Binary Search Tree insert with Parent Pointer in C++

We can insert new node into the BST in recursive manner. In that case we return the address of the root of each subtree. Here we will see another approach, where parent pointer will need to be maintained. Parent pointer will be helpful to find ancestor of a node etc.

The idea is to store the address of the left and right subtrees, we set parent pointers of the returned pointers after recursive call. This confirms that all parent pointers are set during insertion. The parent of root is set to null.

Algorithm

insert(node, key) −

begin
   if node is null, then create a new node and return
      if the key is less than the key of node, then
         create a new node with key
         add the new node with the left pointer or node
      else if key is greater or equal to the key of node, then
            create a new node with key
         add the new node at the right pointer of the node
      end if
   return node
end

Example

#include<iostream>
using namespace std;
class Node {
   public:
      int data;
      Node *left, *right, *parent;
};
struct Node *getNode(int item) {
   Node *temp = new Node;
   temp->data = item;
   temp->left = temp->right = temp->parent = NULL;
   return temp;
}
void inorderTraverse(struct Node *root) {
   if (root != NULL) {
      inorderTraverse(root->left);
      cout << root->data << " ";
      if (root->parent == NULL)
         cout << "NULL" << endl;
      else
         cout << root->parent->data << endl;
      inorderTraverse(root->right);
   }
}
struct Node* insert(struct Node* node, int key) {
   if (node == NULL) return getNode(key);
   if (key < node->data) { //to the left subtree
      Node *left_child = insert(node->left, key);
      node->left = left_child;
      left_child->parent = node;
   }
   else if (key > node->data) { // to the right subtree
      Node *right_child = insert(node->right, key);
      node->right = right_child;
      right_child->parent = node;
   }
   return node;
}
int main() {
   struct Node *root = NULL;
   root = insert(root, 100);
   insert(root, 60);
   insert(root, 40);
   insert(root, 80);
   insert(root, 140);
   insert(root, 120);
   insert(root, 160);
   inorderTraverse(root);
}

Output

40 60
60 100
80 60
100 NULL
120 140
140 100
160 140