Go4Expert - All about Arrays in C/C++

Introduction

This article talks about Multi-Dimensional Arrays in C/C++. About 1-Dimensional Array see All about Arrays in C/C++ - Part I

Background

Multi-Dimensional Arrays

Multi-dimensional arrays are nothing but Array of Arrays.

Characteristics

All the characteristics that are mentioned for 1-Dimensional array hold good for multi-dimensional arrays too. Apart from that

Declaration of a multi-dimension array as a argument to a function, must include explicit size declarations in all of the subscript positions except for the first, which is an option.

Memory Mapping

In C/C++, memory mapping for multi-dimensional arrays, is done using Row Major ordering. It means, memory is allocated for successive elements by incrementing the rightmost index until it reaches the end of the current Row.

2-Dimensional Arrays

A 2-Dimensional Array is nothing but Array of 1-Dimensional Array and is represented with two subscripts as below:

data_typeOfArray arrayName [firstDimension] [secondDimension]

whereas firstDimension represents Rows and secondDimension as Columns.

For Example:
int array[2][3];

Can be said as 2 Arrays of 1-Dimensional Array of size 3;

2-Dimensional array can be imagined as a 2-D table made of elements of same data type as in below picture:

http://www.go4expert.com/images/arti...-2/Arrays4.JPG

Memory Mapping in 2-D Array

For Example if we take int array[2][3];

Here array will be considered as base address and successive address for the elements will be allocated at sizeOfElement*offset in Row Major Ordering. And here if we take the sizeOfElement as 1 then memory mapping for array[2][3] looks as below:

array[0][0] --------> 0
array[0][1] --------> 1
array[0][2] --------> 2

array[1][0] --------> 3
array[1][1] --------> 4
array[1][2] --------> 5

Where 0,1,2 etc at right hand side are offsets

Size of a 2-D Array

Here size of the Array can be calculated as:
firstDimension*secondDimension*sizeOfDataType

For example:
size of the int array[2][3] will be 2*3*4= 24 bytes

Address of an Element

Address of an Element at a given rowIndex and colIndex in an array[rowSize][colSize] can be calculated using the below formula:

Element_Address = Base_Address + (col_size*(rowIndex) + (colIndex)) * Element_Size

For Example:
If we take the above array[2][3] with assumption of baseAddress as 0 and element size as 1, then address of element at indices array[0][1] can be found as:

array[0][1] = 0 + (3*(0)+1)*1
= 0 + (0+1)*1
= 0 + 1
= 1

Initializing a 2-D Array

As like in 1-D Arrays, 2-D Arrays can be initialized as below:

int arry[2][3]={10,20,30,40,50,60}; or

int array[2][3] = {{10,20,30}, {40,50,60}}

And they are assigned with array indices as below:

array[0][0]=10;
array[0][1]=20;
array[0][2]=30;
array[1][0]=40;
array[1][1]=50;
array[1][2]=60;

Like in 1-D array, if some of the indices in an interger 2-D array are not given initilizers, then by default they will be initilized with 0. Also if we assign 2-D Array with {}, it would initialize all it's elements with 0 value.

2-D Array of characters

Like in 1-D Arrays, 2-D Arrays of strings can be initilized in two ways as below:

char array[3][4] = {
'd','o','g'
'c','a','t',
'r','a','t'
}

or

char array[3][4] = {"dog", "cat", "rat"};

As said above for integer arrays, if initilizers are not given for some elements in character arrays, they will be initilized with NULL characters.

Accessing elements in 2-D Arrays

In the above mentioned array i.e. int array[2][3], we can access the 1st Rowth 0th column element as array[1][0] i.e. 40

Dynamic Allocation of 2-D Arrays

Suppose we want an int array[row][col]. where size of row and col are unknown at compile time. Then we can allocate the array dynamically and access it's elements with below steps:

The name array will be a pointer to a pointer to int.
Initialliy allocate memory of size x for pointers to int i.e. array of pointers
Then allocate memory of size y for each of these pointer
Access each of the element of array as array[i][j], where i>=0 and <x and j>=0 and <y.
Free the memory by deallocating for each of the pointers and then array of pointers itself.

For Example:

The code

Code: Cpp

#include<iostream.h>

int main()
{
  int row, col;
  int **array; //array name
  int i, j;
  
  cout<<"Enter the matrix size i.e. row and column"<<endl;
  cin>>row;
  cin>>col;

  //allocate memory of size row to pointer to int
  array = new int*[row];

  //allocate memory of size col to each of the pointer to int 
  for(i=0; i<row; ++i)
  {
    array[i] = new int[col];
  }

  cout<<endl;

  cout<<"Please eneter the values array["<<row<<"]["<<col<<"] :\n";
  
  for(i=0; i<row; ++i)
  {
    for(j=0; j<col; ++j)
    {
      cin>>array[i][j];
    }
  }

  cout<<endl;

  cout<<"Contents of array["<<row<<"]["<<col<<"] are :\n";

  for(i=0; i<row; ++i)
  {
    for(j=0; j<col; ++j)
    {
      cout<<array[i][j];
    }
    cout<<endl;
  }

  cout<<endl;

  //now for each pointer, free its array of ints
  for (i = 0; i < row; i++) 
  {
    delete [] array[i];
  }
   
  //now free the array of pointers
  delete [] array;
  
  return(0);
}

Output:
------------
./a.out
Enter the matrix size i.e. row and column
2
3

Please eneter the values array[2][3] :
1
2
3
4
5
6

Contents of array[2][3] are :
123
456

2-Dimensional array as an Argument to a function

As said above, the size of the first dimension may be omitted, but the second dimension has to be mentioned in the argument.

For Example:

The code

Code: Cpp

#include<iostream.h>

double sum(double array[][3], int size) 
{
  double temp = 0.0;
  
  for(int i = 0 ; i < size ; i++)
  {
    for(int j = 0 ; j < 3 ; j++)        
    {
      temp += array[i][j];
    }
  }
  return temp;
}

int main() {
  double array[2][3] = {
                         { 1.0,  2.0,  3.0},
                         { 5.0,  6.0,  7.0}
                       };

  cout << "Sum of all the elements in array[2][3] is: "<< sum(array, sizeof array/sizeof array[0])<< endl;
  return 0;
}

Output:
----------
Sum of all the elements in array[2][3] is: 24

Easy way of reading a 2-D Array

Imagine a 2-D array as shown in below example picture and then put the elements at the respective indices in the picture. This helps to read/recognise any element of the array at a given indcies.

http://www.go4expert.com/images/arti...-2/2dArray.JPG

3-Dimensional Arrays

Multi Dimensional Arrays are not limited to 2-D alone. They can be of as many indices as needed. But developer must be careful here, as the memory that consume increases with proportional to the dimension.

A 3-Dimensional Array can be said as Array of 2-Dimensional Arrays and is represented with three subscripts as below:

data_typeOfArray arrayName [firstDimension] [secondDimension] [thirdDimension]
whereas firstDiemnsion is also called as depthSize here.

For Example:

int arry[2][3][4]; can be called as
2 Arrays of 2-Dimensional Arrays of size 3X4 !!

Memory Mapping in 3-D Array

If we take int array[2][3][4];

And array will be considered as base address and successive address for the elements will be allocated at sizeOfElement*offset in Row Major Ordering as below:

array[0][0][0] --------> 0
array[0][0][1] --------> 1
array[0][0][2] --------> 2
array[0][0][3] --------> 3

array[0][1][0] --------> 4
.
.
.

array[0][2][0] --------> 8
.
.
.
array[1][0][0] --------> 12
.
.
.
array[1][1][0] --------> 16
.
.
.
array[1][2][0] --------> 20
array[1][2][1] --------> 21
array[1][2][2] --------> 22
array[1][2][3] --------> 23

Where 0,1,2 etc at right hand side are offsets

Size of a 3-D Array

Here size of the Array can be calculated as
firstDimension*secondDimension*thirdDimension*size OfDataType

For example the size of the below array will be:
int array[2][3][4]

2*3*4*4 = 96 bytes

Address of an Element

Address of an Element at a given depthIndex, rowIndex and colIndex in an array[depthSize][rowSize][colSize] can be calculated using the below formula:

Element Address = Base Address + ((depthIndex*colSize+colIndex) * rowSize + rowIndex) * Element_Size

If we take the above array[2][3][4] with assumption of baseAddress as 0 and element size as 1, then address of an element at indices array[1][2][3] can be found as:

array[1][2][3] = 0 + ((1*4+3)*3+2)*1
= 0 + ((4+3)*3+2)1
= 0 + ((7)*3+2)1
= 0 + 23

Easy way of Reading a 3-D Array

Elements in a 3-D array also can easily be read/recognised by imagining the picture of the Array as below:

http://www.go4expert.com/images/arti...-2/3dArray.JPG