Related Articles
Maximum sum after rearranging the array for K queries
• Difficulty Level : Medium
• Last Updated : 28 Jul, 2020

Given two arrays arr[] containing N integers and Q[][] containing K queries where every query represents a range [L, R]. The task is to rearrange the array and find the maximum possible sum of all the subarrays where each subarray is defined by the elements of the array in the range [L, R] given by each query.

Note: 1 based indexing is used in the Q[][] array to signify the ranges.

Examples:

Input: arr[] = { 2, 6, 10, 1, 5, 6 }, Q[] = {{1, 3}, {4, 6}, {3, 4}}
Output: 46
Explanation:
One possible way is to rearrange the array to arr[] = {2, 6, 10, 6, 5, 1}.
In this arrangement:
The sum of the subarray in the range [1, 3] = 2 + 6 + 10 = 18.
The sum of the subarray in the range [4, 6] = 6 + 5 + 1 = 12.
The sum of the subarray in the range [3, 4] = 10 + 6 = 16.
The total sum of all the subarrays = 46 which is the maximum possible.

Input: arr[] = { 1, 2, 3, 4, 5, 6, 7, 8 }, Q[] = {{1, 4}, {5, 5}, {7, 8}, {8, 8}}
Output: 43
Explanation:
One possible way is to rearrange the array to arr[] = {2, 3, 4, 5, 6, 1, 7, 8}.
In this arrangement:
The sum of the subarray in the range [1, 4] = 2 + 3 + 4 + 5 = 14.
The sum of the subarray in the range [5, 5] = 6 = 6.
The sum of the subarray in the range [7, 8] = 7 + 8 = 15.
The sum of the subarray in the range [8, 8] = 8 = 8.
The total sum of all the subarrays = 43 which is the maximum possible.

## Recommended: Please try your approach on {IDE} first, before moving on to the solution.

Approach: On observing clearly, one conclusion which can be made is that we get the maximum sum when the maximum elements are included in as many subarrays as possible. For this, we need to find the number of times every index is included by iterating all the queries.

For example: Let the array be arr[] = {2, 6, 10, 6, 5, 1} and the queries be Q[][] = {{1, 3}, {4, 6}, {3, 4}}.

1. Step 1: Create a count array C[] of size N. So, initially, the count array C[] = {0, 0, 0, 0, 0, 0}.
2. Step 2: For the query [1, 3], the elements at the index [1, 3] are incremented by 1. The count array after this query becomes {1, 1, 1, 0, 0, 0}.
3. Step 3: Similarly, for the next query, the count array becomes {1, 1, 1, 1, 1, 1} and finally, after the third query, the count array becomes {1, 1, 2, 2, 1, 1}.
4. Step 4: After obtaining the count array, the idea is to use sorting to get the maximum sum.
5. Step 5: After sorting, the array C[] = {1, 1, 1, 1, 2, 2} and arr[] = {1, 2, 5, 6, 6, 10}. The maximum possible sum is the weighted sum of both the arrays, i.e.:

sum = ((1 * 1) + (1 * 2) + (1 * 5) + (1 * 6) + (2 * 6) + (2 * 10)) = 46

Below is the implementation of the above approach:

## C++

 `// C++ program to find the maximum sum``// after rearranging the array for K queries`` ` `#include ``using` `namespace` `std;`` ` `// Function to find maximum sum after``// rearranging array elements``int` `maxSumArrangement(``int` `A[], ``int` `R[],``                      ``int` `N, ``int` `M)``{`` ` `    ``// Auxiliary array to find the``    ``// count of each selected elements``    ``int` `count[N];`` ` `    ``// Initialize with 0``    ``memset``(count, 0, ``sizeof` `count);`` ` `    ``// Finding count of every element``    ``// to be selected``    ``for` `(``int` `i = 0; i < M; ++i) {`` ` `        ``int` `l = R[i], r = R[i] + 1;`` ` `        ``// Making it to 0-indexing``        ``l--;``        ``r--;`` ` `        ``// Prefix sum array concept is used``        ``// to obtain the count array``        ``count[l]++;`` ` `        ``if` `(r < N)``            ``count[r]--;``    ``}`` ` `    ``// Iterating over the count array``    ``// to get the final array``    ``for` `(``int` `i = 1; i < N; ++i) {``        ``count[i] += count[i - 1];``    ``}`` ` `    ``// Variable to store the maximum sum``    ``int` `ans = 0;`` ` `    ``// Sorting both the arrays``    ``sort(count, count + N);``    ``sort(A, A + N);`` ` `    ``// Loop to find the maximum sum``    ``for` `(``int` `i = N - 1; i >= 0; --i) {``        ``ans += A[i] * count[i];``    ``}`` ` `    ``return` `ans;``}`` ` `// Driver code``int` `main()``{``    ``int` `A[] = { 2, 6, 10, 1, 5, 6 };``    ``int` `R[]``        ``= { { 1, 3 }, { 4, 6 }, { 3, 4 } };`` ` `    ``int` `N = ``sizeof``(A) / ``sizeof``(A);``    ``int` `M = ``sizeof``(R) / ``sizeof``(R);`` ` `    ``cout << maxSumArrangement(A, R, N, M);`` ` `    ``return` `0;``}`

## Java

 `// Java program to find the maximum sum``// after rearranging the array for K queries``import` `java.util.*;`` ` `class` `GFG``{``      ` `    ``// Function to find maximum sum after``    ``// rearranging array elements``    ``static` `int` `maxSumArrangement(``int` `A[], ``int` `R[][],``                          ``int` `N, ``int` `M)``    ``{``      ` `        ``// Auxiliary array to find the``        ``// count of each selected elements``        ``int` `count[] = ``new` `int``[N];``        ``int` `i;``      ` `        ``// Finding count of every element``        ``// to be selected``        ``for` `( i = ``0``; i < M; ++i) {``      ` `            ``int` `l = R[i][``0``], r = R[i][``1``] + ``1``;``      ` `            ``// Making it to 0-indexing``            ``l--;``            ``r--;``      ` `            ``// Prefix sum array concept is used``            ``// to obtain the count array``            ``count[l]++;``      ` `            ``if` `(r < N)``                ``count[r]--;``        ``}``      ` `        ``// Iterating over the count array``        ``// to get the final array``        ``for` `(i = ``1``; i < N; ++i) {``            ``count[i] += count[i - ``1``];``        ``}``      ` `        ``// Variable to store the maximum sum``        ``int` `ans = ``0``;``      ` `        ``// Sorting both the arrays``        ``Arrays.sort( count);``        ``Arrays.sort(A);``      ` `        ``// Loop to find the maximum sum``        ``for` `(i = N - ``1``; i >= ``0``; --i) {``            ``ans += A[i] * count[i];``        ``}``      ` `        ``return` `ans;``    ``}``      ` `    ``// Driver code``    ``public` `static` `void` `main(String []args)``    ``{``        ``int` `A[] = { ``2``, ``6``, ``10``, ``1``, ``5``, ``6` `};``        ``int` `R[][]``            ``= { { ``1``, ``3` `}, { ``4``, ``6` `}, { ``3``, ``4` `} };``      ` `        ``int` `N = A.length;``        ``int` `M = R.length;``      ` `        ``System.out.print(maxSumArrangement(A, R, N, M));``      ` `    ``}``}`` ` `// This code is contributed by chitranayal`

## Python3

 `# Python3 program to find the maximum sum ``# after rearranging the array for K queries `` ` `# Function to find maximum sum after ``# rearranging array elements ``def` `maxSumArrangement( A,  R,  N,  M):`` ` `    ``# Auxiliary array to find the ``    ``# count of each selected elements ``    ``# Initialize with 0 ``    ``count ``=` `[``0` `for` `i ``in` `range``(N)]`` ` `    ``# Finding count of every element ``    ``# to be selected ``    ``for` `i ``in` `range``(M):`` ` `        ``l ``=` `R[i][``0``]``        ``r ``=` `R[i][``1``] ``+` `1`` ` `        ``# Making it to 0-indexing ``        ``l ``=` `l ``-` `1` `        ``r ``=` `r ``-` `1`` ` `        ``# Prefix sum array concept is used ``        ``# to obtain the count array ``        ``count[l] ``=` `count[l] ``+` `1` ` ` `        ``if` `(r < N):``            ``count[r] ``=` `count[r] ``-` `1` ` ` `    ``# Iterating over the count array ``    ``# to get the final array ``    ``for` `i ``in` `range``(``1``, N): ``        ``count[i] ``=` `count[i] ``+` `count[i ``-` `1``]`` ` `    ``# Variable to store the maximum sum ``    ``ans ``=` `0`` ` `    ``# Sorting both the arrays ``    ``count.sort()``    ``A.sort()`` ` `    ``# Loop to find the maximum sum ``    ``for` `i ``in` `range``(N ``-` `1``, ``-``1``, ``-``1``): ``        ``ans ``=` `ans ``+` `A[i] ``*` `count[i]`` ` `    ``return` `ans`` ` `# Driver code ``A ``=` `[ ``2``, ``6``, ``10``, ``1``, ``5``, ``6` `]``R ``=` `[ [ ``1``, ``3` `], [ ``4``, ``6` `], [ ``3``, ``4` `] ] `` ` `N ``=` `len``(A)``M ``=` `len``(R)`` ` `print``(maxSumArrangement(A, R, N, M))`` ` `# This code is contributed by Sanjit_Prasad`

## C#

 `// C# program to find the maximum sum``// after rearranging the array for K queries``using` `System;`` ` `class` `GFG``{``       ` `    ``// Function to find maximum sum after``    ``// rearranging array elements``    ``static` `int` `maxSumArrangement(``int` `[]A, ``int` `[,]R,``                          ``int` `N, ``int` `M)``    ``{``       ` `        ``// Auxiliary array to find the``        ``// count of each selected elements``        ``int` `[]count = ``new` `int``[N];``        ``int` `i;``       ` `        ``// Finding count of every element``        ``// to be selected``        ``for` `( i = 0; i < M; ++i) {``       ` `            ``int` `l = R[i, 0], r = R[i, 1] + 1;``       ` `            ``// Making it to 0-indexing``            ``l--;``            ``r--;``       ` `            ``// Prefix sum array concept is used``            ``// to obtain the count array``            ``count[l]++;``       ` `            ``if` `(r < N)``                ``count[r]--;``        ``}``       ` `        ``// Iterating over the count array``        ``// to get the readonly array``        ``for` `(i = 1; i < N; ++i) {``            ``count[i] += count[i - 1];``        ``}``       ` `        ``// Variable to store the maximum sum``        ``int` `ans = 0;``       ` `        ``// Sorting both the arrays``        ``Array.Sort( count);``        ``Array.Sort(A);``       ` `        ``// Loop to find the maximum sum``        ``for` `(i = N - 1; i >= 0; --i) {``            ``ans += A[i] * count[i];``        ``}``       ` `        ``return` `ans;``    ``}``       ` `    ``// Driver code``    ``public` `static` `void` `Main(String []args)``    ``{``        ``int` `[]A = { 2, 6, 10, 1, 5, 6 };``        ``int` `[,]R``            ``= { { 1, 3 }, { 4, 6 }, { 3, 4 } };``       ` `        ``int` `N = A.Length;``        ``int` `M = R.GetLength(0);``       ` `        ``Console.Write(maxSumArrangement(A, R, N, M));      ``    ``}``}`` ` `// This code is contributed by Princi Singh`
Output:
```46
```

Time Complexity: O(N* log(N))

My Personal Notes arrow_drop_up