Find any one of the multiple repeating elements in read only array

• Difficulty Level : Hard
• Last Updated : 22 Jun, 2021

Given a read only array of size ( n+1 ), find one of the multiple repeating elements in the array where the array contains integers only between 1 and n.
Read only array means that the contents of the array can’t be modified.
Examples:

Input : n = 5
arr[] = {1, 1, 2, 3, 5, 4}
Output : One of the numbers repeated in the array is: 1

Input : n = 10
arr[] = {10, 1, 2, 3, 5, 4, 9, 8, 5, 6, 4}
Output : One of the numbers repeated in the array is: 4 OR 5

Since, the size of the array is n+1 and elements ranges from 1 to n then it is confirmed that there will be at least one repeating element.
A simple solution is to create a count array and store counts of all elements. As soon as we encounter an element with count more than 1, we return it. This solution works in O(n) time and requires O(n) extra space.
A space optimized solution is to break the given range (from 1 to n) into blocks of size equal to sqrt(n). We maintain the count of elements belonging to each block for every block. Now as the size of array is (n+1) and blocks are of size sqrt(n), then there will be one such block whose size will be more than sqrt(n). For the block whose count is greater than sqrt(n), we can use hashing for the elements of this block to find which element appears more than once.
Explanation
The method described above works because of the following two reasons:

1. There would always be a block which has count greater than sqrt(n) because of one extra element. Even when one extra element has been added it will occupy a position in one of the blocks only, making that block to be selected.
2. The selected block definitely has a repeating element. Consider that ith block is selected. Size of the block is greater than sqrt(n) (Hence, it is selected) Maximum distinct elements in this block = sqrt(n). Thus, size can be greater than sqrt(n) only if there is a repeating element in range ( i*sqrt(n), (i+1)*sqrt(n) ].

Note: The last block formed may or may not have range equal to sqrt(n). Thus, checking if this block has a repeating element will be different than other blocks. However, this difficulty can be overcome from implementation point of view by initialising the selected block with the last block. This is safe because at least one block has to get selected.
Below is the step by step algorithm to solve this problem

1. Divide the array in blocks of size sqrt(n).
2. Make a count array which stores the count of element for each block.
3. Pick up the block which has count more than sqrt(n), setting the last block
as default.
4. For the elements belonging to the selected block, use the method of hashing(explained in next step) to find the repeating element in that block.
5. We can create a hash array of key value pair, where key is the element in the block and value is the count of number of times the given key is appearing. This can be easily implemented using unordered_map in C++ STL.

Below is the implementation of above idea:

C++

 // C++ program to find one of the repeating// elements in a read only array#include using namespace std; // Function to find one of the repeating// elementsint findRepeatingNumber(const int arr[], int n){    // Size of blocks except the    // last block is sq    int sq = sqrt(n);     // Number of blocks to incorporate 1 to    // n values blocks are numbered from 0    // to range-1 (both included)    int range = (n / sq) + 1;     // Count array maintains the count for    // all blocks    int count[range] = {0};     // Traversing the read only array and    // updating count    for (int i = 0; i <= n; i++)    {        // arr[i] belongs to block number        // (arr[i]-1)/sq i is considered        // to start from 0        count[(arr[i] - 1) / sq]++;    }     // The selected_block is set to last    // block by default. Rest of the blocks    // are checked    int selected_block = range - 1;    for (int i = 0; i < range - 1; i++)    {        if (count[i] > sq)        {            selected_block = i;            break;        }    }     // after finding block with size > sq    // method of hashing is used to find    // the element repeating in this block    unordered_map m;    for (int i = 0; i <= n; i++)    {        // checks if the element belongs to the        // selected_block        if ( ((selected_block * sq) < arr[i]) &&                (arr[i] <= ((selected_block + 1) * sq)))        {            m[arr[i]]++;             // repeating element found            if (m[arr[i]] > 1)                return arr[i];        }    }     // return -1 if no repeating element exists    return -1;} // Driver Programint main(){    // read only array, not to be modified    const int arr[] = { 1, 1, 2, 3, 5, 4 };     // array of size 6(n + 1) having    // elements between 1 and 5    int n = 5;     cout << "One of the numbers repeated in"         " the array is: "         << findRepeatingNumber(arr, n) << endl;}

Java

 // Java program to find one of the repeating// elements in a read only arrayimport java.io.*;import java.util.*; class GFG{     // Function to find one of the repeating    // elements    static int findRepeatingNumber(int[] arr, int n)    {        // Size of blocks except the        // last block is sq        int sq = (int) Math.sqrt(n);         // Number of blocks to incorporate 1 to        // n values blocks are numbered from 0        // to range-1 (both included)        int range = (n / sq) + 1;         // Count array maintains the count for        // all blocks        int[] count = new int[range];         // Traversing the read only array and        // updating count        for (int i = 0; i <= n; i++)        {            // arr[i] belongs to block number            // (arr[i]-1)/sq i is considered            // to start from 0            count[(arr[i] - 1) / sq]++;        }         // The selected_block is set to last        // block by default. Rest of the blocks        // are checked        int selected_block = range - 1;        for (int i = 0; i < range - 1; i++)        {            if (count[i] > sq)            {                selected_block = i;                break;            }        }         // after finding block with size > sq        // method of hashing is used to find        // the element repeating in this block        HashMap m = new HashMap<>();        for (int i = 0; i <= n; i++)        {            // checks if the element belongs to the            // selected_block            if ( ((selected_block * sq) < arr[i]) &&                    (arr[i] <= ((selected_block + 1) * sq)))            {                m.put(arr[i], 1);                 // repeating element found                if (m.get(arr[i]) == 1)                    return arr[i];            }        }         // return -1 if no repeating element exists        return -1;} // Driver codepublic static void main(String args[]){    // read only array, not to be modified    int[] arr = { 1, 1, 2, 3, 5, 4 };     // array of size 6(n + 1) having    // elements between 1 and 5    int n = 5;     System.out.println("One of the numbers repeated in the array is: " +                                    findRepeatingNumber(arr, n));}} // This code is contributed by rachana soma

Python3

 # Python 3program to find one of the repeating# elements in a read only arrayfrom math import sqrt # Function to find one of the repeating# elementsdef findRepeatingNumber(arr, n):         # Size of blocks except the    # last block is sq    sq = sqrt(n)     # Number of blocks to incorporate 1 to    # n values blocks are numbered from 0    # to range-1 (both included)    range__= int((n / sq) + 1)     # Count array maintains the count for    # all blocks    count = [0 for i in range(range__)]     # Traversing the read only array and    # updating count    for i in range(0, n + 1, 1):                 # arr[i] belongs to block number        # (arr[i]-1)/sq i is considered        # to start from 0        count[int((arr[i] - 1) / sq)] += 1     # The selected_block is set to last    # block by default. Rest of the blocks    # are checked    selected_block = range__ - 1    for i in range(0, range__ - 1, 1):        if (count[i] > sq):            selected_block = i            break             # after finding block with size > sq    # method of hashing is used to find    # the element repeating in this block    m = {i:0 for i in range(n)}    for i in range(0, n + 1, 1):                 # checks if the element belongs        # to the selected_block        if (((selected_block * sq) < arr[i]) and             (arr[i] <= ((selected_block + 1) * sq))):            m[arr[i]] += 1             # repeating element found            if (m[arr[i]] > 1):                return arr[i]     # return -1 if no repeating element exists    return -1 # Driver Codeif __name__ == '__main__':         # read only array, not to be modified    arr = [1, 1, 2, 3, 5, 4]     # array of size 6(n + 1) having    # elements between 1 and 5    n = 5     print("One of the numbers repeated in the array is:",                             findRepeatingNumber(arr, n))     # This code is contributed by# Sahil_shelangia

C#

 // C# program to find one of the repeating// elements in a read only arrayusing System;using System.Collections.Generic; class GFG{     // Function to find one of the repeating    // elements    static int findRepeatingNumber(int[] arr, int n)    {        // Size of blocks except the        // last block is sq        int sq = (int) Math.Sqrt(n);         // Number of blocks to incorporate 1 to        // n values blocks are numbered from 0        // to range-1 (both included)        int range = (n / sq) + 1;         // Count array maintains the count for        // all blocks        int[] count = new int[range];         // Traversing the read only array and        // updating count        for (int i = 0; i <= n; i++)        {            // arr[i] belongs to block number            // (arr[i]-1)/sq i is considered            // to start from 0            count[(arr[i] - 1) / sq]++;        }         // The selected_block is set to last        // block by default. Rest of the blocks        // are checked        int selected_block = range - 1;        for (int i = 0; i < range - 1; i++)        {            if (count[i] > sq)            {                selected_block = i;                break;            }        }         // after finding block with size > sq        // method of hashing is used to find        // the element repeating in this block        Dictionary m = new Dictionary();        for (int i = 0; i <= n; i++)        {            // checks if the element belongs to the            // selected_block            if ( ((selected_block * sq) < arr[i]) &&                    (arr[i] <= ((selected_block + 1) * sq)))            {                m.Add(arr[i], 1);                 // repeating element found                if (m[arr[i]] == 1)                    return arr[i];            }        }         // return -1 if no repeating element exists        return -1;    } // Driver codepublic static void Main(String []args){    // read only array, not to be modified    int[] arr = { 1, 1, 2, 3, 5, 4 };     // array of size 6(n + 1) having    // elements between 1 and 5    int n = 5;     Console.WriteLine("One of the numbers repeated in the array is: " +                                    findRepeatingNumber(arr, n));}} // This code contributed by Rajput-Ji

Javascript



Output:

One of the numbers repeated in the array is: 1

Time Complexity: O(N)
Auxiliary Space: sqrt(N)
This article is contributed by Aanya Jindal. If you like GeeksforGeeks and would like to contribute, you can also write an article using write.geeksforgeeks.org or mail your article to review-team@geeksforgeeks.org. See your article appearing on the GeeksforGeeks main page and help other Geeks.