Data Modelling & AI Data Structure & Algorithm

Number of elements less than or equal to a number in a subarray : MO’s Algorithm

23 June 2025

0

Given an array arr of size N and Q queries of the form L, R and X, the task is to print the number of elements less than or equal to X in the subarray represented by L to R.

Prerequisites: MO’s Algorithm, Sqrt Decomposition

Examples:

Input: 
arr[] = {2, 3, 4, 5}
Q = {{0, 3, 5}, {0, 2, 2}}
Output:
 4
 1
Explanation:
Number of elements less than or equal to 5
in arr[0..3] is 4 (all elements)

Number of elements less than or equal to 2 
in arr[0..2] is 1 (only 2)

Approach:
The idea of MO’s algorithm is to pre-process all queries so that result of one query can be used in the next query.
Let arr[0…n-1] be input array and Q[0..m-1] be an array of queries.

Sort all queries in a way that queries with L values from 0 to ?n – 1 are put together, then all queries from ?n to 2×?n – 1, and so on. All queries within a block are sorted in increasing order of R values.
Process all queries one by one in a way that every query uses result computed in the previous query.
We will maintain the frequency array that will count the frequency of arr[i] as they appear in the range [L, R].
For example: arr[]=[3, 4, 6, 2, 7, 1], L=0, R=4 and X=5

Initially frequency array is initialized to 0 i.e freq[]=[0….0]
Step 1– add arr[0] and increment its frequency as freq[arr[0]]++ i.e freq[3]++
and freq[]=[0, 0, 0, 1, 0, 0, 0, 0]
Step 2– Add arr[1] and increment freq[arr[1]]++ i.e freq[4]++
and freq[]=[0, 0, 0, 1, 1, 0, 0, 0]
Step 3– Add arr[2] and increment freq[arr[2]]++ i.e freq[6]++
and freq[]=[0, 0, 0, 1, 1, 0, 1, 0]
Step 4– Add arr[3] and increment freq[arr[3]]++ i.e freq[2]++
and freq[]=[0, 0, 1, 1, 1, 0, 1, 0]
Step 5– Add arr[4] and increment freq[arr[4]]++ i.e freq[7]++
and freq[]=[0, 0, 1, 1, 1, 0, 1, 1]
Step 6– Now we need to find the numbers of elements less than or equal to X(here X=5).
Step 7– The answer is equal to $\sum_{i=0}^{X} freq[i]$
To calculate the sum in step 7, we cannot do iteration because that would lead to O(N) time complexity per query so we will use sqrt decomposition technique to find the sum whose time complexity is O(?n) per query.

C++

// C++ program to answer queries to 
// count number of elements smaller 
// than or equal to x.
#include <bits/stdc++.h>
using namespace std;
 
#define MAX 100001
#define SQRSIZE 400
 
// Variable to represent block size.
// This is made global so compare()
// of sort can use it.
int query_blk_sz;
 
// Structure to represent a
// query range
struct Query {
 
    int L;
    int R;
    int x;
};
 
// Frequency array
// to keep count of elements
int frequency[MAX];
 
// Array which contains the frequency
// of a particular block
int blocks[SQRSIZE];
 
// Block size
int blk_sz;
 
// Function used to sort all queries 
// so that all queries of the same
// block are arranged together and 
// within a block, queries are sorted 
// in increasing order of R values.
bool compare(Query x, Query y)
{
    if (x.L / query_blk_sz != 
        y.L / query_blk_sz)
        return x.L / query_blk_sz < 
               y.L / query_blk_sz;
 
    return x.R < y.R;
}
 
// Function used to get the block
// number of current a[i] i.e ind
int getblocknumber(int ind)
{
    return (ind) / blk_sz;
}
 
// Function to get the answer
// of range [0, k] which uses the
// sqrt decomposition technique
int getans(int k)
{
    int ans = 0;
    int left_blk, right_blk;
    left_blk = 0;
    right_blk = getblocknumber(k);
 
    // If left block is equal to
    // right block then we can traverse
    // that block
    if (left_blk == right_blk) {
        for (int i = 0; i <= k; i++)
            ans += frequency[i];
    }
    else {
        // Traversing first block in 
        // range
        for (int i = 0; i < 
             (left_blk + 1) * blk_sz; i++)
            ans += frequency[i];
 
        // Traversing completely overlapped
        // blocks in range
        for (int i = left_blk + 1; 
             i < right_blk; i++)
            ans += blocks[i];
 
        // Traversing last block in range
        for (int i = right_blk * blk_sz;
             i <= k; i++)
            ans += frequency[i];
    }
    return ans;
}
 
void add(int ind, int a[])
{
    // Increment the frequency of a[ind]
    // in the frequency array
    frequency[a[ind]]++;
 
    // Get the block number of a[ind]
    // to update the result in blocks
    int block_num = getblocknumber(a[ind]);
 
    blocks[block_num]++;
}
void remove(int ind, int a[])
{
    // Decrement the frequency of 
    // a[ind] in the frequency array
    frequency[a[ind]]--;
 
    // Get the block number of a[ind]
    // to update the result in blocks
    int block_num = getblocknumber(a[ind]);
 
    blocks[block_num]--;
}
void queryResults(int a[], int n,
                  Query q[], int m)
{
 
    // Initialize the block size
    // for queries
    query_blk_sz = sqrt(m);
 
    // Sort all queries so that queries
    // of same blocks are arranged
    // together.
    sort(q, q + m, compare);
 
    // Initialize current L,
    // current R and current result
    int currL = 0, currR = 0;
 
    for (int i = 0; i < m; i++) {
 
        // L and R values of the 
        // current range
 
        int L = q[i].L, R = q[i].R,
                x = q[i].x;
 
        // Add Elements of current
        // range
        while (currR <= R) {
            add(currR, a);
            currR++;
        }
        while (currL > L) {
            add(currL - 1, a);
            currL--;
        }
 
        // Remove element of previous
        // range
        while (currR > R + 1)
 
        {
            remove(currR - 1, a);
            currR--;
        }
        while (currL < L) {
            remove(currL, a);
            currL++;
        }
        printf("query[%d, %d, %d] : %d\n", 
               L, R, x, getans(x));
    }
}
// Driver code
int main()
{
 
    int arr[] = { 2, 0, 3, 1, 4, 2, 5, 11 };
    int N = sizeof(arr) / sizeof(arr[0]);
 
    blk_sz = sqrt(N);
    Query Q[] = { { 0, 2, 2 }, { 0, 3, 5 },
                { 5, 7, 10 } };
 
    int M = sizeof(Q) / sizeof(Q[0]);
     
    // Answer the queries
    queryResults(arr, N, Q, M);
    return 0;
}

Java

import java.util.*;
 
public class GFG{
 
  static int MAX = 100001;
  static int SQRSIZE = 400;
 
  // Variable to represent block size.
  // This is made global so compare()
  // of sort can use it.
  static int query_blk_sz;
 
  // Frequency array
  // to keep count of elements
  static int[] frequency = new int[MAX];
 
  // Array which contains the frequency
  // of a particular block
  static int[] blocks = new int[SQRSIZE];
  static int blk_sz;
 
  // class to represent a
  // query range
  static class Query {
    int L;
    int R;
    int x;
    Query(int a,int b,int c)
    {
      L = a;
      R = b;
      x = c;
    }
  }
 
  // Function used to get the block
  // number of current a[i] i.e ind
  static int getblocknumber(int ind) {
    return ind / blk_sz;
  }
 
  // Function to get the answer
  // of range [0, k] which uses the
  // sqrt decomposition technique
  static int getans(int k) {
    int ans = 0;
    int left_blk, right_blk;
    left_blk = 0;
    right_blk = getblocknumber(k);
 
    // If left block is equal to
    // right block then we can traverse
    // that block
    if (left_blk == right_blk) {
      for (int i = 0; i <= k; i++) {
        ans += frequency[i];
      }
    } else {
      // Traversing first block in
      // range
      for (int i = 0; i < (left_blk + 1) * blk_sz; i++) {
        ans += frequency[i];
      }
      // Traversing completely overlapped
      // blocks in range
      for (int i = left_blk + 1; i < right_blk; i++) {
        ans += blocks[i];
      }
      // Traversing last block in range
      for (int i = right_blk * blk_sz; i <= k; i++) {
        ans += frequency[i];
      }
    }
    return ans;
  }
 
  static void add(int ind, int[] a) {
    // Increment the frequency of a[ind]
    // in the frequency array
    frequency[a[ind]]++;
    // Get the block number of a[ind]
    // to update the result in blocks
    int block_num = getblocknumber(a[ind]);
    blocks[block_num]++;
  }
 
  static void remove(int ind, int[] a) 
  {
 
    // Decrement the frequency of
    // a[ind] in the frequency array
    frequency[a[ind]]--;
 
    // Get the block number of a[ind]
    // to update the result in blocks
    int block_num = getblocknumber(a[ind]);
    blocks[block_num]--;
  }
 
  static void queryResults(int[] a, int n, Query[] q, int m) {
    query_blk_sz = (int) Math.sqrt(m);
    Arrays.sort(q, new Comparator<Query>() {
      @Override
 
      // Function used to sort all queries
      // so that all queries of the same
      // block are arranged together and
      // within a block, queries are sorted
      // in increasing order of R values.
      public int compare(Query x, Query y) {
        if (x.L / query_blk_sz != y.L / query_blk_sz) {
          return x.L / query_blk_sz - y.L / query_blk_sz;
        }
        return x.R - y.R;
      }
    });
 
    // Initialize current L,
    // current R and current result
    int currL = 0, currR = 0;
    for (int i = 0; i < m; i++) 
    {
 
      // L and R values of the
      // current range
      int L = q[i].L, R = q[i].R, x = q[i].x;
      while (currR <= R) {
        add(currR, a);
        currR++;
      }
      while (currL > L) {
        add(currL - 1, a);
        currL--;
      }
      while (currR > R + 1) {
        remove(currR - 1, a);
        currR--;
      }
      while (currL < L) {
        remove(currL, a);
        currL++;
      }
      System.out.println("query[" + L + ", " + R + ", " + x + "] : " + getans(x));
 
    }
  }
 
  public static void main(String[] args) {
    int arr[] = { 2, 0, 3, 1, 4, 2, 5, 11 };
    int n = arr.length;
    blk_sz = (int)Math.sqrt(n);
    Query q[] ={new Query(0,2,2),new Query(0,3,5),new Query(5,7,10)};
    int m = q.length;
    queryResults(arr,n,q,m);
 
  }
}
 
// This code is contributed by anskalyan3

Python3

# Python equivalent
MAX = 100001
SQRSIZE = 400
 
# Variable to represent block size.
# This is made global so compare()
# of sort can use it.
query_blk_sz = 0
 
# Frequency array
# to keep count of elements
frequency = [0] * MAX
 
# Array which contains the frequency
# of a particular block
blocks = [0] * SQRSIZE
blk_sz = 0
 
# class to represent a
# query range
class Query:
    def __init__(self, a, b, c):
        self.L = a
        self.R = b
        self.x = c
 
# Function used to get the block
# number of current a[i] i.e ind
def getblocknumber(ind):
    return ind // blk_sz
 
# Function to get the answer
# of range [0, k] which uses the
# sqrt decomposition technique
def getans(k):
    ans = 0
    left_blk, right_blk = 0, getblocknumber(k)
 
    # If left block is equal to
    # right block then we can traverse
    # that block
    if left_blk == right_blk:
        for i in range(0, k+1):
            ans += frequency[i]
    else:
        # Traversing first block in
        # range
        for i in range(0, (left_blk + 1) * blk_sz):
            ans += frequency[i]
        # Traversing completely overlapped
        # blocks in range
        for i in range(left_blk + 1, right_blk):
            ans += blocks[i]
        # Traversing last block in range
        for i in range(right_blk * blk_sz, k+1):
            ans += frequency[i]
    return ans
 
def add(ind, a):
    # Increment the frequency of a[ind]
    # in the frequency array
    frequency[a[ind]] += 1
    # Get the block number of a[ind]
    # to update the result in blocks
    block_num = getblocknumber(a[ind])
    blocks[block_num] += 1
 
def remove(ind, a):
    # Decrement the frequency of
    # a[ind] in the frequency array
    frequency[a[ind]] -= 1
    # Get the block number of a[ind]
    # to update the result in blocks
    block_num = getblocknumber(a[ind])
    blocks[block_num] -= 1
 
def queryResults(a, n, q, m):
    query_blk_sz = int(m ** 0.5)
    q.sort(key = lambda x: (x.L // query_blk_sz, x.R))
 
    # Initialize current L,
    # current R and current result
    currL = 0
    currR = 0
    for i in range(m):
        # L and R values of the
        # current range
        L = q[i].L
        R = q[i].R
        x = q[i].x
        while currR <= R:
            add(currR, a)
            currR += 1
        while currL > L:
            add(currL - 1, a)
            currL -= 1
        while currR > R + 1:
            remove(currR - 1, a)
            currR -= 1
        while currL < L:
            remove(currL, a)
            currL += 1
        print("query[{}, {}, {}] : {}".format(L, R, x, getans(x)))
 
def main():
    global blk_sz
    arr = [2, 0, 3, 1, 4, 2, 5, 11]
    n = len(arr)
    blk_sz = int(n ** 0.5)
    q = [Query(0,2,2),Query(0,3,5),Query(5,7,10)]
    m = len(q)
    queryResults(arr,n,q,m)
 
if __name__ == "__main__":
    main()

C#

using System;
using System.Collections.Generic;
 
class GFG {
    static int MAX = 100001;
    static int SQRSIZE = 400;
 
    // Variable to represent block size.
    // This is made global so compare()
    // of sort can use it.
    static int query_blk_sz;
 
    // Frequency array
    // to keep count of elements
    static int[] frequency = new int[MAX];
 
    // Array which contains the frequency
    // of a particular block
    static int[] blocks = new int[SQRSIZE];
    static int blk_sz;
 
    // class to represent a
    // query range
    public class Query {
        public int L;
        public int R;
        public int x;
 
        public Query(int a, int b, int c)
        {
            L = a;
            R = b;
            x = c;
        }
    }
 
    // Function used to get the block
    // number of current a[i] i.e ind
    static int getblocknumber(int ind)
    {
        return ind / blk_sz;
    }
 
    // Function to get the answer
    // of range [0, k] which uses the
    // sqrt decomposition technique
    static int getans(int k)
    {
        int ans = 0;
        int left_blk, right_blk;
        left_blk = 0;
        right_blk = getblocknumber(k);
 
        // If left block is equal to
        // right block then we can traverse
        // that block
        if (left_blk == right_blk) {
            for (int i = 0; i <= k; i++) {
                ans += frequency[i];
            }
        }
        else {
            // Traversing first block in
            // range
            for (int i = 0; i < (left_blk + 1) * blk_sz;
                 i++) {
                ans += frequency[i];
            }
            // Traversing completely overlapped
            // blocks in range
            for (int i = left_blk + 1; i < right_blk; i++) {
                ans += blocks[i];
            }
            // Traversing last block in range
            for (int i = right_blk * blk_sz; i <= k; i++) {
                ans += frequency[i];
            }
        }
        return ans;
    }
 
    static void add(int ind, int[] a)
    {
        // Increment the frequency of a[ind]
        // in the frequency array
        frequency[a[ind]]++;
        // Get the block number of a[ind]
        // to update the result in blocks
        int block_num = getblocknumber(a[ind]);
        blocks[block_num]++;
    }
 
    static void remove(int ind, int[] a)
    {
 
        // Decrement the frequency of
        // a[ind] in the frequency array
        frequency[a[ind]]--;
 
        // Get the block number of a[ind]
        // to update the result in blocks
        int block_num = getblocknumber(a[ind]);
        blocks[block_num]--;
    }
 
    static void queryResults(int[] a, int n, Query[] q,
                             int m)
    {
        query_blk_sz = (int)Math.Sqrt(m);
        Array.Sort(q, new Comparison<Query>((x, y) => {
                       if (x.L / query_blk_sz
                           != y.L / query_blk_sz) {
                           return x.L / query_blk_sz
                               - y.L / query_blk_sz;
                       }
                       return x.R - y.R;
                   }));
 
        // Initialize current L,
        // current R and current result
        int currL = 0, currR = 0;
        for (int i = 0; i < m; i++) {
            // L and R values of the
            // current range
            int L = q[i].L, R = q[i].R, x = q[i].x;
            while (currR <= R) {
                add(currR, a);
                currR++;
            }
            while (currL > L) {
                add(currL - 1, a);
                currL--;
            }
            while (currR > R + 1) {
                remove(currR - 1, a);
                currR--;
            }
            while (currL < L) {
                remove(currL, a);
                currL++;
            }
            Console.WriteLine("query[" + L + ", " + R + ", "
                              + x + "] : " + getans(x));
        }
    }
     
      // Driver code
    static void Main(string[] args)
    {
        int[] arr = { 2, 0, 3, 1, 4, 2, 5, 11 };
        int n = arr.Length;
        blk_sz = (int)Math.Sqrt(n);
        Query[] q
            = { new Query(0, 2, 2), new Query(0, 3, 5),
                new Query(5, 7, 10) };
        int m = q.Length;
          // Function call
        queryResults(arr, n, q, m);
    }
}
 
// This code is contributed by phasing17

Javascript

const MAX = 100001;
const SQRSIZE = 400;
let query_blk_sz = 0;
let frequency = new Array(MAX).fill(0);
let blocks = new Array(SQRSIZE).fill(0);
let blk_sz = 0;
 
class Query {
  constructor(a, b, c) {
    this.L = a;
    this.R = b;
    this.x = c;
  }
}
 
function getblocknumber(ind) {
  return Math.floor(ind / blk_sz);
}
 
function getans(k) {
  let ans = 0;
  let left_blk = 0;
  let right_blk = getblocknumber(k);
  if (left_blk == right_blk) {
    for (let i = 0; i <= k; i++) {
      ans += frequency[i];
    }
  } else {
    for (let i = 0; i <= (left_blk + 1) * blk_sz - 1; i++) {
      ans += frequency[i];
    }
    for (let i = left_blk + 1; i <= right_blk - 1; i++) {
      ans += blocks[i];
    }
    for (let i = right_blk * blk_sz; i <= k; i++) {
      ans += frequency[i];
    }
  }
  return ans;
}
 
function add(ind, a) {
  frequency[a[ind]]++;
  let block_num = getblocknumber(a[ind]);
  blocks[block_num]++;
}
 
function remove(ind, a) {
  frequency[a[ind]]--;
  let block_num = getblocknumber(a[ind]);
  blocks[block_num]--;
}
 
function queryResults(a, n, q, m) {
  query_blk_sz = Math.floor(Math.sqrt(m));
  q.sort((a, b) => {
    return a.L / query_blk_sz - b.L / query_blk_sz || a.R - b.R;
  });
  let currL = 0;
  let currR = 0;
  for (let i = 0; i < m; i++) {
    let L = q[i].L;
    let R = q[i].R;
    let x = q[i].x;
    while (currR <= R) {
      add(currR, a);
      currR++;
    }
    while (currL > L) {
      add(currL - 1, a);
      currL--;
    }
    while (currR > R + 1) {
      remove(currR - 1, a);
      currR--;
    }
    while (currL < L) {
      remove(currL, a);
      currL++;
    }
    console.log(`query[${L}, ${R}, ${x}] : ${getans(x)}`);
  }
}
 
function main() {
  let arr = [2, 0, 3, 1, 4, 2, 5, 11];
  let n = arr.length;
  blk_sz = Math.floor(Math.sqrt(n));
  let q = [new Query(0, 2, 2), new Query(0, 3, 5), new Query(5, 7, 10)];
  let m = q.length;
  queryResults(arr, n, q, m);
}
 
main();

Output:

query[0, 2, 2] : 2
query[0, 3, 5] : 4
query[5, 7, 10] : 2

Time Complexity: O(Q × ?N).
It takes O(Q × ?N) time for MO’s algorithm and O(Q × ?N) time for sqrt decomposition technique to answer the sum of freq[0]+….freq[k], therefore total time complexity is O(Q × ?N + Q × ?N) which is O(Q × ?N).

Space Complexity: O(N + sqrt(N) + M), where N is the size of the input array, M is the number of queries, and sqrt(N) is the size of the blocks used in the sqrt decomposition technique.

Feeling lost in the world of random DSA topics, wasting time without progress? It’s time for a change! Join our DSA course, where we’ll guide you on an exciting journey to master DSA efficiently and on schedule.
Ready to dive in? Explore our Free Demo Content and join our DSA course, trusted by over 100,000 neveropen!

Number of elements less than or equal to a number in a subarray : MO’s Algorithm

C++

Java

Python3

C#

Javascript

From Word2Vec to LLM2Vec: How to Choose the Right Embedding Model for RAG

How to Debug Slow Search Requests in Milvus

When Context Engineering Is Done Right, Hallucinations Can Be the Spark of AI Creativity

LEAVE A REPLY Cancel reply

Most Popular

Google defends Android’s controversial sideloading policy

Nothing’s new AI Tools let you build apps with just a prompt

Apple iPhone 17 Pro Max review: As subtle as a sledgehammer, and just as powerful

The offline Google Maps routine that saves me on every road trip

EDITOR PICKS

Google defends Android’s controversial sideloading policy

Nothing’s new AI Tools let you build apps with just a prompt

Apple iPhone 17 Pro Max review: As subtle as a sledgehammer, and just as powerful

POPULAR POSTS

Google defends Android’s controversial sideloading policy

Nothing’s new AI Tools let you build apps with just a prompt

Apple iPhone 17 Pro Max review: As subtle as a sledgehammer, and just as powerful

POPULAR CATEGORY

ABOUT US

FOLLOW US