done.cpp // Creating and using a programmer-defined function. #include <iostream> using std::cout. } // end main // indicates successful termination  2003 Prentice Hall. 3.3: fig03_03. x <= 10. using std::endl. // function call cout << endl. for ( int x = 1.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 // Fig. x++ ) cout << square( x ) << " ". int main() { Parentheses () cause // loop 10 times and calculate and output function to be called.

Returns y * y. . All rights reserved.  2003 Prentice Hall.cpp (2 of 2) fig03_03.cpp output (1 of 1) 1 4 9 16 25 36 49 64 81 100 Definition of square. or y squared.23 24 25 26 27 28 // square function definition returns square of an integer int square( int y ) // y is a copy of argument to function { return y * y. y is a copy of the argument passed. Inc. // returns square of y as an int } // end function square 14 Outline fig03_03.

using std::endl.cpp (1 of 2) Function maximum takes 3 arguments (all double) and returns a double. double number2. Inc.4: fig03_04. . // number1. double. 3. return 0. // indicates successful termination  2003 Prentice Hall. #include <iostream> using std::cout. cout << "Enter three floating-point numbers: ". double number3. 15 Outline fig03_04. number2 and number3 are arguments to // the maximum function call cout << "Maximum is: " << maximum( number1. cin >> number1 >> number2 >> number3. double ). double maximum( double. number2. using std::cin. // function prototype int main() { double number1.cpp // Finding the maximum of three floating-point numbers. number3 ) << endl. All rights reserved.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 // Fig.

31 88.1 3.99 Maximum is: 88.22 Maximum is: 3. . double z ) { double max = x.333 Enter three floating-point numbers: 27.32 37. // x. Inc.333 2. // if y is larger.32 Enter three floating-point numbers: 1. // assume x is largest if ( y > max ) max = y. fig03_04. return max. double y. if ( z > max ) max = z.cpp (2 of 2) fig03_04.1928 Maximum is: 99.cpp output (1 of 1) // function maximum definition.26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 16 } // end main Outline Comma separated list for multiple parameters. y and z are parameters double maximum( double x.9 14. // assign y to max // if z is larger. // assign z to max // max is largest value } // end function maximum Enter three floating-point numbers: 99.3 27.99  2003 Prentice Hall. All rights reserved.

6 Function Prototypes  Function prototype contains ± Function name ± Parameters (number and data type) ± Return type (void if returns nothing) ± Only needed if function definition after function call  Prototype must match function definition ± Function prototype double maximum( double. double.17 3. . double y. double ). Inc. double z ) { « }  2003 Prentice Hall. All rights reserved. ± Definition double maximum( double x.

double.0) ± Conversion rules  Arguments usually converted automatically  Changing from double to int can truncate data ± 3. . Inc. double ).6 Function Prototypes  Function signature ± Part of prototype with name and parameters  double maximum( double. All rights reserved.18 3.4 to 3 ± Mixed type goes to highest type (promotion)  Int * double  2003 Prentice Hall.  Argument Coercion cout << sqrt(4) Function signature ± Force arguments to be of proper type  Converting int (4) to double (4.

3.5 Pro m o tio n hie ra rc hy fo r b uilt-in d a ta typ es.19 3. . All rights reserved.  2003 Prentice Hall. Inc. true becomes 1) bool Fig .6 Da ta typ es long double Function Prototypes double float (synonymous with unsigned long) unsigned long int (synonymous with long) long int (synonymous with unsigned) unsigned int int (synonymous with unsigned short) unsigned short int (synonymous with short) short int unsigned char char (false becomes 0.

8 Random Number Generation  rand function (<cstdlib>) ± i = rand().  ³Rand() % 6´ generates a number between 0 and 5 (scaling)  ³+ 1´ makes the range 1 to 6 (shift) ± Next: program to roll dice  2003 Prentice Hall. ± Generates unsigned integer between 0 and RAND_MAX (usually 32767)  Scaling and shifting ± Modulus (remainder) operator: %  10 % 3 is 1  x % y is between 0 and y ± 1 ± Example i = rand() % 6 + 1. All rights reserved.21 3. . Inc.

cpp // Shifted. All rights reserved. counter <= 20. begin new line of output if ( counter % 5 == 0 ) cout << endl. scaled integers produced by 1 + rand() % 6. counter++ ) { // pick random number from 1 to 6 and output cout << setw( 10 ) << ( 1 + rand() % 6 ).7: fig03_07. 3. of rand() scaled and shifted to be a number it between 1 and 6. #include <cstdlib> // contains function prototype for rand 22 Outline fig03_07. #include <iostream> using std::cout. } // end for structure  2003 Prentice Hall.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 // Fig. using std::endl. #include <iomanip> using std::setw. // if counter divisible by 5.cpp (1 of 2) int main() { // loop 20 times Output for ( int counter = 1. Inc. .

. Inc.cpp output (1 of 1)  2003 Prentice Hall.27 28 29 30 23 return 0. All rights reserved. } // end main 6 5 6 6 6 1 6 2 5 1 2 3 5 5 4 4 6 3 2 1 // indicates successful termination Outline fig03_07.cpp (2 of 2) fig03_07.

8  Next Random Number Generation ± Program to show distribution of rand() ± Simulate 6000 rolls of a die ± Print number of 1¶s. . rolled ± Should be roughly 1000 of each  2003 Prentice Hall. All rights reserved. Inc.24 3. etc. 2¶s. 3¶s.

8: fig03_08. int frequency4 = 0. int face.cpp (1 of 3) int main() { int frequency1 = 0. int frequency3 = 0. using std::endl. // represents one roll of the die  2003 Prentice Hall. #include <cstdlib> // contains function prototype for rand 25 Outline fig03_08. . #include <iostream> using std::cout. Inc. #include <iomanip> using std::setw. All rights reserved. 3. int frequency2 = 0. int frequency6 = 0. int frequency5 = 0.cpp // Roll a six-sided die 6000 times.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 // Fig.

case 5: // rolled 5 ++frequency5. roll++ ) { face = 1 + rand() % 6. 26 Outline fig03_08. case 4: // rolled 4 ++frequency4. break. break. case 3: // rolled 3 ++frequency3. roll <= 6000. Inc.24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 // loop 6000 times and summarize results for ( int roll = 1. case 2: // rolled 2 ++frequency2.cpp (2 of 3)  2003 Prentice Hall. break. All rights reserved. break. // random number from 1 to 6 // determine face value and increment appropriate counter switch ( face ) { case 1: // rolled 1 ++frequency1. break. .

This is a matter of good coding style tabular format << "Frequency" ) << frequency1 ) << frequency2 ) << frequency3 ) << frequency4 ) << frequency5 ) << frequency6 << endl. break. Inc. } // end main Default case included even though it should never be reached. All rights reserved. .50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 27 case 6: // rolled 6 ++frequency6. } // end switch } // end for Outline fig03_08. // indicates successful termination  2003 Prentice Hall. default: // invalid value cout << "Program should never get here!".cpp (3 of 3) // display results in cout << "Face" << setw( 13 ) << "\n 1" << setw( 13 << "\n 2" << setw( 13 << "\n 3" << setw( 13 << "\n 4" << setw( 13 << "\n 5" << setw( 13 << "\n 6" << setw( 13 return 0.

All rights reserved. Inc. .Face 1 2 3 4 5 6 Frequency 1003 1017 983 994 1004 999 28 Outline fig03_08.cpp output (1 of 1)  2003 Prentice Hall.

. Inc.29 3. All rights reserved.8 Random Number Generation  Calling rand() repeatedly ± Gives the same sequence of numbers  Pseudorandom numbers ± Preset sequence of "random" numbers ± Same sequence generated whenever program run  To get different random sequences ± Provide a seed value  Like a random starting point in the sequence  The same seed will give the same sequence ± srand(seed).  <cstdlib>  Used before rand() to set the seed  2003 Prentice Hall.

// seed random number generator Setting the seed with srand().  2003 Prentice Hall. 3. // contains prototypes for functions srand and rand #include <cstdlib> // main function begins program execution int main() { unsigned seed. All rights reserved. cin >> seed.cpp // Randomizing die-rolling program. #include <iostream> using std::cout. . srand( seed ). 30 Outline fig03_09. using std::cin. #include <iomanip> using std::setw. Inc. using std::endl.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 // Fig.cpp (1 of 2) cout << "Enter seed: ".9: fig03_09.

counter++ ) { // pick random number from 1 to 6 and output it cout << setw( 10 ) << ( 1 + rand() % 6 ). All rights reserved.cpp output (1 of 1) rand() gives the same sequence if it has the same initial seed. // if counter divisible by 5.cpp (2 of 2) fig03_09. 6 6 2 4 Enter seed: 67 6 1 Enter seed: 432 4 3 Enter seed: 67 6 1 1 6 4 1 6 1 3 5 1 4 6 2 1 6 4 1 6 6 2 4  2003 Prentice Hall. begin new line of output if ( counter % 5 == 0 ) cout << endl. counter <= 10. Inc. } // end for return 0. .25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 // loop 10 times for ( int counter = 1. } // end main // indicates successful termination 31 Outline fig03_09.

. All rights reserved. Inc.8 Random Number Generation  Can use the current time to set the seed ± No need to explicitly set seed every time ± srand( time( 0 ) ).  <ctime>  Returns current time in seconds  General shifting and scaling ± Number = shiftingValue + rand() % scalingFactor ± shiftingValue = first number in desired range ± scalingFactor = width of desired range  2003 Prentice Hall. ± time( 0 ).32 3.

Status enumVar.33 3. enumVar = WON. constant2«}.9 Example: Game of Chance and Introducing enum  Enumeration ± Set of integers with identifiers enum typeName {constant1. ± Constants start at 0 (default). WON. All rights reserved. LOST}. incremented by 1 ± Constants need unique names ± Cannot assign integer to enumeration variable  Must use a previously defined enumeration type  Example enum Status {CONTINUE. . Inc. // cannot do enumVar = 1  2003 Prentice Hall.

JUN. NOV. 9. 10  Value becomes player's "point"  Player must roll his point before rolling 7 to win  2003 Prentice Hall. ± Starts at 1. 8. 5. . MAR.9 Example: Game of Chance and Introducing enum  Enumeration constants can have preset values enum Months { JAN = 1. AUG. DEC}. increments by 1  Next: craps simulator ± ± ± ± Roll two dice 7 or 11 on first throw: player wins 2. APR. 6.34 3. MAY. All rights reserved. SEP. Inc. OCT. or 12 on first throw: player loses 4. JUL. 3. FEB.

{ // enumeration constants represent game status enum Status { CONTINUE. All rights reserved. int rollDice( void ). 3.cpp // Craps. WON.10: fig03_10. // contains function prototypes for functions srand and rand #include <cstdlib> #include <ctime> // contains 35 Outline fig03_10. Enumeration to keep track of // can contain CONTINUE. using std::endl. // function prototype int main() the current game.cpp (1 of 5) Function to roll 2 dice and return for function time prototype the result as an int.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 // Fig. . LOST }. #include <iostream> using std::cout. Inc. int sum. int myPoint. WON or LOST  2003 Prentice Hall. Status gameStatus.

cpp (2 of 5) // determine game status and switch ( sum ) { // win on first roll case 7: case 11: gameStatus = WON. . Inc. determines outcome based on point based on sum of dice die roll.25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 // randomize random number generator using current time srand( time( 0 ) ). sum = rollDice(). break. // lose on first roll case 2: case 3: case 12: gameStatus = LOST. All rights reserved. // first switch the dice roll of statement 36 Outline fig03_10.  2003 Prentice Hall. break.

else if ( sum == 7 ) gameStatus = LOST. Inc.. All rights reserved. // optional } // end switch // while game not complete .46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 // remember point default: gameStatus = CONTINUE. } // end while 37 Outline fig03_10.cpp (3 of 5) // win by making point // lose by rolling 7  2003 Prentice Hall. myPoint = sum. cout << "Point is " << myPoint << endl. . break. while ( gameStatus == CONTINUE ) { sum = rollDice().. // roll dice again // determine game status if ( sum == myPoint ) gameStatus = WON.

// indicates successful termination 38 Outline fig03_10. calculate sum and int rollDice( void ) { int die1. Inc. else cout << "Player loses" << endl. int workSum. workSum = die1 + die2. so has void in the parameter list. return 0. All rights reserved. } // end main // roll dice.cpp (4 of 5) Function rollDice takes no arguments. display results // pick random die1 value // pick random die2 value // sum die1 and die2  2003 Prentice Hall. int die2.68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 // display won or lost message if ( gameStatus == WON ) cout << "Player wins" << endl. die2 = 1 + rand() % 6. die1 = 1 + rand() % 6. .

cpp output (1 of 2)  2003 Prentice Hall. All rights reserved.89 90 91 92 93 94 95 // display results of this roll cout << "Player rolled " << die1 << " + " << die2 << " = " << workSum << endl.cpp (5 of 5) Player rolled 2 + 5 = 7 Player wins Player rolled 6 + 6 = 12 Player loses Player rolled Point is 6 Player rolled Player rolled Player rolled Player rolled Player wins 3 + 3 = 6 5 4 2 1 + + + + 3 5 1 5 = = = = 8 9 3 6 fig03_10. } // end function rollDice // return sum of dice 39 Outline fig03_10. return workSum. . Inc.

. Inc. All rights reserved.cpp output (2 of 2)  2003 Prentice Hall.Player rolled Point is 4 Player rolled Player rolled Player rolled Player rolled Player rolled Player rolled Player rolled Player rolled Player loses 1 + 3 = 4 4 2 6 2 2 1 4 4 + + + + + + + + 6 4 4 3 4 1 4 3 = = = = = = = = 10 6 10 5 6 2 8 7 40 Outline fig03_10.

10 Storage Classes  Variables have attributes ± Have seen name. Inc. All rights reserved. type. which files can use it  2003 Prentice Hall. value ± Storage class  How long variable exists in memory ± Scope  Where variable can be referenced in program ± Linkage  For multiple-file program (see Ch. . size. 6).41 3.

Inc. not both  register int counter = 1. .  2003 Prentice Hall. compiler optimizes ± Specify either register or auto.10 Storage Classes  Automatic storage class ± Variable created when program enters its block ± Variable destroyed when program leaves block ± Only local variables of functions can be automatic  Automatic by default  keyword auto explicitly declares automatic ± register keyword  Hint to place variable in high-speed register  Good for often-used items (loop counters)  Often unnecessary.42 3. All rights reserved.

Inc. .43 3. scope rules still apply (more later)  static keyword ± Local variables in function ± Keeps value between function calls ± Only known in own function  extern keyword ± Default for global variables/functions  Globals: defined outside of a function block ± Known in any function that comes after it  2003 Prentice Hall. All rights reserved.10 Storage Classes  Static storage class ± Variables exist for entire program  For functions. name exists for entire program ± May not be accessible.

g. function definitions and prototypes  Function scope ± Can only be referenced inside defining function ± Only labels. . identifiers with a colon (case:)  2003 Prentice Hall.44 3. known in all functions ± Global variables. All rights reserved.11 Scope Rules  Scope ± Portion of program where identifier can be used  File scope ± Defined outside a function. Inc.. e.

45 3. function parameters ± static variables still have block scope  Storage class separate from scope  Function-prototype scope ± Parameter list of prototype ± Names in prototype optional  Compiler ignores ± In a single prototype.11 Scope Rules  Block scope ± Begins at declaration. All rights reserved. name can be used once  2003 Prentice Hall. ends at right brace }  Can only be referenced in this range ± Local variables. . Inc.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

// Fig. 3.12: fig03_12.cpp // A scoping example. #include <iostream> using std::cout; using std::endl; void useLocal( void ); // function prototype Declared outside of function; void useStaticLocal( void ); // function prototype globalfunction with file void useGlobal( void ); // variable prototype

46

Outline
fig03_12.cpp (1 of 5)

scope.
int x = 1; int main() { int x = 5; // global variable

Local variable with function scope.
// local variable to main

cout << "local x in main's outer scope new" << x giving x Create a is block, << endl; { // start new scope int x = 7; cout << "local x in main's inner scope is " << x << endl; } // end new scope

block scope. When the block ends, this x is destroyed.

 2003 Prentice Hall, Inc.

27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

47
cout << "local x in main's outer scope is " << x << endl; useLocal(); useStaticLocal(); useGlobal(); useLocal(); useStaticLocal(); useGlobal(); // // // // // // useLocal has local x useStaticLocal has static local x useGlobal uses global x useLocal reinitializes its local x static local x retains its prior value global x also retains its value

Outline
fig03_12.cpp (2 of 5)

cout << "\nlocal x in main is " << x << endl; return 0; } // end main // indicates successful termination

 2003 Prentice Hall, Inc.

43 44 45 46 47 48 49 50 51 52 53 54 55

// useLocal reinitializes local variable x during each call void useLocal( void ) { int x = 25; // initialized each time useLocal is called cout << << ++x; cout << << endl << "local x is Automatic variable (local " << x variable endl; " on entering useLocal" << of function). This

48

Outline
fig03_12.cpp (3 of 5)

is destroyed when the function "local x is " << x exits, and reinitialized when " on exiting useLocal" function begins. the << endl;

} // end function useLocal

 2003 Prentice Hall, Inc.

value of x is saved // between calls to this function void useStaticLocal( void ) { // initialized only first time useStaticLocal is called static int x = 50. All rights reserved.  2003 Prentice Hall. it is initialized " on exiting useStaticLocal" << endl. Inc. } // end function useStaticLocal only once. and retains its value between function calls. cout << << endl << "local static x is " << x " on entering useStaticLocal" << endl. . cout << << ++x.cpp (4 of 5) Static local variable of "local static x is " << x function.56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 // useStaticLocal initializes static local variable x only the // first time the function is called. 49 Outline fig03_12.

fig03_12. All rights reserved. Inc.cpp << " on exiting useGlobal" << endl.72 73 74 75 76 77 78 79 80 81 // useGlobal modifies global variable x during each call void useGlobal( void ) { cout << endl << "global x is " << x This function does not declare fig03_12. . It uses the (5 of 5) x *= 10. } // end function useGlobal 50 Outline output (1 of 2) local x in main's outer scope is 5 local x in main's inner scope is 7 local x in main's outer scope is 5 local x is 25 on entering useLocal local x is 26 on exiting useLocal local static x is 50 on entering useStaticLocal local static x is 51 on exiting useStaticLocal global x is 1 on entering useGlobal global x is 10 on exiting useGlobal  2003 Prentice Hall.cpp << " on entering useGlobal" << endl. any variables. global x declared in the cout << "global x is " << x beginning of the program.

51 local x is 25 on entering useLocal local x is 26 on exiting useLocal local static x is 51 on entering useStaticLocal local static x is 52 on exiting useStaticLocal global x is 10 on entering useGlobal global x is 100 on exiting useGlobal local x in main is 5 Outline fig03_12. All rights reserved. . Inc.cpp output (2 of 2)  2003 Prentice Hall.

12 Recursion  Recursive functions ± Functions that call themselves ± Can only solve a base case  If not base case ± Break problem into smaller problem(s) ± Launch new copy of function to work on the smaller problem (recursive call/recursive step)  Slowly converges towards base case  Function makes call to itself inside the return statement ± Eventually base case gets solved  Answer works way back up. . Inc. All rights reserved.52 3. solves entire problem  2003 Prentice Hall.

53 3.12 Recursion  Example: factorial n! = n * ( n ± 1 ) * ( n ± 2 ) * « * 1 ± Recursive relationship ( n! = n * ( n ± 1 )! ) 5! = 5 * 4! 4! = 4 * 3!« ± Base case (1! = 0! = 1)  2003 Prentice Hall. All rights reserved. Inc. .

i++ ) cout << setw( 2 ) << i << "! = " << factorial( i ) << endl. #include <iostream> using std::cout. return 0. // function prototype int main() { // Loop 10 times. During each iteration. Inc. for ( int i = 0. . 3. #include <iomanip> using std::setw. using std::endl. 54 Outline fig03_14. calculate // factorial( i ) and display result.cpp (1 of 2) Data type unsigned long can hold an integer from 0 to 4 billion.cpp // Recursive factorial function.14: fig03_14. unsigned long factorial( unsigned long ). } // end main // indicates successful termination  2003 Prentice Hall. i <= 10.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 // Fig. All rights reserved.

// recursive step else return number * factorial( number .cpp output (1 of 1)  2003 Prentice Hall. All other { cases must be split up // base case if ( number <= 1 ) (recursive step). } // end function factorial = = = = = = = = = = = 1 1 2 6 24 120 720 5040 40320 362880 3628800 Outline fig03_14.25 26 27 28 29 30 31 32 33 34 35 36 37 0! 1! 2! 3! 4! 5! 6! 7! 8! 9! 10! 55 // recursive definition of function factorial The base unsigned long factorial( unsigned long number ) case occurs when we have 0! or 1!.1 ). . All rights reserved. Inc.cpp (2 of 2) fig03_14. return 1.

.13 Example Using Recursion: Fibonacci Series  Fibonacci series: 0..1 ) + fibonacci( n ± 2 ). Inc. 2. else return fibonacci( n . 1. 8. 1. 3.. }  2003 Prentice Hall. ± Each number sum of two previous ones ± Example of a recursive formula:  fib(n) = fib(n-1) + fib(n-2)  C++ code for Fibonacci function long fibonacci( long n ) { if ( n == 0 || n == 1 ) // base case return n. All rights reserved.56 3. 5.

13 Example Using Recursion: Fibonacci Series f( 3 ) return f( 2 ) + f( 1 ) return f( 1 ) + f( 0 ) return 1 return 1 return 0  2003 Prentice Hall.57 3. . All rights reserved. Inc.

.2 ). Inc.1 ) + fibonacci( n .  Do not know which one executed first ± C++ does not specify ± Only &&.58 3. || and ?: guaranteed left-to-right evaluation  Recursive function calls ± Each level of recursion doubles the number of function calls  30th number = 2^30 ~ 4 billion function calls ± Exponential complexity  2003 Prentice Hall. All rights reserved.13 Example Using Recursion: Fibonacci Series  Order of operations ± return fibonacci( n .

unsigned long fibonacci( unsigned long ). // calculate fibonacci value for number input by user result = fibonacci( number ). Inc. Thus. cin >> number. 3. return 0. using std::cin. #include <iostream> using std::cout. // obtain integer from user cout << "Enter an integer: ". // display result cout << "Fibonacci(" << number << ") = " << result << endl. using std::endl.cpp The Fibonacci numbers get (1 of 2) large very quickly.cpp // Recursive fibonacci function.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 // Fig.15: fig03_15.  2003 Prentice Hall. number. // int main() { unsigned long result. and are all non-negative integers. // indicates successful termination 59 Outline fig03_15. All rights reserved. we use the unsigned function prototype long data type. .

2 ).1 ) + fibonacci( n .cpp (2 of 2) fig03_15. All rights reserved.cpp output (1 of 2) Enter an integer: 0 Fibonacci(0) = 0 Enter an integer: 1 Fibonacci(1) = 1 Enter an integer: 2 Fibonacci(2) = 1 Enter an integer: 3 Fibonacci(3) = 2  2003 Prentice Hall. . } // end function fibonacci Outline fig03_15. Inc.26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 60 } // end main // recursive definition of function fibonacci unsigned long fibonacci( unsigned long n ) { // base case if ( n == 0 || n == 1 ) return n. // recursive step else return fibonacci( n .

cpp output (2 of 2)  2003 Prentice Hall. . All rights reserved.Enter an integer: 4 Fibonacci(4) = 3 Enter an integer: 5 Fibonacci(5) = 5 Enter an integer: 6 Fibonacci(6) = 8 Enter an integer: 10 Fibonacci(10) = 55 Enter an integer: 20 Fibonacci(20) = 6765 Enter an integer: 30 Fibonacci(30) = 832040 Enter an integer: 35 Fibonacci(35) = 9227465 61 Outline fig03_15. Inc.

. All rights reserved. Inc. Iteration  Repetition ± Iteration: explicit loop ± Recursion: repeated function calls  Termination ± Iteration: loop condition fails ± Recursion: base case recognized  Both can have infinite loops  Balance between performance (iteration) and good software engineering (recursion)  2003 Prentice Hall.14 Recursion vs.62 3.

15 Functions with Empty Parameter Lists  Empty parameter lists ± void or leave parameter list empty ± Indicates function takes no arguments ± Function print takes no arguments and returns no value  void print(). . Inc. All rights reserved.  void print( void ).  2003 Prentice Hall.63 3.

int main() { function1(). All rights reserved.18: fig03_18.cpp // Functions that take no arguments. 3. void function1(). . #include <iostream> using std::cout. using std::endl.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 // Fig.cpp (1 of 2) // call function1 with no arguments // call function2 with no arguments // indicates successful termination  2003 Prentice Hall. } // end main // function prototype // function prototype 64 Outline fig03_18. function2(). return 0. void function2( void ). Inc.

} // end function2 65 Outline fig03_18. Inc. } // end function1 // function2 uses a void parameter list to specify that // the function receives no arguments void function2( void ) { cout << "function2 also takes no arguments" << endl.cpp output (1 of 1) function1 takes no arguments function2 also takes no arguments  2003 Prentice Hall.20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 // function1 uses an empty parameter list to specify that // the function receives no arguments void function1() { cout << "function1 takes no arguments" << endl.cpp (2 of 2) fig03_18. All rights reserved. .

All rights reserved. Inc.66 3. } ± const tells compiler that function does not modify s  Discussed in chapters 6-7  2003 Prentice Hall.16 Inline Functions  Inline functions ± Keyword inline before function ± Asks the compiler to copy code into program instead of making function call  Reduce function-call overhead  Compiler can ignore inline ± Good for small. often-used functions  Example inline double cube( const double s ) { return s * s * s. .

inline double cube( const double side ) { return side * side * side. Definition of function // appears before function is called. First line of function definition acts as // the prototype. using std::endl. // Definition of inline function cube. // calculate cube } // end function cube 67 Outline fig03_19.cpp (1 of 2)  2003 Prentice Hall. // the volume of a cube. Inc. so a function prototype // is not required.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 // Fig.cpp // Using an inline function to calculate. All rights reserved. . 3. using std::cin. #include <iostream> using std::cout.19: fig03_19.

875  2003 Prentice Hall.cpp output (1 of 1) Enter the side length of your cube: 3.cpp (2 of 2) fig03_19.5 is 42.20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 int main() { cout << "Enter the side length of your cube: ". return 0. cin >> sideValue. Inc. .5 Volume of cube with side 3. } // end main // indicates successful termination 68 Outline fig03_19. double sideValue. // calculate cube of sideValue and display result cout << "Volume of cube with side " << sideValue << " is " << cube( sideValue ) << endl. All rights reserved.

Inc. All rights reserved.69 3. .17 References and Reference Parameters  Call by value ± Copy of data passed to function ± Changes to copy do not change original ± Prevent unwanted side effects  Call by reference ± Function can directly access data ± Changes affect original  2003 Prentice Hall.

Inc. . original can now be changed  2003 Prentice Hall.70 3. All rights reserved.17 References and Reference Parameters  Reference parameter ± Alias for argument in function call  Passes parameter by reference ± Use & after data type in prototype  void myFunction( int &data )  Read ³data is a reference to an int´ ± Function call format the same  However.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

// Fig. 3.20: fig03_20.cpp // Comparing pass-by-value and pass-by-reference // with references. #include <iostream> using std::cout; using std::endl; int squareByValue( int ); void squareByReference( int & ); int main() { int x = 2; int z = 4; // demonstrate squareByValue cout << "x = " << x << " before squareByValue\n"; cout << "Value returned by squareByValue: " << squareByValue( x ) << endl; cout << "x = " << x << " after squareByValue\n" << endl;

71

Outline
fig03_20.cpp (1 of 2)

Notice the & operator, indicating pass-by-reference.
// function prototype // function prototype

 2003 Prentice Hall, Inc.

23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

// demonstrate squareByReference cout << "z = " << z << " before squareByReference" << endl; squareByReference( z ); cout << "z = " << z << " after squareByReference" << endl; return 0; // indicates successful termination } // end main

72

Outline
fig03_20.cpp (2 of 2)

// // int squareByValue( int number ) { return number *= number; // caller's argument not modified } // end function squareByValue

Changes number, but original parameter (x) is not squareByValue multiplies number by itself, stores the modified. result in number and returns the new value of number

Changes numberRef, an
// squareByReference multiplies numberRef by itself and alias for the original // stores the result in the variable to which numberRef parameter. Thus, z is // refers in function main changed. void squareByReference( int &numberRef ) { numberRef *= numberRef; // caller's argument modified } // end function squareByReference

 2003 Prentice Hall, Inc.

x = 2 before squareByValue Value returned by squareByValue: 4 x = 2 after squareByValue z = 4 before squareByReference z = 16 after squareByReference

73

Outline
fig03_20.cpp output (1 of 1)

 2003 Prentice Hall, Inc.

74 3. . All rights reserved. compiler error ± Dangling reference  Reference to undefined variable  2003 Prentice Hall.17 References and Reference Parameters  Pointers (chapter 5) ± Another way to pass-by-refernce  References as aliases to other variables ± Refer to same variable ± Can be used within a function int count = 1. // declare integer variable count Int &cRef = count. Inc. // create cRef as an alias for count ++cRef. // increment count (using its alias)  References must be initialized when declared ± Otherwise.

using std::endl. #include <iostream> using std::cout. cout << "x = " << x << endl << "y = " << y << endl.cpp // References must be initialized. .cpp output (1 of 1) y declared as a reference to x. Inc. All rights reserved. 75 Outline fig03_21. return 0. cout << "x = " << x << endl << "y = " << y << endl.cpp (1 of 1) fig03_21.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 x y x y = = = = // Fig. int main() { int x = 3. // y refers to (is an alias for) x int &y = x. } // end main 3 3 7 7 // indicates successful termination  2003 Prentice Hall. 3. y = 7.21: fig03_21.

cpp (1 of 1) Uninitialized reference ± compiler error. y = 7.cpp 11: Reference variable 'y' must be initialized. Inc. All rights reserved. int &y. using std::endl. 76 Outline fig03_22.cpp // References must be initialized. } // end main // indicates successful termination Borland C++ command-line compiler error message: Error E2304 Fig03_22. cout << "x = " << x << endl << "y = " << y << endl. int main() { int x = 3. 3. .in function main() Microsoft Visual C++ compiler error message: D:\cpphtp4_examples\ch03\Fig03_22. return 0.cpp(11) : error C2530: 'y' : references must be initialized  2003 Prentice Hall.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 // Fig. #include <iostream> using std::cout.cpp output (1 of 1) cout << "x = " << x << endl << "y = " << y << endl.22: fig03_22. // Error: y must be initialized fig03_22.

y and z get defaults (rightmost) ± myFunction(3. All rights reserved. Inc. . int z = 3 ). 5)  x = 3. y = 5 and z gets default  2003 Prentice Hall.77 3. global variables. or function calls  Set defaults in function prototype int myFunction( int x = 1. ± myFunction(3)  x = 3. rightmost go to their defaults ± Default values  Can be constants.18 Default Arguments  Function call with omitted parameters ± If not enough parameters. int y = 2.

int width = 1. using std::endl. default height cout << "\n\nThe volume of a box with length 10.  2003 Prentice Hall. #include <iostream> using std::cout. 78 Outline Set defaults in function prototype.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 // Fig.cpp (1 of 2) // function prototype that specifies default arguments int boxVolume( int length = 1. 5 ). All rights reserved. fig03_23. // specify length and width. int main() { // no arguments--use default values for all dimensions cout << "The default box volume is: " << boxVolume(). 3. .\n" << "width 5 and height 1 is: " << boxVolume( 10. // specify length. default width and height cout << "\n\nThe volume of a box with length 10. Inc. Function calls with some parameters missing ± the rightmost parameters get their defaults. int height = 1 ).cpp // Using default arguments.23: fig03_23.\n" << "width 1 and height 1 is: " << boxVolume( 10 ).

. int height ) { return length * width * height.\n" << "width 5 and height 2 is: " << boxVolume( 10. int width. All rights reserved. width 1 and height 1 is: 10 The volume of a box with length 10. width 5 and height 1 is: 50 The volume of a box with length 10.cpp output (1 of 1) The default box volume is: 1 The volume of a box with length 10. } // end function boxVolume // indicates successful termination 79 Outline fig03_23. width 5 and height 2 is: 100  2003 Prentice Hall. 5. } // end main // function boxVolume calculates the volume of a box int boxVolume( int length.24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 // specify all arguments cout << "\n\nThe volume of a box with length 10. Inc. return 0.cpp (2 of 2) fig03_23. 2 ) << endl.

19 Unitary Scope Resolution Operator  Unary scope resolution operator (::) ± Access global variable if local variable has same name ± Not needed if names are different ± Use ::variable  y = ::x + 3. .80 3. Inc. ± Good to avoid using same names for locals and globals  2003 Prentice Hall. All rights reserved.

cpp // Using the unary scope resolution operator. // define global constant PI const double PI = 3.14159265358979. . This example will show the difference between float and double.cpp (1 of 2) Access the global PI with ::PI.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 // Fig. int main() { // define local constant PI const float PI = static_cast< float >( ::PI ). All rights reserved. Cast the global PI to a float for the local PI. return 0. // indicates successful termination  2003 Prentice Hall. Inc. 81 Outline fig03_24. 3. // display values of local and global PI constants cout << setprecision( 20 ) << " Local float value of PI = " << PI << "\nGlobal double value of PI = " << ::PI << endl. #include <iomanip> using std::setprecision. #include <iostream> using std::cout.24: fig03_24. using std::endl.

141592741012573242 Global double value of PI = 3. All rights reserved. Inc.cpp output (1 of 1) Borland C++ command-line compiler output: Local float value of PI = 3.1415927410125732 Global double value of PI = 3.14159265358979  2003 Prentice Hall. .26 27 82 } // end main Outline fig03_24.141592653589790007 Microsoft Visual C++ compiler output: Local float value of PI = 3.cpp (2 of 2) fig03_24.

25: fig03_25. return y * y. using std::endl.cpp (1 of 2) // function square for int values int square( int x ) { cout << "Called square with int argument: " << x << endl. . but the different parameters distinguish them. All rights reserved. } // end double version of function square Overloaded functions have the same name.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 // Fig. 84 Outline fig03_25. Inc. 3. } // end int version of function square // function square for double values double square( double y ) { cout << "Called square with double argument: " << y << endl. #include <iostream> using std::cout.  2003 Prentice Hall.cpp // Using overloaded functions. return x * x.

cpp (2 of 2) fig03_25.5 The square of integer 7 is 49 The square of double 7. // calls int version double doubleResult = square( 7.24 25 26 27 28 29 30 31 32 33 34 35 int main() { int intResult = square( 7 ).cpp output (1 of 1) (int or double). } // end main // indicates successful termination Called square with int argument: 7 Called square with double argument: 7.25  2003 Prentice Hall.5 is " << doubleResult based upon the argument << endl. return 0. Inc.5 ). All rights reserved. 85 Outline fig03_25. // calls double version cout << "\nThe square of integer 7 is " << intResult The proper function is called << "\nThe square of double 7.5 is 56. .

cpp // Name mangling. float b.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 // Fig. int *d ) { // empty function body } 86 Outline fig03_26. float. } // function that receives arguments of types // int. char and int * void nothing1( int a.cpp (1 of 2)  2003 Prentice Hall. char c. Inc. // function square for int values int square( int x ) { return x * x. All rights reserved.26: fig03_26. 3. . } // function square for double values double square( double y ) { return y * y.

cpp output (1 of 1) // indicates successful termination Mangled names produced in assembly language. double *d ) { return 0. c is char. } int main() { return 0. int. . d is double.cpp (2 of 2) fig03_26. float *c. etc. int b.23 24 25 26 27 28 29 30 31 32 33 34 // function that receives arguments of types // char. Inc. i is int. \$q separates the function name from its parameters. float * and double * char *nothing2( char a. All rights reserved. _main @nothing2\$qcipfpd @nothing1\$qifcpi @square\$qd @square\$qi  2003 Prentice Hall. pf is a pointer to a float. } // end main 87 Outline fig03_26.

. All rights reserved. Inc. int) or user-defined types  Specify arguments types.e. return types. declare variables ± Function definition like normal. .21 Function Templates  Compact way to make overloaded functions ± Generate separate function for different data types  Format ± Begin with keyword template ± Formal type parameters in brackets <>  Every type parameter preceded by typename or class (synonyms)  Placeholders for built-in types (i. except formal types used  2003 Prentice Hall.88 3.

Inc. all T's become ints int x.89 3. T replaced by real type  If int. int y = square(x). used as parameter type  Above function returns variable of same type as parameter ± In function call. All rights reserved. } ± T is a formal type.21 Function Templates  Example template < class T > // or template< typename T > T square( T value1 ) { return value1 * value1. .  2003 Prentice Hall.

if ( value3 > max ) max = value3. . All rights reserved. T value2.cpp (1 of 3) // definition of function template < class T > // or template < typename T > T maximum( T value1. Inc. using std::endl. if ( value2 > max ) max = value2.cpp // Using a function template. using std::cin.  2003 Prentice Hall. 90 Outline fig03_27. maximum expects all parameters to be of the same type. T value3 ) { T max = value1.27: fig03_27.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 // Fig. return max. } // end function template maximum Formal type parameter T placeholder for type of data to be tested template maximumby maximum. #include <iostream> using std::cout. 3.

double3. double2. cout << "\n\nInput three double values: ". double3 ). cin >> int1 >> int2 >> int3. Inc. double2. int2.cpp (2 of 3) maximum called with various data types.  2003 Prentice Hall. int3. All rights reserved. // demonstrate maximum with double values double double1. int3 ). 91 Outline fig03_27. int2. cin >> double1 >> double2 >> double3. // invoke double version of maximum cout << "The maximum double value is: " << maximum( double1. . cout << "Input three integer values: ".25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 int main() { // demonstrate maximum with int values int int1. // invoke int version of maximum cout << "The maximum integer value is: " << maximum( int1.

Inc. cin >> char1 >> char2 >> char3. All rights reserved.cpp output (1 of 1) Input three integer values: 1 2 3 The maximum integer value is: 3 Input three double values: 3.3 2.1 The maximum double value is: 3. char3 ) << endl. .47 48 49 50 51 52 53 54 55 56 57 58 59 60 // demonstrate maximum with char values char char1. char2. return 0. char3.3 Input three characters: A C B The maximum character value is: C  2003 Prentice Hall. cout << "\n\nInput three characters: ". } // end main // indicates successful termination 92 Outline fig03_27. // invoke char version of maximum cout << "The maximum character value is: " << maximum( char1. char2.2 1.cpp (3 of 3) fig03_27.