Functionsce.sharif.edu/courses/98-99/1/ce153-8/resources/root/Slides/Lecture9_1… · • Function definition format return-value-type function-name( parameter-list ) {declarations

$Page 1: Functionsce.sharif.edu/courses/98-99/1/ce153-8/resources/root/Slides/Lecture9_1… · • Function definition format return-value-type function-name( parameter-list ) {declarations$
Sharif University of TechnologyDepartment of Computer Engineering 1

Functions

Input and Output – Lecture 4


Outline• Standard (Predefined) Functions

• User-Defined Functions

• Function Definition

• Flow of Execution

• …



Introduction• Until now, we learned to develop simple

algorithms– Interactions, Mathematics, Decisions, and Loops

• Real problems: very complex– Calculator

– Games, MS Word, Firefox, …

• Cannot be developed at once– Divide the problem into smaller sub-problems

– Solve the sub-problems

– Put the solutions altogether to get the final solution



Introduction• Divide and conquer

– Construct a program from smaller pieces or components

• Smaller pieces sometimes called functions

– Each piece more manageable than the original program

Divide and Conquer



Modular programming• Solving a large and complex problem

• Design the overall algorithm

• Some portions are black-box

We know it does something

But we don't worry how

Later, we think about the black-boxes and

develop them

• Black-boxes are implemented by functions



Why?• Helps manage complexity

– Smaller blocks of code

– Easier to read

• Encourages re-use of code

– Within a particular program or across differentprograms

• Provides a layer of ‘abstraction’



Building Programs from Existing Programs

• Programmers seldom start off with an empty program.

• Reasons:

– Most basic or frequently used functions have been written by other programmers.

– If you write every function by yourself, the source code may be messy and hard to maintained.

– Less code, less bugs.

– Many existing libraries have tune the performance, and thus are more reliable and robust.



Functions - Mathematical View

32)( 2

xxxf

11 is )2(

113443)2(2)2()2( 2

f

f

f(2)? isWhat

)(xf2 11

X FunctionReturned

value



Function Input and Output



Functions• Every C program starts with main() function

• Functions could be

– Pre-defined library functions

• e.g., printf, sin, tan

– Programmer-defined functions

• e.g., my_printf, area

int main(){

…}



Pre-defined library functions • <math.h>

– Defines common mathematical functions– e.g. sin, cos. sqrt, pow

• <stdio.h>– Defines core input and output functions– e.g. printf, scanf

• <time.h>– Defines date and time handling functions– e.g. time, clock

• <stdlib.h>– Defines pseudo-random numbers generation functions– e.g. rand, srand

http://en.wikipedia.org/wiki/C_mathematical_functions

http://en.wikipedia.org/wiki/C_file_input/output

http://en.wikipedia.org/wiki/C_date_and_time_functions

http://en.wikipedia.org/wiki/C_mathematical_functionsstdlib.h



C mathematical functions• double fmod( double x, double y );

– Computes the remainder of the division operation x/y

• double exp( double arg );

– Computes the e (Euler's number, 2.7182818) raised to the given power arg

• double log( double arg );

– Computes the natural (base e) logarithm of arg

• double log10( double arg );

– Computes the common (base 10) logarithm of arg

• double sqrt( double arg );

– Computes square root of arg

http://en.cppreference.com/w/c/numeric/math/fmod

http://en.cppreference.com/w/c/numeric/math/exp

http://en.cppreference.com/w/c/numeric/math/log

http://en.cppreference.com/w/c/numeric/math/log10

http://en.cppreference.com/w/c/numeric/math/sqrt



C mathematical functions• double pow( double base, double exp);

– Computes the value of base raised to the power exp

• double sin( double arg );– Computes sine of arg (representing angle in radians)

• double cos( double arg );• Computes cosine of arg (representing angle in radians)

• double tan( double arg );– Computes tangent of arg (representing angle in radians)

• ...

http://en.cppreference.com/w/c/numeric/math/pow

http://en.cppreference.com/w/c/numeric/math/sin

http://en.cppreference.com/w/c/numeric/math/cos

http://en.cppreference.com/w/c/numeric/math/tan



An example#include <stdio.h>

#include <math.h>

int main(void)

{

double angle;

printf("Input angle in radians: \n");

scanf("%lf", &angle);

printf("The sine of the angle is %f\n", sin(angle) );

return 0;

}



An example#include <stdio.h>#include <math.h>

int main(void){

double x1,y1,x2,y2, dist;printf("Enter x1 y1 x2 y2 :");scanf("%lf %lf %lf %lf", &x1, &y1, &x2, &y2);

dist = sqrt(pow((x2-x1),2) + pow((y2-y1),2)); //

printf("Distance is %lf\n", dist);return 0;

}



Random numbers generation functions• int rand();

– Returns a uniformly distributed pseudo-random integral valuebetween 0 and RAND_MAX (0 and RAND_MAX included)

– RAND_MAX : Expands to an integer constant expression equal to themaximum value returned by the function rand(). This value isimplementation dependent.

• #define RAND_MAX 32767 /*implementation defined*/

– srand() should be called before any calls to rand() to initialize therandom number generator

• void srand( unsigned seed );– Initializes the built-in random number generator used to generate

values for rand() with the seed value seed

http://en.cppreference.com/w/c/numeric/random/rand

http://en.cppreference.com/w/c/numeric/random/RAND_MAX

http://en.cppreference.com/w/c/numeric/random/srand

http://en.cppreference.com/w/c/numeric/random/srand

http://en.cppreference.com/w/c/numeric/random/rand



An Example#include <stdio.h>#include <stdlib.h>

int main(void){

unsigned int seed; /* Declare variables. */int k;/* Get seed value from the user. */printf("Enter a positive integer seed value: \n");scanf("%u",&seed);srand(seed);

/* Generate and print ten random numbers. */printf("Random Numbers: \n");for (k=1; k<=10; k++)

printf("%i ", rand());printf("\n");

return 0; /* Exit program. */}



An Example#include <stdio.h>#include <stdlib.h>#include <time.h>

int main(void){

srand(time(0)); //use current time as seed for random generator /* Generate and print ten random numbers. */printf("Random Numbers: \n");for (k=1; k<=10; k++)

printf("%i ", rand());printf("\n");

return 0; /* Exit program. */}


Sharif University of TechnologyDepartment of Computer Engineering 19 19

Random Numbers in [a b]• Generate a random number [0 .. 7]

– x = rand() % 8;

• Generate a random number [10 ..17]

– x = 10 + rand() % 8;

• rand() % (b-a+1) + a;

2000 Prentice Hall, Inc.All rights reserved.

OutlineOutline

1. Initialize seed

2. Input value for seed

2.1 Use srand to

change random

sequence

2.2 Define Loop

3. Generate and

output random

numbers

1 /* Fig. 5.9: fig05_09.c

2 Randomizing die-rolling program */

3 #include <stdlib.h>

4 #include <stdio.h>

5

6 int main()

7 {

8 int i;

9 unsigned seed;

10

11 printf( "Enter seed: " );

12 scanf( "%u", &seed );

13 srand( seed );

14

15 for ( i = 1; i <= 10; i++ ) {

16 printf( "%10d", 1 + ( rand() % 6 ) );

17

18 if ( i % 5 == 0 )

19 printf( "\n" );

20 }

21

22 return 0;

23 } Enter seed: 867

2 4 6 1 6

1 1 3 6 2

Enter seed: 67

6 1 4 6 2

1 6 1 6 4



Time Example/*This is a simple program that illustrates that calling the time function at distinct

moments and noting the different times is a simple method of timing fragments of code:*/

#include <stdio.h>

#include <sys/types.h>

#include <time.h>

int main(){

int i;

time_t t1,t2;

time(&t1);

for (i=1;i<=300;++i)

printf("%d %d %d\n",i, i*i, i*i*i);

time(&t2);

printf("\n Time to do 300 squares and cubes= %d seconds\n", (int) t2-t1);

return 0;

}



Functions in C• Queries: Return a value

sin(), fabs()

• Commands: do some tasks, do not return any

value or we don’t use the value

printf(...)

scanf(...)



User- defined

OR

Programmer-defined function


Sharif University of TechnologyDepartment of Computer Engineering 24 24

Functions in CThree steps to use functions in C

Function prototype (declaration)

Introduce the function to compiler

Function definition

What the function does

Function call

Use the function



Functions - Definition Structure• Function 'header'

– Return data type (if any)

– Name• Descriptive

– Arguments (or parameter list)• Notice: data type and name

• Statements– Variable declaration– Operations– Return value (if any)

type function_name (type arg1, type arg2 ){

statements;}

double product(double x, double y)

{

double result;

result = x * y;

return result;

}

A function that calculates the product of two numbers



Function Definitions• Function definition format

return-value-type function-name( parameter-list ){

declarations and statements}

– Function-name: any valid identifier

– Return-value-type: data type of the result• void – indicates that the function returns nothing

– Parameter-list: comma separated list, declares parameters

• A type must be listed explicitly for each parameter

• No input: empty parameter list ()or void



Function Definitions• Function definition format (continued)

return-value-type function-name( parameter-list ){


– Declarations and statements: function body (block)

• Variables can be declared inside blocks

• Functions can not be defined inside other functions



Producing output

Query functions

Produce output

To produce an output

Declare output type

Generate the output by return

Function definition format (continued)return-value-type function-name( parameter-list ){




The return command• To generate a result by a function

– return <value>;

– return <expression>;

• Only one value can be returned

• If noting returned

• return;

• Or until reaches right brace



The return command• return finishes running the function

• Function can have multiple return– Only one of them runs each time

• The type of the returned value = the result type– Otherwise, cast



Functions that do not return a value

• Use the return type of void

– void functionName( DataType arg_1,…)

– void functionName()

– void functionName( void)



Function – An Example

• Write a function named 'sum'– sums two integers

– returns the sum

Steps1. Function header

• return data type• function name• argument list with data types

2. Statements in function definition• variable declaration• operations• return value

int sum_int(int x, int y)

{

int result;

result = x + y;

return result;

}



Function callCommand function

<function name> (inputs);

Query function

<variable> = <function name>(inputs);

Inputs should match by function definition

Functions are called by another function

Function call comes inside in a function



Examplevoid my_info(void){

printf("My name is Ahmad Ahmadi\n");

printf("My student number: 95222222\n");

}

int main(void){

my_info();

printf("--------------\n");

my_info();

return 0;

}



Function Call – An Example• If the function returns a value, then the returned value need to be

assigned to a variable so that it can be stored

int GetUserInput (void); /* function prototype*/int main(void) {

int input;input = GetUserInput( );return(0); /* return 0; */

}• However, it is perfectly okay (syntax wise) to just call the function without

assigning it to any variable if we want to ignore the returned value

• We can also call a function inside another functionprintf("User input is: %d", GetUserInput( ));



Using Functions• Let int f(double x, int a) be (the beginning

of) a declaration of a function.

• Then f(expr1, expr2) can be used,

N = f(pi*pow(r,2), b+c) + d;



• Let int f(double x, int a) be (the beginning


• Then f(expr1, expr2) can be used ,

N = f(pi*pow(r,2), b+c) + d;

Using Functions (continued)This is a parameter

This is an argumentThis is also an

argument



Definitions• Parameter:– a declaration of an identifier

within the '()' of a function declaration• Used within the body of the function as a variable of

that function

• Argument:– an expression passed when a function is called; becomes the initial value of the corresponding parameter






N = f(pi*pow(r,2), b+c) + d;

Using Functions (continued)






N = f(pi*pow(r,2), b+c) + d;

Using Functions (continued)

Function f is executed and

returns a value of type intResult of f is added to dSum is assigned to N



Formal and Actual Parameters• Formal parameter

– Variables declared in the formal list of the function header (written in function prototype &function definition)

• Actual parameter– Constants, variables, or expression in a function call that correspond to its formal parameter

• The number of actual parameters in a function call must be the same as thenumber of formal parameters in the function definition

• A one-to-one correspondence must occur among the actual and formalparameters. The first actual parameter must correspond to the first formalparameter and the second to the second formal parameter, an so on

• The type of each actual parameter must be the same as that of the correspondingformal parameter



An Example• If the function requires some arguments to be passed along, then the

arguments need to be listed in the bracket ( ) according to the specifiedorder

void Calc(int …, double …, char…, int…){…}

int main(void) {

int a, b;double c;char d;…Calc(a, c, d, b);

return (0);}

Function Call



Function Call – An Example#include <stdio.h>

//function prototype//global variable declaration

int main(void){

local variable declaration;statements;fn1( );fn2( );

return (0);}

void fn1(void){

local variable declaration;statements;

}

void fn2(void){

local variable declaration;statements;return;

}

1

2

3

4



Function prototype<output type> <functionname>(<input parameter

types>);

<output type> Queries: int, float,…

Command: void

<function name> is an identifier

<input parameter list> <type>, <type>, …

int, float, …

Void



Function declaration is optional if program is developed in a single filevoid my_info(void){

printf("My name is Ahmad Ahmadi\n");

printf("My student number: 95222222\n");

}

int main(void){

my_info();

printf("--------------\n");

my_info();

return 0;

}



Example

/* Function declaration */

void my_info(void);

int main(void){

printf("This is my info");

my_info(); /* Function call */

printf("=============");

return 0;

}

/* Function definition */

void my_info(void){

printf("Student name is Ali Hosseini\n");

printf("Student number: 9522222\n");

}



An Example#include <stdio.h>int calSum(int,int); /*function prototype*/

int main(void){

…..…..sum = calSum(num1,num2); /* function call */…..

}

int calSum(int val1, int val2) /*function header*/{

………………

}

Formal Parameters

Formal Parameters

Actual Parameters

Introduction to Functions CS-2301, B-Term 200948

Purposes of Function Prototype

• So compiler knows how to compile calls to that function, i.e.,– number and types of arguments– type of result

• As part of a “contract” between developer and programmer who uses the function

• As part of hiding details of how it works and exposing what it does.

• A function serves as a “black box.”

• Prototype only needed if function definition comes after use in program



An ExampleFunction prototype

– Like a variable declaration• Tells compiler that the function will be defined later• Note semicolon!!

Function definition– See previous slide– Note, NO semicolon

Function return– return statement terminates execution of the

current function– Control returns to the calling function– if return expression;

• then value of expression is returned as the value of the function call

• Only one value can be returned this way

Function call– main() is the 'calling function'– product() is the 'called function'– Control transferred to the function code– Code in function definition is executed

#include <stdio.h>

/* function prototype */

double product(double x, double y);

int main()

{

double var1 = 3.0, var2 = 5.0;

double ans;

ans = product(var1, var2);

printf("var1 = %.2f\n"

"var2 = %.2f\n",var1,var2);

printf("var1*var2 = %g\n", ans);

return 0;

}

/* function definition */


{

double result;

result = x * y;

return result;

}



Flow of Control• First statement executed in any program is the

first statement in the function main( )

• When another function called– logical control passed to first statement in that

function’s body

– program proceeds through sequence of statements within the function

• When last statement of function executed– control returns to where function was called

– control given to next command after the call

2000 Prentice Hall, Inc.All rights reserved.

OutlineOutline

1. Function prototype

(3 parameters)

2. Input values

2.1 Call function

3. Function definition

Program Output

1 /* Fig. 5.4: fig05_04.c

2 Finding the maximum of three integers */

3 #include <stdio.h>

4

5 int maximum( int, int, int ); /* function prototype */

6

7 int main()

8 {

9 int a, b, c;

10

11 printf( "Enter three integers: " );

12 scanf( "%d%d%d", &a, &b, &c );

13 printf( "Maximum is: %d\n", maximum( a, b, c ) );

14

15 return 0;

16 }

17

18 /* Function maximum definition */

19 int maximum( int x, int y, int z )

20 {

21 int max = x;

22

23 if ( y > max )

24 max = y;

25

26 if ( z > max )

27 max = z;

28

29 return max;

30 }

Enter three integers: 22 85 17

Maximum is: 85



Receive nothing and return nothing#include <stdio.h>

void greeting(void); /* function prototype */

int main(void)

{

greeting( );

greeting( );

return(0);

}

void greeting(void)

{

printf("Have fun!! \n");

}

Have fun!!

Have fun!!

Press any key to continue



Receive nothing and return int#include <stdio.h>

int getInput(void) /* ignore function prototype */

{

int number;

printf("Enter a number:");

scanf("%d",&number);

return number;

}

int main(void)

{

int num1, num2, sum;

num1 = getInput( );

num2 = getInput( );

sum = num1 + num2;

printf("Sum is %d\n",sum);

return(0);

}

Enter a number: 5

Enter a number: 4

Sum is 9




Receive parameter(s) and return nothing#include <stdio.h>

int getInput(void);

void displayOutput(int);

int main(void)

{

int num1, num2, sum;

num1 = getInput();

num2 = getInput();

sum = num1 + num2;

displayOutput(sum);

return(0);

}int getInput(void){

int number;printf("Enter a number:");scanf("%d",&number);return number;

}void displayOutput(int sum){

printf("Sum is %d \n",sum);}

Enter a number: 5

Enter a number: 4

Sum is 9




Exmaple: my_fabs (Version 1)double my_fabs(double x){

double res;

if(x >= 0)

res = x;

else

res

return

= -1 * x;

res;

}

* b));

void main(void){

double d = -10;

double b;

b = my_fabs(d);

printf("%f\n", b);

printf("%f\n", my_fabs(-2

}

10

20




if(x >= 0)

return x;

return (-1 * x);

}

void main(void){

double d = -10;

double b;

b = my_fabs(d);

printf("b = %f\n", b);

b = my_fabs(-2 * d);


}



Output of functionsA function can produce at most one output

Output of functions can be dropped

drop the output of sinsin(f); //we

Casting in functionsCast for input Prototype: void f(int a, double b);

Call: f(10.1, 20.2);

Cast for outputf(int a);

= f(10);

Prototype: int

Call: double d

Cast in return

int f(int a){

...

return 10.20

}



Scope of VariablesUntil now, Variables

Are declared in the start of functions

Are used any where in the function after declaration

Can we used them outside of function?

Can we used them in other functions?

Scope of variable

A range of code that the variable can be used

Variable cannot not be used outside of itsscope

Compile error



Scopes vs. BlocksScopes are determined by Blocks

Start with { and finished by }

Example: statements of a function, statement of aif or while, …

Variables

Can be declared in a block

Can be used in the declared block

Cannot be used outside the declared block

The Declared block is the scope of the variable



Variables in Blocks

61

", i);

#include <stdio.h>

int main(void){

int i;

for(i = 1; i <= 10; i++){

int number;

printf("Enter %d-th number:

scanf("%d", &number);

if((number % 2) == 0)

printf("Your number is even\n");

else

printf("Your number is odd\n");

}

/* compile error

printf("The last number is %d\n", number); */

return 0;

}



Nested Scopes/BlocksScopes can be nested

Example: Nested if, nested for, …

int main(){ //block 1

int i;

{ //block 2

int j;

{ //block 3

int k;

}

int m;

}

return 0;

}



Variables in Nested BlocksAll variables from outer block can be used

inner blocks

Scope of outer block contains the inner block

Variables in inner block cannot be used in

outer block

Scope of the inner block does not contains the

outer block



Variables in Nested Blocks: Exampleint k;

for(int i = 0; i < 10; i++){

/* block 1 */

if(i > 5){

/* block 2 */

int j = i;

...

j; compile error */

}

while(k > 10){

/* block 3 */

int l = i;

/* int m =

...

}

/* k = l; compile error */

}



Same Variables in Nested Block If a variable in inner block has the same identifier of a

variable in outer block

The inner variable hides the outer variable

Changing inner variables does not change outer variable

int i=10 ,j = 20;

printf("outer i =%d, %d\n",i,j);if(true){

int i=100;

j = 25;

printf("outer i =%d, %d\n",i,j);

}

printf("outer i =%d, %d\n",i,j);



Local VariablesAll defined variables in a function are the local

variable of the function

Can ONLY be used in the function, not other functions

void func(void){

int i, j;

float f;

/* These are local variables */

error,

error,

why? */

why? */

}

int main(void){

i = 10; /* compile

f = 0; /* compile

}



Global VariablesGlobal variables are defined outside of all functions

Global variables are initialized to zero

Global variables are available to all subsequent

functions

void f(){

i = 0; // compile error

}

int i;

void g(){

int j = i; // g can use i

}



Global Variables: Example

\n", i);

\n", f);

int i, j;

float f;

void func(void){

printf("i = %d

printf("f = %f

i = 20;

}

void f(){

printf("%d", i);

}

int main(void){

f = 1000;

func();

f();

return 0;

}




double res;

if(x >= 0)

res = x;

else

res

return

= -1 * x;

res;

}

* b));

void main(void){

double d = -10;

double b;

b = my_fabs(d);

printf("%f\n", b);

printf("%f\n", my_fabs(-2

}

10

20




if(x >= 0)

return x;

return (-1 * x);

}

void main(void){

double d = -10;

double b;

b = my_fabs(d);


b = my_fabs(-2 * d);


}



Parameter Passing by Global Variables: my_fabs (V.3)

double x;

void my_fabs(void){

x = (x > 0) ? x : -1 * x;

}

void main(void){

= -10;double b, d

x = d;

my_fabs();

b = x;


}

Don’t use this method.

Parameters should be passed

by input parameter list.

Global variable are used to

define (large) variables that

are used in many functions



Prototype Example#include <stdio.h>

int max(int x,int y);

int main(void)

{

int a = 10, b = 20;

int m = max(a, b);

printf("m is %d", m);

return 0;

}

int max(int x, int y)

{

if (x > y)

return x;

else

return y;

}




int max(int,int);

int main(void)

{

int a = 10, b = 20;

int m = max(a, b);


return 0;

}


{

if (x > y)

return x;

else

return y;

}




int max(int a,int b);

int main(void)

{

int a = 10, b = 20;

int m = max(a, b);


return 0;

}


{

if (x > y)

return x;

else

return y;

}



How Function Call Works Function call is implemented by “stack”

Stack is a part of main memory

When a function calls

Its variables including the inputs are allocated in stack

The value of input parameters from caller function is pushed to stack of called function

When function finished, its stack is freed



Example: print_sub function#include <stdio.h>

void print_sub(double a, double b){

double res;

res = a - b;

printf("Sub of %f and %f is %f\n", a, b, res);

}

int main(void){

double d1 = 10, d2 = 20;

print_sub(56.0,6.0); //What is the output?

print_sub(d1,d2); //output?

print_sub(d1,d2 + d2); //output?

return 0;



print_sub: What happen?

double a = 56.0;

double b = 6.0;

double res;

res = a - b;

print_sub(56.0, 6.0);

56.0 is copied the memory location a

6.0 is copied to memory location b



print_sub: What happen?print_sub(d1, d2);

Value of d1 is copied the memory location a

Value of d2 is copied to memory location b

double a = 10;

double b = 20;

double res;

res = a - b;

Call by Value



Call by valueIn call by value mechanism

The values are copied to the function

Only the copy of variable’s value (copy of actual parameter’s value) is passed to the function.

If we change values in the function

The copied version is changed

The original value does not affected

Any modification to the passed value inside the function will not affect the actual value



Call by value – An Example• How are the arguments passed

into functions?– 'Pass by value'– function arguments are

expressions

– In the function call:

• Expressions are evaluated and copies of their values are put into temporary memory locations

• The names of the corresponding parameters in the function definition are made to be the names of the copies

– The values of the expressions in the function call are notchanged

#include <stdio.h>


int main()

{

int a = 10;


double ans;





}


double product(double A, double B)

{

double result;

result = A * B;

return result;

}



Call by value – An Example#include <stdio.h>int calSum(int,int); /*function protototype*/

int main(void){

int sum, num1, num2;

printf("Enter two numbers to calculate its sum:\n");scanf("%d%d",&num1,&num2);

sum = calSum(num1,num2); /* function call */printf("\n %d + %d = %d", num1, num2, sum);return(0);

}

int calSum(int val1, int val2) /*function definition*/{

int sum; sum = val1 + val2; val2 = 100;return sum;

}

Enter two numbers to calculate its sum:

4

9

4 + 9 = 13


?num2

?num1

?sum

4num2

9num1

?sum

4val2

9val1

?sum

4val2

9val1

13sum

100val2

9val1

13sum

4num2

9num1

13sum



Example: print_sub function#include <stdio.h>

void print_sub(double a, double b){

double res;

res = a - b;

printf("Sub of %f and %f is %f\n", a, b, res);

}

int main(void){

double d1 = 10, d2 = 20;

print_sub(56.0,6.0); //What is the output?

print_sub(d1,d2); //output?

print_sub(d1,d2 + d2); //output?

return 0;



add function (wrong version)res){void add(double a, double b, double

res = a + b;

return;

}

int main(void){

10.1, d2 = 20.2;double d1 =

double result = 0;

add(56.0, 6.7,

printf("result

result);

= %f\n", result);

add(d1, d2, result);

printf("result = %f\n", result);

}

result = 0

result = 0



Call by reference• Call by reference

– In this method, the reference (memory address)of the variable is passed to the function. Anymodification passed done to the variable insidethe function will affect the actual value



Call by reference#include <stdio.h>

void CalByVal(a, b)

{

a = 0; b = 10;

}

void CalByRef(int *a, int *b) // CalByRef(int *p, int *q)

{

*a = 0; *b = -5;

}

int main(void)

{

int a = 1, b = 5;

printf("Before cal CalByVal: a = %d, b = %d\n", a, b);

CalByVal(a, b);

printf("After cal CalByVal: a = %d, b = %d\n", a, b);

printf("Before cal CalByRef: a = %d, b = %d\n", a, b);

CalByRef(&a, &b);

printf("After cal CalByRef: a = %d, b = %d\n", a, b);

return 0; /* Exit program. */

}

main5b

1a

CalByVal5b

1a

CalByVal10b

0aCalByRefb

a

main-5b

0a



Call by reference

• Instead of product()– prod_sum()

– How can I get the function to give both product and sum?

• put * in front of variable name in prototype and function definition

• put & in front of variable names in function call

#include <stdio.h>

void prod_sum(double x, double y,

double *ptr1, double *ptr2);

int main()

{


double prod, sum;

prod_sum(var1, var2, &prod, &sum);

printf("var1= %g\n"

"var2= %g\n",var1, var2);

printf("prod= %g\n" "sum= %g\n", prod, sum);

}


void prod_sum(double A, double B,

double *rslt_prod, double *rslt_sum)

{

*rslt_prod = A * B;

*rslt_sum = A + B;

}



Recursion• Recursion is a technique that solves a problem by solving a smaller

problem of the same type

• A recursive function is a function invoking itself, either directly orindirectly

– Recursion: A → B → C → D → A

• Concept of recursive function (generally):

– A recursive function is called to solve a problem

– The function only knows how to solve the simplest case of the problem. Whenthe simplest case is given as an input, the function will immediately returnwith an answer if (stopping case)

solve itelse

reduce the problem using recursion



RecursionRecursiveAlgorithm

An algorithm uses itself to solve the problem

There is a basic problem with known solution

Recursive Algorithms are implemented by

recursive functions

Recursive function

A function which calls itself

There is a condition that it does not call itself



Recursion• Any problem that can be solved recursively can also be

solved iteratively (using loop)

• Recursive functions are slow and takes a lot of memoryspace compared to iterative functions

• So why bother with recursion? There are 2 reasons:– Recursion approach more naturally resembles the problem

and therefore the program is easier to understand anddebug

– Iterative solution might not be apparent



An Example: xy

• In this example, we want to calculate x to the power of y– i.e. xy

• If we analyze the formula for xy, we could see that xy

could be written as (x being multiplied to itself, y times)– An example is 24, which can be written as

24 = 2 x 2 x 2 x 2 (in this case, x = 2, y = 4)

– 24 could also be rewritten as24 = 21 x 23 where 21 = 2 (i.e the number itself)

• Therefore, we could divide the problem into two stage:

– Simplest case: when y = 1, the answer is x

– Recursive case, we need to solve for x * x(y-1)



Introduction

Factorial

n! = n x n-1 x … x 2 x 1

n! = n x (n-1) !

GCD

GCD(a, b) = EuclideanAlgorithm

GCD(a, b) = GCD(b, a mod b)

There is a simple (basic) problem which we can

solve it directly (without recursion)

Factorial: 1! = 1

GCD: b == 0



Euclid’s algorithm in Cint gcd(int m, int n){int r;while ( (r = m % n) != 0) {m = n;n = r;

}return n;

}



Recursion solution of xy

#include <stdio.h>

double XpowerY(double, int);

int main(void)

{

double power, x; int y;

printf("Enter the value of x and y:\n");

scanf("%f%d", &x, &y);

power = XpowerY(x, y);

printf("%.2f to the power of %d is %.2f\n\n", x, y, power);

return(0);

}

double XpowerY(double x, int y)

{

if (y ==1)

return x;

else

return x * XpowerY(x, y-1);

}

Enter the value of x and y:

2

3

2.00 to the power of 3 is 8.00




#include <stdio.h>

int factorial(int n){

int res, tmp;

if(n == 1)

/* The basic problem */

res = 1;

else{

/* recursive call */

tmp = factorial(n - 1);

res = n * tmp;

}

return res;

}

fac);

void main(void){

int i = 4;

int fac = factorial(i);

printf("%d! = %d\n", i,

}



Factorial#include <stdio.h>double fact(double);int main(void){

double n, result;printf("Please enter the value of n:");scanf("%lf", &n);

result = fact(n);printf(" %.f! = %.2f\n", n, result);return(0);

}double fact(double n){

if (n <= 1)return 1;

elsereturn n * fact(n-1);

}

Enter the value of n: 3

3! = 6.00


int fact(int n){

int factres = 1;while(n>1){

factres = factres*n;n--;

}return (factres);

}


Sharif University of TechnologyDepartment of Computer Engineering 96 10-96

How C Maintains the Recursive Steps

• C keeps track of the values of variables by thestack data structure.

– Recall that stack is a data structure where the lastitem added is the first item processed

– There are two operations (push and pop) associatedwith stack

a

b

c

b

c

d

b

c

pop push d



How C Maintains the Recursive Steps

• Each time a function is called, the executionstate of the caller function (e.g., parameters,local variables, and memory address) arepushed onto the stack

• When the execution of the called function isfinished, the execution can be restored bypopping up the execution state from the stack



#include <stdio.h>

int factorial(int n){

int res, tmp;

if(n == 1)

/* The basic problem */

res = 1;

else{

/* recursive call */

tmp = factorial(n - 1);

res = n * tmp;

}

return res;

}

fac);

void main(void){

int i = 4;

int fac = factorial(i);

printf("%d! = %d\n", i,

}



Function Call Graphfactorial(4)

n = 4, tmp = ?, res = ?

factorial(3)

n = 3, tmp = ?, res = ?

factorial(2)

n = 2, tmp = ?, res = ?

factorial(1)

n = 1, res = ?



Function Call Graphfactorial(4)

n = 4, tmp = ?, res = ?

factorial(3)

n = 3, tmp = ?, res = ?

factorial(2)

n = 2, tmp = ?, res = ?

factorial(1)

n = 1, res = 1

1



factorial(4)

n = 4, tmp = ?, res = ?

factorial(3)

n = 3, tmp = ?, res = ?

factorial(2)

n = 2, tmp = 1, res = 2

factorial(1)

n = 1, res = 1

1

2

Function Call Graph



factorial(4)

n = 4, tmp = ?, res = ?

factorial(3)

n = 3, tmp = 2, res = 6

factorial(2)

n = 2, tmp = 1, res = 2

factorial(1)

n = 1, res = 1

1

2

6

Function Call Graph



factorial(4)

n = 4,tmp = 6,res = 24

factorial(3)

n = 3, tmp = 2, res = 6

factorial(2)

n = 2, tmp = 1, res = 2

factorial(1)

n = 1, res = 1

1

2

6

Function Call Graph



Recursive Steps of xy

#include <stdio.h>

double XpowerY(double, int);

int main(void)

{

double power, x; int y;

printf("Enter the value of x and y:\n");

scanf("%lf%d", &x, &y);

power = XpowerY(x, y);

printf("%.2f to the power of %d is %.2f\n\n", x, y, power);

return(0);

}

double XpowerY(double x, int y)

{

if (y ==1)

return x;

else

return x * XpowerY(x, y-1);

}

x = 2; y = 4;

x * XpowerY(2, 3)

x = 2; y = 3;

x * XpowerY(2, 2)

x = 2; y = 2;

x * XpowerY(2, 1)

x = 2; y = 1;

return x;

2*8

2

2 * 2

2 * 4



ExamplesRecursive version of GCD?

Recursive version of Fibonacci numbers

Fibonacci numbers

1, 1, 2, 3, 5, 8, ...

Print digits: left-to-right and right-to-left



#include <stdio.h>

int GCD(int a, int b){

if(b == 0)

return a;

else

return GCD(b, a % b);

}

int main(void){

printf("GCD(1, 10)

printf("GCD(10, 1)

= %d \n",

= %d \n",

GCD(1, 10));

GCD(10, 1));

printf("GCD(15, 100)

printf("GCD(201, 27)

= %d \n", GCD(15, 100));

= %d \n", GCD(201, 27));

return 0;

}

#include <stdio.h>

return 0;

}

int fibo(int n){

if(n

else

== 1)

return 1;

if(n == 2)

return 1;

else

return fibo(n - 1) + fibo(n - 2);

}

int main(void){

printf("fibo(1)

printf("fibo(3)

printf("fibo(5)

printf("fibo(8)

= %d\n",

= %d\n",

= %d\n",

= %d\n",

fibo(1));

fibo(3));

fibo(5));

fibo(8));



Fibonacciint Fib (int n){

int temp = 1; /* handles base cases */ if (n > 2)

temp = Fib(n-1) + Fib(n-2);return temp;

}

/* Iterative Version */

int fibonacci(int k)

{

int a,b,c,i;

if (k <= 1) return 1;

else

{

a = 1;

b = 1;

i = 2;

while (i <= k)

{

c = a + b;

a = b;

b = c;

i++;

}

return(c);

}

}



Fibonacci

int Fib (int x){

/* handles base cases of */ /* Fib(1) and Fib(2) */ int temp = 1; if (x > 2) /* other cases */

temp = Fib(x-1) + Fib(x-2);return temp;

}

x = 5; temp = 1;

temp = Fib(4) + Fib(3);return temp;

x = 1; temp = 1;

return temp;

x = 4; temp = 1;


x = 5; temp = 1;


x = 1; temp = 1;

return temp;x = 1; temp = 1;

return temp;

x = 5; temp = 1;


x = 1; temp = 1;

return temp;

x = 1; temp = 1;

return temp;

x = 5; temp = 1;

temp = 1 + Fib(1);return temp;

x = 5; temp = 1;

temp = 1 + 1;return temp;

x = 4; temp = 1;


x = 4; temp = 1;


x = 5; temp = 1;


x = 5; temp = 1;


x = 5; temp = 1;


x = 5; temp = 1;




Fibonacci

fib(5)

fib(4) fib(3)

fib(3) fib(2)

fib(2) fib(1)

fib(2) fib(1)

1

1

1

12 1

3 2

5int Fib (int n){

/* handles base cases of */ /* Fib(1) and Fib(2) */ int temp = 1; if (n > 2) /* other cases */

temp = Fib(n-1) + Fib(n-2);return temp;

}

76

#include <stdio.h>

return 0;

}

void print_digit_right_left(int n){

int digit = n % 10;

printf("%d ", digit);

if(n >= 10)

print_digit_right_left(n / 10);

}

int main(void){

printf("\n print_digit_right_left(123): ");

print_digit_right_left(123);

printf("\n print_digit_right_left(1000): ");

print_digit_right_left (1000);

#include <stdio.h>

return 0;

}

void print_digit_left_right(int n){

if(n >= 10)

print_digit_left_right(n / 10);

int digit = n % 10;

printf("%d ", digit);

}

int main(void){

printf("\n print_digit_left_right(123): ");

print_digit_left_right(123);

printf("\n print_digit_left_right(1000): ");

print_digit_left_right (1000);



Indirect recursionWhat we have seen are direct recursion

A function calls itself directly

Indirect recursion

A function calls itself using another function

Example:

Function A calls functionB

Function B calls functionA

#include <stdio.h>

#include <stdbool.h>

bool

bool

is_even(int n);

is_odd(int n);

bool is_even(int n){

true;

false;

if(n == 0)

return

if(n == 1)

return

else

return is_odd(n - 1);

}

bool is_odd(int n){

if(n == 0)

return false;

if(n == 1)

return true;

else

return is_even(n - 1);

}



int main(void){

is even\n");

if(is_even(20))

printf("20

else

printf("20 is odd\n");

is odd\n");

if(is_odd(23))

printf("23

else

printf("23 is even\n");

return 0;

}



Bugs & Avoiding Them

Recursion must finish, be careful a bout basic

problem in the recursive functions

No base problem Stack Overflow



Exercise• What is the output of the following program

void function2()

{

printf("In function 2\n");

}

void function1()

{

function2();


}

#include <stdio.h>

void function3()

{


function2();

}

int main()

{

function1();

function3();

return 0;

}

In function 2In function 1In function 3In function 2



Storage Classes• Variables have attributes

– Have seen name, type, value

– Scope

• Where variable can be referenced in program

– Storage class

• How long variable exists in memory

– Linkage

• For multiple-file program, which files can use it



Organizing Multi-File Programs• A large C program should be divided into multiple files

// main.c#include <stdio.h>void Test(){

// …}int main(){

// …return 0;

}

// math.cdouble mathVar;double sin(){

double tempSin;// …

return tempSin;}



Scope of Identifiers• Scope of a declaration of an identifier

– The region of the program that the declaration isactive (i.e., can access the variable, function, label,etc.)

• Five types of scope:– Program (global scope)– File– Function prototype– Function– Block ("between the { } scope")



Scope of Identifiers - Program Scope• Program (global) scope

– if declared outside of allfunctions

– "Visible" to all functions from point of declaration

– Use only when necessaryand then very carefully!!

– If there exist a local variableand a global variable withthe same name, thecompiler will refer to thelocal variable

– Visible to functions in othersource files

#include <stdio.h>

int a = 10;


int main()

{


double ans;


// …

}



{

double result; a = 20;

result = x * y;

return result;

}

a = 10

a = 20



An Example// File name: main.c

#include <stdio.h>

int a = 10;



{

double result;

// …

a = 70;

return result;

}

int main()

{

a = 80;

}

// File name: ExternFile.c

extern int a = 10;


void TestExtern()

{

// …

a = 90;

// …

}



Scope of Identifiers - File Scope

• File scope

– Keyword static• Makes variable a ‘visible’

only within this source file

– Use file scope to avoidnaming conflict if multiplesource files are used

#include <stdio.h>

static int a = 10;


int main()

{


double ans;


// …

}



{

double result;

result = x * y;

return result;

}



An Example// File name: main.c

#include <stdio.h>

static int a = 10;



{

double result;

// …

a = 70;

return result;

}

int main()

{

a = 80;

}

// File name: ExternFile.c

extern int a = 10;


void TestExtern()

{

// …

a = 90;

// …

}



• Function prototype scope– Identifiers x and y are not

visible outside the prototype

– Thus, names in the prototypedo not have to match names inthe function definition

• MUST match types, however!

Scope of Identifiers - Function Prototype Scope

#include <stdio.h>


int main()

{

int a = 10;


double ans;





}


double product(double A, double B)

{

double result;

result = A * B;

return result;

}



• Function scope– Active from the beginning to the end of a function

Scope of Identifiers - Function Scope

#include <stdio.h>

int main()

{

int a;

// …

return 0;

}

int FunctionScopeTest()

{

int b;

// …

return 0;

}



Scope of Identifiers - Block Scope• Block (local) scope

– A block is a series of statements enclosed in braces { }

– The identifier scope is active from the point of declaration to the end of the block ( } )

– Nested blocks canboth declare the same variable name and not interfere

#include <stdio.h>


int main()

{

int a = 10;


double ans;


// …

}



{

double result; // a = 60; Error

result = x * y;

return result;

}



Storage ClassesStorage class(Refers to the lifetime of a variable)

How memory is allocated for the variable

Until when the variable exists

How it is initialized

Storage classes in C

Automatic

External

Static

Register



Storage Classes• Refers to the lifetime of a variable

• Local variables only exist within a function by default. When calling a function repeatedly, we might want to– Start from scratch – reinitialize the variables

• The storage class is ‘auto’

– Continue where we left off – remember the last value• The storage class is ‘static’

• Another two storage classes (seldomly used)– register (ask to use hardware registers if available)– extern (global variables are external)

How use?

storage_classes variable_type variable_name

autoregisterstaticextern

intfloatdoublechar



Storage Classes: Automatic• 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



Storage Classes: Automatic• The example above defines two variables with the

same storage class. auto can only be used within functions, i.e. local variables.

• auto is the default storage class for local variables.

{int Count;auto int Month;

}



Auto - Example#include <stdio.h>void auto_example(void);

int main(void){

int i;

printf("Auto example:\n");

auto_example( );auto_example( );auto_example( );

return(0);

}

void auto_example(void)

{

auto int num = 1;

printf(" %d\n",num);

num = num + 2;

}

Auto example:

1

1

1




Storage Classes: ExternalAll global variables are external by default

Are generated when program starts

Are destroyed when program finishes

Usage of keyword “extern”

To use global variables in other files

To emphasize that variable is global

This usage is optional



Storage Classes: StaticKeyword “static” comes before them

For local variables:

1) Generated in the first run of the block

2) Destroyed when program finishes

3) Initialized

If no value initialized by 0

Only initialized in the first run of the block



Static storage class• However the static keyword can be applied to

a local variable so that the variable still existeven though the program has gone out of thefunction. As a result, whenever the programenters the function again, the value in thestatic variable still holds



Static - Example#include <stdio.h>void auto_example(void);

int main(void){

int i;

printf("Static example:\n");static_example( );static_example( );static_example( );return(0);

}

void static_example(void)

{

static int num = 1;

printf(" %d\n",num);

num = num + 2;

}

Static example:

1

3

5




Storage Classes: Static

Keyword “static” comes before them

For global variables:

1) Generated when program starts

2) Destroyed when program finishes

3) Always initialized

If no value initialized by 0

4) Is not accessible for other files



Storage Classes: RegisterKeyword “register” comes before them

Can be used for local variables

Compiler tries to allocated the variable in

registers of CPU

But does not guaranteed

Registers are very fast and small memories

Improve performance

The register Storage Class

• The register storage class is used to define local variables that should be stored in a register instead of RAM.

• The register should only be used for variables that require quick access such as counters. It should also be noted that defining 'register' does not mean that the variable will be stored in a register. It means that it MIGHT be stored in a register depending on hardware and implementation restrictions.



Storage Classes, Register: Examples

register int i;

i++)for(i = 0; i < 100;

…



Storage Classes: Examplesint i;

void func(void){

int j;

printf("i = %d \n", i);

printf("j = %d \n", j);

i = 20;

}

int main(void){

func();

func();

i = 30;

func();

return 0;

}

i =???

j =???

i =???

j =???

i =???

j =???

i = 0

j = 0

i = 20

j = 0

i = 30

j = 0



Storage Classes, Static:Examplesvoid func(void){

int j;

static int i;



i = 20;

}

int main(void){

func();

func();

/* i = 30; compile error, why? */

func();

return 0;

}

i = ???

j = ???

i = ???

j = ???

i = ???

j = ???

i = 0

j = 0

i = 20

j = 0

i = 20

j = 0



Storage Classes, Static:Examplesvoid func(void){

int j;

static int i=10;



i = 20;

}

int main(void){

func();

func();

return 0;

}

i = ???

j = ???

i = ???

j = ???

i = 10

j = 0

i = 20

j = 0



Storage Classes, Extern: Examplesint i = 10, j = 20;

void print(void){

printf("i = %d, j = %d\n", i, j);

}

int main(void){

extern int i; // i refers the global i

int j; // j is new variable

print();

i = 1000;

j = 2000;

print();

return 0;

}

i = ?, j = ?

i = ?, j = ?

i = 10, j = 20

i = 1000, j = 20



Inline Functions & MacroFunction call using stack has its overhead

2 approaches to reduce the overhead

inline function

To ask from compiler to compile it as inline, but

no guarantee

int f(float x)inline

Macros

#define PRINT_INT(X) printf("%d\n", X)


If the same sequence of steps or instructions is required in several different places in a program, you will normally write a function for the steps and call the function whenever these steps are required. But this involves time overhead.

Also can place the actual sequence of steps wherever they are needed in the program, but this increase the program size and memory required to store the program. Also need retyping process or copying a block of program.

Use function if the sequence of steps is long. If small, use macros or inline function, to eliminate the need for retyping and time overhead.

Inline Functions



Example: GCD#define PRINT_INT(x) printf("%d\n",x)

int b){ /* return gcd of a and b */inline int gcd(int a,

int temp;

while(b != 0){

temp = a % b;

a = b;

b = temp;

}

return a;

}

void main(void){

int i = 20, j = 35, g;

g = gcd(i, j);

", i , j);printf("GCD of %d and %d =

PRINT_INT(g);

", j , i);

g = gcd(j, i);

printf("GCD of %d and %d =

PRINT_INT(g);

}


Need #define compiler directive. For example, to obtain just the area of a triangle, we could create a macro,

Macros


Example for finding an average of 4 numbers (a, b, c and

d),

#define avg(x, y) (x + y)/2.0

Then in the program we can define something like this,

avg1 = avg(avg(a, b), avg(c, d))

Doing the substitution,

avg4 = ((a + b)/2.0 + (c + d)/2.0) / 2.0

The drawback: nesting of macros may result in code that

difficult to read.

C FUNCTIONS



151

Header Files• Header files contain

– Function prototypes

– Definitions of data types and constants

• Header files ending with .h

– Programmer-defined header files#include “myheader.h”



Typical C Programming Style• A lot of small C programs, rather than a few

large ones• Each .c file contains closely related functions

• Usually a small number of functions

• Header files to tie them together



Header files• In applications with multiple C programs,

function prototypes are typically provided in header files

• I.e., the ‘.h’ files that programmers include in their code

• Grouped by related functions and features• To make it easier for developers to understand

• To make it easier for team development

• To make a package that can be used by someone else

Documents

Functionsce.sharif.edu/courses/98-99/1/ce153-8/resources/root/Slides/Lecture9_1… · • Function definition format return-value-type function-name( parameter-list ) {declarations