CS102/ECE 206 Lecture Notes
More on Arrays

Output Parameters -- a Preview

Example: Swap array elements.


  void bad_swap(int, int);     // prototypes 
  void a_swap(int[], int, int);

  #define SIZE 100

  int main()  {
    int x, y,       // items to exchange
        i, j,       // indexes of array items to swap
        A[SIZE];    // assume initialization 

    cin >> x >> y;
    bad_swap(x, y);
    cout << "x and y in main: " << x << " " <<  y << endl << endl; 

    cin >> i >> j;
    if(i >= 0 && i < SIZE && j >= 0 && j < SIZE)  {  // are indexes in bounds?  
      a_swap(A, i, j);
      cout << "A[" << i << "] and A[" << j << "] in main: " 
                  <<  A[i] << " " << A[y] << endl; 
    }
    else  {
      cout << "Illegal array index; exiting program\n";
      exit(1);
    }

  return 0;
  }

  // Function definition - does NOT work 
  void bad_swap(int a, int b)  {
    int temp;

    temp = a;  a = b;  b = temp;
    cout << "a and b in bad_swap: " << a << " " << b << endl;  
  }

  // swap two array elements 
  void a_swap(int A[], int i, int j)  {
    int temp;

    temp = A[i];
    A[i] = A[j];
    A[j] = temp;
    cout << "A[" << i << "] and A[" << j << "] in a_swap: " 
         << A[i] << " " << A[j] << endl;  
  }


  Output if x and i are 6, y and j are 8, A[6] is 58, A[8] is -40:  
  
    a and b in bad_swap: 8 6
    x and y in main: 6 8

    A[6] and A[8] in a_swap: -40 58
    A[6] and A[8] in main: -40 58

________________________________________

Output Parameters and Scope

Input arguments are passed to functions by value. These arguments, along with any local variables are of the automatic storage class. This means that they are automatically created when the function is invoked and automatically destroyed when the function terminates. They are available only to the called function, not to the calling function. A copy of the argument is made and the function does its magic on the copy, not on the original.

In order to modify the original variable, the function must have access to the address of the variable. As you have seen, this happens when we pass an array to a function. The array name acts like a pointer to (the address of) the first array element. This means that A is the same thing as &A[0], or A == &A[0]. When we pass an array to a function, a copy of this address is made, and through this copy we have access to the array elements.

What about simple variables? Can we get access to the originals by passing (by value) their addresses? Yes! The correct version of swap() below illustrates this. Notice that a and b are input/output parameters since one value is held going in and that value is changed.

Example--corrected swap() with output parameters Call with the addresses of two integers.


  #include <iostream>
  using namespace std;  

  void swap(int *, int *);
  
  int main()  { 
    int x = 3, y = 4;

    swap(&x, &y);
    cout << x << " " << y << endl;

    return 0;
  }

  // swap two integers 
  void swap(int *a, int *b)  {  
    int temp;

    temp = *a; 
    *a = *b; 
    *b = temp;
    cout << *a << " " << *b << endl;  
  }


  What does each call so?  See the comments!

  swap(&x, &y);            // swap two integer variables  
  swap(&A[x], &A[y]);      // swap two array elements
  swap(&A[0], &A[SIZE-1]); // swap the first and last elements in A[]  
  swap(A, &A[SIZE-1]);     // swap the first and last elements in A[] (WHY?)

Multidimensional Arrays


  <type> a_name [number_rows][number_cols];

Example

char grid[3][4];

grid[1][3] = 'R';
grid[2][0] = '$';

0 1 2 3

? ? ? ?

? ? ? R

$ ? ? ?

Passing Multi-Dimensional Arrays to Functions
Use the same approach as for 1-D arrays, but include sizes of all but the first dimension.

#define DAYS 31 // Sample function header void get_data(int x[][DAYS]) // Prototype void get_data(int [][DAYS]);
Example: Fill a 2-D array in row-major order

#define NROWS 3 #define NCOLS 4 // Read values in row-major order char grid[NROWS][NCOLS]; int row, col; for(row = 0; row < NROWS; row++) for(col = 0; col < NCOLS; col++) cin >> grid[row][col];
If input is: a b c d e f g h i j k l m n o p, we would get:

0 1 2 3
a b c d

e f g h

i j k l

Question: Can you implement column-major?

Passing an entire array to a function

// Prototype void func(int[], char[][4]); // Call to func func(histo, grid); // Header for func void func(int histo[], char grid[][4])

Examples: an N x N array of integers
1. Print the diagonal:

// print A[0][0], A[1][1], A[2][2], ... for(i = 0; i < N; i++) cout << A[i][i] << endl;

2. Print the reverse diagonal,

// print A[0][N-1], A[1][N-2], ..., A[N-1][0] for(i = 0; i < N; i++) cout << A[i][N-(i+1)]) << endl;

3. Frame the array with asterisks:

// Frame top and bottom rows for(col = 0; col < N; col++) { A[0][col] = '*'; // top row A[N-1][col] = '*'; // bottom row } // Frame sides, no overwrites for(row = 1; row < N-1; row++) { A[row][0] = '*'; // left side A[row][N-1] ='*'; // right side }
4. Print the elements in each of the odd-numbered columns, one column per line:

// Print odd-numbered columns int row, col, A[NROWS][NCOLS]; // assume initialization

for(row = 0; row < NROWS; row++) { for(col = 1; col < NCOLS; col += 2) cout << setw(5) << A[row][col] << endl; cout << endl; }

Try these at home. Drawing the diagrams will help.

In an N by N array where N is odd, write code to put an asterisk in the center.
You have an M by N array, initialized. Write a function to display the values in the odd-numbered rows, one row per line.
You have an M by N array, initialized. Write a function to display the values in the even-numbered columns, one column per line.

CS102/ECE 206 Lecture Notes More on Arrays

Output Parameters -- a Preview

Output Parameters and Scope

Multidimensional Arrays

Passing Multi-Dimensional Arrays to Functions

Passing an entire array to a function

Examples: an N x N array of integers

CS102/ECE 206 Lecture Notes
More on Arrays