Tutorialspoint.dev

Largest BST in a Binary Tree | Set 2

Given a Binary Tree, write a function that returns the size of the largest subtree which is also a Binary Search Tree (BST). If the complete Binary Tree is BST, then return the size of whole tree.

Examples:

Input: 
      5
    /  
   2    4
 /  
1    3

Output: 3 
The following subtree is the 
maximum size BST subtree 
   2  
 /  
1    3


Input: 
       50
     /    
  30       60
 /       /   
5   20   45    70
              /  
            65    80
Output: 5
The following subtree is the
maximum size BST subtree 
      60
     /   
   45    70
        /  
      65    80

We have discussed a two methods in below post.
Find the largest BST subtree in a given Binary Tree | Set 1

In this post a different O(n) solution is discussed. This solution is simpler than the solutions discussed above and works in O(n) time.

The idea is based on method 3 of check if a binary tree is BST article.



A Tree is BST if following is true for every node x.

  1. The largest value in left subtree (of x) is smaller than value of x.
  2. The smallest value in right subtree (of x) is greater than value of x.

We traverse tree in bottom up manner. For every traversed node, we return maximum and minimum values in subtree rooted with it. If any node follows above properties and size of

C++

// C++ program to find largest BST in a
// Binary Tree.
#include <bits/stdc++.h>
using namespace std;
  
/* A binary tree node has data,
pointer to left child and a
pointer to right child */
struct Node
{
    int data;
    struct Node* left;
    struct Node* right;
};
  
/* Helper function that allocates a new
node with the given data and NULL left
and right pointers. */
struct Node* newNode(int data)
{
    struct Node* node = new Node;
    node->data = data;
    node->left = node->right = NULL;
  
    return(node);
}
  
// Information to be returned by every
// node in bottom up traversal.
struct Info
{
    int sz; // Size of subtree
    int max; // Min value in subtree
    int min; // Max value in subtree
    int ans; // Size of largest BST which
    // is subtree of current node
    bool isBST; // If subtree is BST
};
  
// Returns Information about subtree. The
// Information also includes size of largest
// subtree which is a BST.
Info largestBSTBT(Node* root)
{
    // Base cases : When tree is empty or it has
    // one child.
    if (root == NULL)
        return {0, INT_MIN, INT_MAX, 0, true};
    if (root->left == NULL && root->right == NULL)
        return {1, root->data, root->data, 1, true};
  
    // Recur for left subtree and right subtrees
    Info l = largestBSTBT(root->left);
    Info r = largestBSTBT(root->right);
  
    // Create a return variable and initialize its
    // size.
    Info ret;
    ret.sz = (1 + l.sz + r.sz);
  
    // If whole tree rooted under current root is
    // BST.
    if (l.isBST && r.isBST && l.max < root->data &&
            r.min > root->data)
    {
        ret.min = min(l.min, min(r.min, root->data));
        ret.max = max(r.max, max(l.max, root->data));
  
        // Update answer for tree rooted under
        // current 'root'
        ret.ans = ret.sz;
        ret.isBST = true;
  
        return ret;
    }
  
    // If whole tree is not BST, return maximum
    // of left and right subtrees
    ret.ans = max(l.ans, r.ans);
    ret.isBST = false;
  
    return ret;
}
  
/* Driver program to test above functions*/
int main()
{
    /* Let us construct the following Tree
        60
       
      65  70
     /
    50 */
  
    struct Node *root = newNode(60);
    root->left = newNode(65);
    root->right = newNode(70);
    root->left->left = newNode(50);
    printf(" Size of the largest BST is %d ",
           largestBSTBT(root).ans);
    return 0;
}
  
// This code is contributed by Vivek Garg in a
// comment on below set 1.
// www.geeksforgeeks.org/find-the-largest-subtree-in-a-tree-that-is-also-a-bst/

Python3

# Python program to find largest 
# BST in a Binary Tree.
  
INT_MIN = -2147483648
INT_MAX = 2147483647
  
# Helper function that allocates a new 
# node with the given data and None left 
# and right pointers. 
class newNode: 
  
    # Constructor to create a new node 
    def __init__(self, data): 
        self.data = data 
        self.left = None
        self.right = None
  
# Returns Information about subtree. The 
# Information also includes size of largest 
# subtree which is a BST
def largestBSTBT(root):
      
# Base cases : When tree is empty or it has 
    # one child. 
    if (root == None):
        return 0, INT_MIN, INT_MAX, 0, True
    if (root.left == None and root.right == None) :
        return 1, root.data, root.data, 1, True
  
    # Recur for left subtree and right subtrees 
    l = largestBSTBT(root.left) 
    r = largestBSTBT(root.right) 
  
    # Create a return variable and initialize its 
    # size. 
    ret = [0, 0, 0, 0, 0
    ret[0] = (1 + l[0] + r[0]) 
  
    # If whole tree rooted under current root is 
    # BST. 
    if (l[4] and r[4] and l[1] < 
        root.data and r[2] > root.data) :
      
        ret[2] = min(l[2], min(r[2], root.data)) 
        ret[1] = max(r[1], max(l[1], root.data)) 
  
        # Update answer for tree rooted under 
        # current 'root' 
        ret[3] = ret[0
        ret[4] = True
  
        return ret 
      
  
    # If whole tree is not BST, return maximum 
    # of left and right subtrees 
    ret[3] = max(l[3], r[3]) 
    ret[4] = False
  
    return ret
  
# Driver Code 
if __name__ == '__main__'
      
    """Let us construct the following Tree
        60 
        /  
        65 70 
    
    50 """
    root = newNode(60
    root.left = newNode(65
    root.right = newNode(70
    root.left.left = newNode(50)
    print("Size of the largest BST is",
                    largestBSTBT(root)[3]) 
                              
# This code is contributed
# Shubham Singh(SHUBHAMSINGH10)

Output:

Size of largest BST is 2

Time Complexity : O(n)

Please write comments if you find anything incorrect, or you want to share more information about the topic discussed above.



This article is attributed to GeeksforGeeks.org

leave a comment

code

0 Comments

load comments

Subscribe to Our Newsletter