Trees

FindtheclosestleafinaBinaryTree
GivenaBinaryTreeandakeyk,finddistanceoftheclosestleaffromk.
Examples:
A
/\
BC
/\
EF
/\
GH
/\/
IJK
Closestleafto'H'is'K',sodistanceis1for'H'
Closestleafto'C'is'B',sodistanceis2for'C'
Closestleafto'E'iseither'I'or'J',sodistanceis2for'E'
Closestleafto'B'is'B'itself,sodistanceis0for'B'
Westronglyrecommendtominimizeyourbrowserandtrythisyourselffirst
Themainpointtonotehereisthataclosestkeycaneitherbeadescendentofgivenkeyorcanbereached
throughoneoftheancestors.
Theideaistotraversethegiventreeinpreorderandkeeptrackofancestorsinanarray.Whenwereach
thegivenkey,weevaluatedistanceoftheclosestleafinsubtreerootedwithgivenkey.Wealsotraverseall
ancestorsonebyoneandfinddistanceoftheclosestleafinthesubtreerootedwithancestor.Wecompare
alldistancesandreturnminimum.
// A C++ program to find the closesr leaf of a given key in Binary Tree
#include <iostream>
#include <climits>
using namespace std;
/* A binary tree Node has key, pocharer to left and right children */
struct Node
{
char key;
struct Node* left, *right;
};
/* Helper function that allocates a new node with the

given data and NULL left and right pocharers. */
Node *newNode(char k)
{
Node *node = new Node;
node->key = k;
node->right = node->left = NULL;
return node;
}
// A utility function to find minimum of x and y

int getMin(int x, int y)
{
return (x < y)? x :y;
}
// A utility function to find distance of closest leaf of the tree

// rooted under given root
int closestDown(struct Node *root)
{
// Base cases
if (root == NULL)
return INT_MAX;
if (root->left == NULL && root->right == NULL)
return 0;
// Return minimum of left and right, plus one

return 1 + getMin(closestDown(root->left), closestDown(root->right));
}
// Returns distance of the cloest leaf to a given key 'k'. The array
// ancestors is used to keep track of ancestors of current node and
// 'index' is used to keep track of curremt index in 'ancestors[]'
int findClosestUtil(struct Node *root, char k, struct Node *ancestors[],
int index)
{
// Base case
if (root == NULL)
return INT_MAX;
// If key found
if (root->key == k)
{
// Find the cloest leaf under the subtree rooted with given key
int res = closestDown(root);
// Traverse all ancestors and update result if any parent node

// gives smaller distance
for (int i = index-1; i>=0; i--)
res = getMin(res, index - i + closestDown(ancestors[i]));
return res;
}
// If key node found, store current node and recur for left and
// right childrens
ancestors[index] = root;
return getMin(findClosestUtil(root->left, k, ancestors, index+1),
findClosestUtil(root->right, k, ancestors, index+1));
// The main function that returns distance of the closest key to 'k'. It
// mainly uses recursive function findClosestUtil() to find the closes
// distance.
int findClosest(struct Node *root, char k)
{
// Create an array to store ancestors
// Assumptiom: Maximum height of tree is 100
struct Node *ancestors[100];
return findClosestUtil(root, k, ancestors, 0);

}
/* Driver program to test above functions*/

int main()
{
// Let us construct the BST shown in the above figure
struct Node *root
= newNode('A');
root->left
= newNode('B');
root->right
= newNode('C');
root->right->left
= newNode('E');
root->right->right
= newNode('F');
root->right->left->left = newNode('G');
root->right->left->left->left = newNode('I');
root->right->left->left->right = newNode('J');
root->right->right->right
= newNode('H');
root->right->right->right->left = newNode('K');
char k = 'H';
cout << "Distace of the closest key from " << k << " is "
<< findClosest(root, k) << endl;
k = 'C';
k = 'E';
k = 'B';
return 0;
}
Output:
DistaceoftheclosestkeyfromHis1
DistaceoftheclosestkeyfromCis2
DistaceoftheclosestkeyfromEis2
DistaceoftheclosestkeyfromBis0
Theabovecodecanbeoptimizedbystoringtheleft/rightinformationalsoinancestorarray.Theideais,if
givenkeyisinleftsubtreeofanancestors,thenthereisnopointtocallclosestDown().Also,theloopcan
thattraversesancestorsarraycanbeoptimizedtonottraverseancestorswhichareatmoredistancethan
currentresult.
=======================================================
DiagonalSumofaBinaryTree
Considerlinesofslope1passingbetweennodes(dottedlinesinbelowdiagram).Diagonalsuminabinary
treeissumofallnodesdatalyingbetweentheselines.GivenaBinaryTree,printalldiagonalsums.
Forthefollowinginputtree,outputshouldbe9,19,42.
9issumof1,3and5.
19issumof2,6,4and7.
42issumof9,10,11and12.

Westronglyrecommendtominimizeyourbrowserandtrythisyourselffirst
Algorithm:
Theideaistokeeptrackofverticaldistancefromtopdiagonalpassingthroughroot.Weincrementthe
verticaldistancewegodowntonextdiagonal.
1.Addrootwithverticaldistanceas0tothequeue.
2.Processthesumofallrightchildandrightofrightchildandsoon.
3.Addleftchildcurrentnodeintothequeueforlaterprocessing.Theverticaldistanceofleftchildisvertical
distanceofcurrentnodeplus1.
4.Keepdoing2nd,3rdand4thsteptillthequeueisempty.
FollowingisJavaimplementationofaboveidea.
// Java Program to find diagonal sum in a Binary Tree
import java.util.*;
import java.util.Map.Entry;
//Tree node
class TreeNode
{
int data; //node data
int vd; //vertical distance diagonally
TreeNode left, right; //left and right child's reference
// Tree node constructor

public TreeNode(int data)
{
this.data = data;
vd = Integer.MAX_VALUE;
left = right = null;
}
}
// Tree class
class Tree
{
TreeNode root;//Tree root
// Tree constructor
public Tree(TreeNode root) { this.root = root; }
// Diagonal sum method

public void diagonalSum()
{
// Queue which stores tree nodes
Queue<TreeNode> queue = new LinkedList<TreeNode>();
// Map to store sum of node's data lying diagonally

Map<Integer, Integer> map = new TreeMap<>();
// Assign the root's vertical distance as 0.

root.vd = 0;
// Add root node to the queue

queue.add(root);
// Loop while the queue is not empty

while (!queue.isEmpty())
{
// Remove the front tree node from queue.
TreeNode curr = queue.remove();
// Get the vertical distance of the dequeued node.

int vd = curr.vd;
// Sum over this node's right-child, right-of-right-child

// and so on
while (curr != null)
{
int prevSum = (map.get(vd) == null)? 0: map.get(vd);
map.put(vd, prevSum + curr.data);
// If for any node the left child is not null add

// it to the queue for future processing.
if (curr.left != null)
{
curr.left.vd = vd+1;
queue.add(curr.left);
}
// Move to the current node's right child.

curr = curr.right;
}
}
// Make an entry set from map.

Set<Entry<Integer, Integer>> set = map.entrySet();
// Make an iterator
Iterator<Entry<Integer, Integer>> iterator = set.iterator();
// Traverse the map elements using the iterator.

while (iterator.hasNext())
{
Map.Entry<Integer, Integer> me = iterator.next();
System.out.print(me.getValue()+" ");
}
}
}
//Driver class
public class DiagonalSum
{
public static void main(String[] args)
{
TreeNode root = new TreeNode(1);
root.left = new TreeNode(2);
root.right = new TreeNode(3);
root.left.left = new TreeNode(9);
root.left.right = new TreeNode(6);
root.right.left = new TreeNode(4);
root.right.right = new TreeNode(5);
root.right.left.left = new TreeNode(12);
root.right.left.right = new TreeNode(7);
root.left.right.left = new TreeNode(11);
root.left.left.right = new TreeNode(10);
Tree tree = new Tree(root);
tree.diagonalSum();
}
}
Output:
9,19,42
=======================================================
BottomViewofaBinaryTree
GivenaBinaryTree,weneedtoprintthebottomviewfromlefttoright.Anodexisthereinoutputifxisthe
bottommostnodeatitshorizontaldistance.Horizontaldistanceofleftchildofanodexisequaltohorizontal
distanceofxminus1,andthatofrightchildishorizontaldistanceofxplus1.
Examples:
20
/\
822
/\\
5325
/\
1014
Fortheabovetreetheoutputshouldbe5,10,3,14,25.
Iftherearemultiplebottommostnodesforahorizontaldistancefromroot,thenprintthelateroneinlevel
traversal.Forexample,inthebelowdiagram,3and4areboththebottommostnodesathorizontaldistance
0,weneedtoprint4.
20
/\
822
/\/\
53425
/\
1014
Fortheabovetreetheoutputshouldbe5,10,4,14,25.
Westronglyrecommendtominimizeyourbrowserandtrythisyourselffirst.
ThefollowingarestepstoprintBottomViewofBianryTree.
1.Weputtreenodesinaqueueforthelevelordertraversal.
2.Startwiththehorizontaldistance(hd)0oftherootnode,keeponaddingleftchildtoqueuealongwiththe
horizontaldistanceashd1andrightchildashd+1.
3.Also,useaTreeMapwhichstoreskeyvaluepairsortedonkey.
4.Everytime,weencounteranewhorizontaldistanceoranexistinghorizontaldistanceputthenodedata
forthehorizontaldistanceaskey.Forthefirsttimeitwilladdtothemap,nexttimeitwillreplacethevalue.
Thiswillmakesurethatthebottommostelementforthathorizontaldistanceispresentinthemapandifyou
seethetreefrombeneaththatyouwillseethatelement.
AJavabasedimplementationisbelow:
// Java Program to print Bottom View of Binary Tree
import java.util.*;
import java.util.Map.Entry;
// Tree node class

class TreeNode
{
int data; //data of the node
int hd; //horizontal distance of the node
TreeNode left, right; //left and right references
// Constructor of tree node

public TreeNode(int key)
{
data = key;
hd = Integer.MAX_VALUE;
}
}
//Tree class
class Tree
{
TreeNode root; //root node of tree
// Default constructor
public Tree() {}
// Parameterized tree constructor

public Tree(TreeNode node)
{
root = node;
}
// Method that prints the bottom view.

public void bottomView()
{
if (root == null)
return;
// Initialize a variable 'hd' with 0 for the root element.

int hd = 0;
// TreeMap which stores key value pair sorted on key value

Map<Integer, Integer> map = new TreeMap<>();
// Queue to store tree nodes in level order traversal

Queue<TreeNode> queue = new LinkedList<TreeNode>();
// Assign initialized horizontal distance value to root

// node and add it to the queue.
root.hd = hd;
queue.add(root);
// Loop until the queue is empty (standard level order loop)

while (!queue.isEmpty())
{
TreeNode temp = queue.remove();
// Extract the horizontal distance value from the

// dequeued tree node.
hd = temp.hd;
// Put the dequeued tree node to TreeMap having key

// as horizontal distance. Every time we find a node
// having same horizontal distance we need to replace
// the data in the map.
map.put(hd, temp.data);
// If the dequeued node has a left child add it to the

// queue with a horizontal distance hd-1.
if (temp.left != null)
{
temp.left.hd = hd-1;
queue.add(temp.left);
}
// If the dequeued node has a left child add it to the
// queue with a horizontal distance hd+1.
if (temp.right != null)
{
temp.right.hd = hd+1;
queue.add(temp.right);
}
}
// Extract the entries of map into a set to traverse

// an iterator over that.
Set<Entry<Integer, Integer>> set = map.entrySet();
// Make an iterator
Iterator<Entry<Integer, Integer>> iterator = set.iterator();
// Traverse the map elements using the iterator.

while (iterator.hasNext())
{
Map.Entry<Integer, Integer> me = iterator.next();
System.out.print(me.getValue()+" ");
}
}
}
// Main driver class

public class BottomView
{
{

root.left.left = new TreeNode(5);
root.right.left = new TreeNode(4);
root.right.right = new TreeNode(25);
root.left.right.left = new TreeNode(10);
root.left.right.right = new TreeNode(14);
Tree tree = new Tree(root);
System.out.println("Bottom view of the given binary tree:");
tree.bottomView();
}
}
Output:
Bottomviewofthegivenbinarytree:
51041425
=======================================================
BinaryIndexedTreeorFenwicktree
LetusconsiderthefollowingproblemtounderstandBinaryIndexedTree.
Wehaveanarrayarr[0...n1].Weshouldbeableto
1Findthesumoffirstielements.
2Changevalueofaspecifiedelementofthearrayarr[i]=xwhere0<=i<=n1.
Asimplesolutionistorunaloopfrom0toi1andcalculatesumofelements.Toupdateavalue,simplydo
arr[i]=x.ThefirstoperationtakesO(n)timeandsecondoperationtakesO(1)time.Anothersimplesolution
istocreateanotherarrayandstoresumfromstarttoiattheithindexinthisarray.Sumofagivenrange
cannowbecalculatedinO(1)time,butupdateoperationtakesO(n)timenow.Thisworkswellifthenumber
ofqueryoperationsarelargeandveryfewupdates.
CanweperformboththeoperationsinO(logn)timeoncegiventhearray?
OneEfficientSolutionistouseSegmentTreethatdoesbothoperationsinO(Logn)time.
UsingBinaryIndexedTree,wecandobothtasksinO(Logn)time.TheadvantagesofBinaryIndexedTree
overSegmentare,requireslessspaceandveryeasytoimplement..
Representation
BinaryIndexedTreeisrepresentedasanarray.LetthearraybeBITree[].EachnodeofBinaryIndexed
Treestoressumofsomeelementsofgivenarray.SizeofBinaryIndexedTreeisequaltonwherenissize
ofinputarray.Inthebelowcode,wehaveusedsizeasn+1foreaseofimplementation.
Construction
WeconstructtheBinaryIndexedTreebyfirstinitializingallvaluesinBITree[]as0.Thenwecallupdate()
operationforallindexestostoreactualsums,updateisdiscussedbelow.
Operations
getSum(index):Returnssumofarr[0..index]
//Returnssumofarr[0..index]usingBITree[0..n].Itassumesthat
//BITree[]isconstructedforgivenarrayarr[0..n1]
1)Initializesumas0andindexasindex+1.
2)Dofollowingwhileindexisgreaterthan0.
...a)AddBITree[index]tosum
...b)GotoparentofBITree[index].Parentcanbeobtainedbyremoving
thelastsetbitfromindex,i.e.,index=index(index&(index))
3)Returnsum.

TheabovediagramdemonstratesworkingofgetSum().Followingaresomeimportantobservations.
Nodeatindex0isadummynode.
Anodeatindexyisparentofanodeatindexx,iffycanbeobtainedbyremovinglastsetbitfrombinary
representationofx.
Achildxofanodeystoressumofelementsfromofy(exclusivey)andofx(inclusivex).
update(index,val):UpdatesBITforoperationarr[index]+=val
//Notethatarr[]isnotchangedhere.Itchanges
//onlyBITreeforthealreadymadechangeinarr[].
1)Initializeindexasindex+1.
2)Dofollowingwhileindexissmallerthanorequalton.
...a)AddvaluetoBITree[index]
...b)GotoparentofBITree[index].Parentcanbeobtainedbyremoving
thelastsetbitfromindex,i.e.,index=index(index&(index))

TheupdateprocessneedstomakesurethatallBITreenodesthathavearr[i]aspartofthesectionthey
covermustbeupdated.WegetallsuchnodesofBITreebyrepeatedlyaddingthedecimalnumber
correspondingtothelastsetbit.
HowdoesBinaryIndexedTreework?
Theideaisbasedonthefactthatallpositiveintegerscanberepresentedassumofpowersof2.For
example19canberepresentedas16+2+1.EverynodeofBITreestoressumofnelementswherenisa
powerof2.Forexample,intheabovefirstdiagramforgetSum(),sumoffirst12elementscanbeobtained
bysumoflast4elements(from9to12)plussumof8elements(from1to8).Thenumberofsetbitsin
binaryrepresentationofanumbernisO(Logn).Therefore,wetraverseatmostO(Logn)nodesinboth
getSum()andupdate()operations.TimecomplexityofconstructionisO(nLogn)asitcallsupdate()foralln
elements.
Implementation:
FollowingisC++implementationofBinaryIndexedTree.
// C++ code to demonstrate operations of Binary Index Tree
#include <iostream>
/*
n --> No. of elements present in input array.

BITree[0..n] --> Array that represents Binary Indexed Tree.
arr[0..n-1] --> Input array for whic prefix sum is evaluated. */
// Returns sum of arr[0..index]. This function assumes

// that the array is preprocessed and partial sums of
// array elements are stored in BITree[].
int getSum(int BITree[], int n, int index)
{
int sum = 0; // Iniialize result
// index in BITree[] is 1 more than the index in arr[]

index = index + 1;
// Traverse ancestors of BITree[index]

while (index>0)
{
// Add current element of BITree to sum
sum += BITree[index];
// Move index to parent node

index -= index & (-index);
}
return sum;
}
// Updates a node in Binary Index Tree (BITree) at given index

// in BITree. The given value 'val' is added to BITree[i] and
// all of its ancestors in tree.
void updateBIT(int *BITree, int n, int index, int val)
{
// index in BITree[] is 1 more than the index in arr[]

index = index + 1;
// Traverse all ancestors and add 'val'

while (index <= n)
{
// Add 'val' to current node of BI Tree
BITree[index] += val;
// Update index to that of parent

index += index & (-index);
}
}
// Constructs and returns a Binary Indexed Tree for given

// array of size n.
int *constructBITree(int arr[], int n)
{
// Create and initialize BITree[] as 0
int *BITree = new int[n+1];
for (int i=1; i<=n; i++)
BITree[i] = 0;
// Store the actual values in BITree[] using update()

for (int i=0; i<n; i++)
updateBIT(BITree, n, i, arr[i]);
// Uncomment below lines to see contents of BITree[]

//for (int i=1; i<=n; i++)
//
cout << BITree[i] << " ";
return BITree;
}
// Driver program to test above functions

int main()
{
int freq[] = {2, 1, 1, 3, 2, 3, 4, 5, 6, 7, 8, 9};
int n = sizeof(freq)/sizeof(freq[0]);
int *BITree = constructBITree(freq, n);
cout << "Sum of elements in arr[0..5] is "
<< getSum(BITree, n, 5);
// Let use test the update operation

freq[3] += 6;
updateBIT(BITree, n, 3, 6); //Update BIT for above change in arr[]
cout << "\nSum of elements in arr[0..5] after update is "

<< getSum(BITree, n, 5);
return 0;
}
Output:
Sumofelementsinarr[0..5]is12
Sumofelementsinarr[0..5]afterupdateis18
CanweextendtheBinaryIndexedTreeforrangeSuminLogntime?
Thisissimpletoanswer.TherangeSum(l,r)canbeobtainedasgetSum(r)getSum(l1).
=======================================================
HowtoImplementReverseDNS
LookUpCache?
ReverseDNSlookupisusinganinternetIPaddresstofindadomainname.Forexample,ifyoutype
74.125.200.106inbrowser,itautomaticallyredirectstogoogle.in.
HowtoimplementReverseDNSLookUpcache?Followingaretheoperationsneededfromcache.
1)AddaIPaddresstoURLMappingincache.
2)FindURLforagivenIPaddress.
OnesolutionistouseHashing.
Inthispost,aTriebasedsolutionisdiscussed.OneadvantageofTriebasedsolutionsis,worstcaseupper
boundisO(1)forTrie,forhashing,thebestpossibleaveragecasetimecomplexityisO(1).Also,withTrie
wecanimplementprefixsearch(findingallurlsforacommonprefixofIPaddresses).
ThegeneraldisadvantageofTrieislargeamountofmemoryrequirement,thisisnotamajorproblemhere
asthealphabetsizeisonly11here.Tencharactersareneededfordigitsfrom0to9andonefordot(.).
TheideaistostoreIPaddressesinTrienodesandinthelastnodewestorethecorrespondingdomain
name.FollowingisCstyleimplementationinC++.
// C based program to implement reverse DNS lookup
#include<stdio.h>
#include<stdlib.h>
#include<string.h>
// There are atmost 11 different chars in a valid IP address

#define CHARS 11
// Maximum length of a valid IP address

#define MAX 50
// A utility function to find index of child for a given character 'c'

int getIndex(char c) { return (c == '.')? 10: (c - '0'); }
// A utility function to find character for a given child index.

char getCharFromIndex(int i) { return (i== 10)? '.' : ('0' + i); }
// Trie Node.
struct trieNode
{
bool isLeaf;
char *URL;
struct trieNode *child[CHARS];
};
// Function to create a new trie node.

struct trieNode *newTrieNode(void)
{
struct trieNode *newNode = new trieNode;
newNode->isLeaf = false;
newNode->URL = NULL;
for (int i=0; i<CHARS; i++)
newNode->child[i] = NULL;
return newNode;
}
// This method inserts an ip address and the corresponding

// domain name in the trie. The last node in Trie contains the URL.
void insert(struct trieNode *root, char *ipAdd, char *URL)
{
// Length of the ip address
int len = strlen(ipAdd);
struct trieNode *pCrawl = root;
// Traversing over the length of the ip address.

for (int level=0; level<len; level++)
{
// Get index of child node from current character
// in ipAdd[]. Index must be from 0 to 10 where
// 0 to 9 is used for digits and 10 for dot
int index = getIndex(ipAdd[level]);
// Create a new child if not exist already

if (!pCrawl->child[index])
pCrawl->child[index] = newTrieNode();
// Move to the child

pCrawl = pCrawl->child[index];
}
//Below needs to be carried out for the last node.

//Save the corresponding URL of the ip address in the
//last node of trie.
pCrawl->isLeaf = true;
pCrawl->URL = new char[strlen(URL) + 1];
strcpy(pCrawl->URL, URL);
}
// This function returns URL if given IP address is present in DNS cache.

// Else returns NULL
char *searchDNSCache(struct trieNode *root, char *ipAdd)
{
// Root node of trie.
struct trieNode *pCrawl = root;
int len = strlen(ipAdd);
// Traversal over the length of ip address.

for (int level=0; level<len; level++)
{
int index = getIndex(ipAdd[level]);
if (!pCrawl->child[index])
return NULL;
pCrawl = pCrawl->child[index];
}
// If we find the last node for a given ip address, print the URL.
if (pCrawl!=NULL && pCrawl->isLeaf)
return pCrawl->URL;
return NULL;
}
//Driver function.
int main()
{
/* Change third ipAddress for validation */
char ipAdd[][MAX] = {"107.108.11.123", "107.109.123.255",
"74.125.200.106"};
char URL[][50] = {"www.samsung.com", "www.samsung.net",
"www.google.in"};
int n = sizeof(ipAdd)/sizeof(ipAdd[0]);
struct trieNode *root = newTrieNode();
// Inserts all the ip address and their corresponding
// domain name after ip address validation.

insert(root,ipAdd[i],URL[i]);
// If reverse DNS look up succeeds print the domain

// name along with DNS resolved.
char ip[] = "107.108.11.123";
char *res_url = searchDNSCache(root, ip);
if (res_url != NULL)
printf("Reverse DNS look up resolved in cache:\n%s --> %s",
ip, res_url);
else
printf("Reverse DNS look up not resolved in cache ");
return 0;
}
Output:
ReverseDNSlookupresolvedincache:
107.108.11.123>www.samsung.com
=======================================================
Givennappointments,findall
conflictingappointments
Givennappointments,findallconflictingappointments.
Examples:
Input:appointments[]={{1,5}{3,7},{2,6},{10,15},{5,6},{4,100}}
Output:Followingareconflictingintervals
[3,7]Conflictswith[1,5]
Anappointmentisconflicting,ifitconflictswithanyofthepreviousappointmentsinarray.
Westronglyrecommendtominimizethebrowserandtrythisyourselffirst.
ASimpleSolutionistoonebyoneprocessallappointmentsfromsecondappointmenttolast.Forevery
appointmenti,checkifitconflictswithi1,i2,0.ThetimecomplexityofthismethodisO(n2).
WecanuseIntervalTreetosolvethisprobleminO(nLogn)time.Followingisdetailedalgorithm.
1)CreateanIntervalTree,initiallywiththefirstappointment.
2)Dofollowingforallotherappointmentsstartingfromthesecondone.
a)Checkifthecurrentappointmentconflictswithanyoftheexisting
appointmentsinIntervalTree.Ifconflicts,thenprintthecurrent
appointment.ThisstepcanbedoneO(Logn)time.
b)InsertthecurrentappointmentinIntervalTree.Thisstepalsocan
bedoneO(Logn)time.
FollowingisC++implementationofaboveidea.
// C++ program to print all conflicting appointments in a
// given set of appointments
#include <iostream>
// Structure to represent an interval

struct Interval
{
int low, high;
};
// Structure to represent a node in Interval Search Tree

struct ITNode
{
Interval *i; // 'i' could also be a normal variable
int max;
ITNode *left, *right;
};
// A utility function to create a new Interval Search Tree Node

ITNode * newNode(Interval i)
{
ITNode *temp = new ITNode;
temp->i = new Interval(i);
temp->max = i.high;
temp->left = temp->right = NULL;
};
// A utility function to insert a new Interval Search Tree

// Node. This is similar to BST Insert. Here the low value
// of interval is used tomaintain BST property
ITNode *insert(ITNode *root, Interval i)
{
// Base case: Tree is empty, new node becomes root
if (root == NULL)
return newNode(i);
// Get low value of interval at root

int l = root->i->low;
// If root's low value is smaller, then new interval

// goes to left subtree
if (i.low < l)
root->left = insert(root->left, i);
// Else, new node goes to right subtree.

else
root->right = insert(root->right, i);
// Update the max value of this ancestor if needed

if (root->max < i.high)
root->max = i.high;
return root;
}
// A utility function to check if given two intervals overlap

bool doOVerlap(Interval i1, Interval i2)
{
if (i1.low < i2.high && i2.low < i1.high)
return true;
return false;
}
// The main function that searches a given interval i

// in a given Interval Tree.
Interval *overlapSearch(ITNode *root, Interval i)
{
// Base Case, tree is empty
if (root == NULL) return NULL;
// If given interval overlaps with root

if (doOVerlap(*(root->i), i))
return root->i;
// If left child of root is present and max of left child

// is greater than or equal to given interval, then i may
// overlap with an interval is left subtree
if (root->left != NULL && root->left->max >= i.low)
return overlapSearch(root->left, i);
// Else interval can only overlap with right subtree

return overlapSearch(root->right, i);
}
// This function prints all conflicting appointments in a given

// array of apointments.
void printConflicting(Interval appt[], int n)
{
// Create an empty Interval Search Tree, add first
// appointment
ITNode *root = NULL;
root = insert(root, appt[0]);
// Process rest of the intervals

{
// If current appointment conflicts with any of the
// existing intervals, print it
Interval *res = overlapSearch(root, appt[i]);
if (res != NULL)
cout << "[" << appt[i].low << "," << appt[i].high
<< "] Conflicts with [" << res->low << ","
<< res->high << "]\n";
// Insert this appointment

root = insert(root, appt[i]);
}
}

int main()
{
// Let us create interval tree shown in above figure
Interval appt[] = { {1, 5}, {3, 7}, {2, 6}, {10, 15},
{5, 6}, {4, 100}};
int n = sizeof(appt)/sizeof(appt[0]);
cout << "Following are conflicting intervals\n";
printConflicting(appt, n);
return 0;
}
Output:
Followingareconflictingintervals
=======================================================
PerfectBinaryTreeSpecificLevel
OrderTraversal
GivenaPerfectBinaryTreelikebelow:
(clickonimagetogetaclearview)
Printthelevelorderofnodesinfollowingspecificmanner:
12347568159141013111216311730182919282027212622252324
i.e.printnodesinlevelorderbutnodesshouldbefromleftandrightsidealternatively.Here1stand2ndlevels
aretrivial.
While3rdlevel:4(left),7(right),5(left),6(right)areprinted.
While4thlevel:8(left),15(right),9(left),14(right),..areprinted.
While5thlevel:16(left),31(right),17(left),30(right),..areprinted.
InstandardLevelOrderTraversal,weenqueuerootintoaqueue1st,thenwedequeueONEnodefrom
queue,process(print)it,enqueueitschildrenintoqueue.Wekeepdoingthisuntilqueueisempty.
Approach1:
Wecandostandardlevelordertraversalheretoobutinsteadofprintingnodesdirectly,wehavetostore
nodesincurrentlevelinatemporaryarrayorlist1standthentakenodesfromalternateends(leftandright)
andprintnodes.Keeprepeatingthisforalllevels.
Thisapproachtakesmorememorythanstandardtraversal.
Approach2:
Thestandardlevelordertraversalideawillslightlychangehere.InsteadofprocessingONEnodeatatime,
wewillprocessTWOnodesatatime.Andwhilepushingchildrenintoqueue,theenqueueorderwillbe:1st
nodesleftchild,2ndnodesrightchild,1stnodesrightchildand2ndnodesleftchild.
/* C++ program for special order traversal */
#include <iostream>
#include <queue>
/* A binary tree node has data, pointer to left child

and a pointer to right child */
struct Node
{
int data;
Node *left;
Node *right;
};

given data and NULL left and right pointers. */
Node *newNode(int data)
{
node->data = data;
node->right = node->left = NULL;
return node;
}
/* Given a perfect binary tree, print its nodes in specific

level order */
void printSpecificLevelOrder(Node *root)
{
if (root == NULL)
return;
// Let us print root and next level first

cout << root->data;
// / Since it is perfect Binary Tree, right is not checked

if (root->left != NULL)
cout << " " << root->left->data << " " << root->right->data;
// Do anything more if there are nodes at next level in

// given perfect Binary Tree
if (root->left->left == NULL)
return;
// Create a queue and enqueue left and right children of root

queue <Node *> q;
q.push(root->left);
q.push(root->right);
// We process two nodes at a time, so we need two variables

// to store two front items of queue
Node *first = NULL, *second = NULL;
// traversal loop
while (!q.empty())
{
// Pop two items from queue
first = q.front();
q.pop();
second = q.front();
q.pop();
// Print children of first and second in reverse order

cout << " " << first->left->data << " " << second->right->data;
cout << " " << first->right->data << " " << second->left->data;
// If first and second have grandchildren, enqueue them

// in reverse order
if (first->left->left != NULL)
{
q.push(first->left);
q.push(second->right);
q.push(first->right);
q.push(second->left);
}
}
}

int main()
{
//Perfect Binary Tree of Height 4
Node *root = newNode(1);
root->left
= newNode(2);
root->right
= newNode(3);
root->left->left = newNode(4);
root->left->right = newNode(5);
root->right->left = newNode(6);
root->right->right = newNode(7);
root->left->left->left = newNode(8);
root->left->left->right = newNode(9);
root->left->right->left = newNode(10);
root->left->right->right = newNode(11);
root->right->left->left = newNode(12);
root->right->left->right = newNode(13);
root->right->right->left = newNode(14);
root->right->right->right = newNode(15);
root->left->left->left->left = newNode(16);
root->left->left->left->right = newNode(17);
root->left->left->right->left = newNode(18);
root->left->left->right->right = newNode(19);
root->left->right->left->left = newNode(20);
root->left->right->left->right = newNode(21);
root->left->right->right->left = newNode(22);
root->left->right->right->right = newNode(23);
root->right->left->left->left = newNode(24);
root->right->left->left->right = newNode(25);
root->right->left->right->left = newNode(26);
root->right->left->right->right = newNode(27);
root->right->right->left->left = newNode(28);
root->right->right->left->right = newNode(29);
root->right->right->right->left = newNode(30);
root->right->right->right->right = newNode(31);
cout << "Specific Level Order traversal of binary tree is \n";

printSpecificLevelOrder(root);
return 0;
}
Output:
SpecificLevelOrdertraversalofbinarytreeis
12347568159141013111216311730182919282027212622252324
=======================================================
PrintNodesinTopViewofBinary
Tree
Topviewofabinarytreeisthesetofnodesvisiblewhenthetreeisviewedfromthetop.Givenabinary
tree,printthetopviewofit.Theoutputnodescanbeprintedinanyorder.ExpectedtimecomplexityisO(n)
Anodexisthereinoutputifxisthetopmostnodeatitshorizontaldistance.Horizontaldistanceofleftchild
ofanodexisequaltohorizontaldistanceofxminus1,andthatofrightchildishorizontaldistanceofxplus
1.
1
/\
23
/\/\
4567
Topviewoftheabovebinarytreeis
42137
1
/\
23
\
4
\
5
\
6
Topviewoftheabovebinarytreeis
2136
TheideaistodosomethingsimilartoverticalOrderTraversal.LikeverticalOrderTraversal,weneedto
nodesofsamehorizontaldistancetogether.Wedoalevelordertraversalsothatthetopmostnodeata
horizontalnodeisvisitedbeforeanyothernodeofsamehorizontaldistancebelowit.Hashingisusedto
checkifanodeatgivenhorizontaldistanceisseenornot.
// Java program to print top view of Binary tree
import java.util.*;
// Class for a tree node

class TreeNode
{
// Members
int key;
TreeNode left, right;
// Constructor
public TreeNode(int key)
{
this.key = key;
}
}
// A class to represent a queue item. The queue is used to do Level

// order traversal. Every Queue item contains node and horizontal
// distance of node from root
class QItem
{
TreeNode node;
int hd;
public QItem(TreeNode n, int h)
{
node = n;
hd = h;
}
}
// Class for a Binary Tree

class Tree
{
TreeNode root;
// Constructors
public Tree() { root = null; }
public Tree(TreeNode n) { root = n; }
// This method prints nodes in top view of binary tree

public void printTopView()
{
// base case
if (root == null) { return; }
// Creates an empty hashset

HashSet<Integer> set = new HashSet<>();
// Create a queue and add root to it

Queue<QItem> Q = new LinkedList<QItem>();
Q.add(new QItem(root, 0)); // Horizontal distance of root is 0
// Standard BFS or level order traversal loop

while (!Q.isEmpty())
{
// Remove the front item and get its details
QItem qi = Q.remove();
int hd = qi.hd;
TreeNode n = qi.node;
// If this is the first node at its horizontal distance,

// then this node is in top view
if (!set.contains(hd))
{
set.add(hd);
System.out.print(n.key + " ");
}
// Enqueue left and right children of current node

if (n.left != null)
Q.add(new QItem(n.left, hd-1));
if (n.right != null)
Q.add(new QItem(n.right, hd+1));
}
}
}
// Driver class to test above methods

public class Main
{
{
/* Create following Binary Tree
1
/ \
2 3
\
4
\
5
\
6*/
root.left.right.right = new TreeNode(5);
root.left.right.right.right = new TreeNode(6);
Tree t = new Tree(root);
System.out.println("Following are nodes in top view of Binary Tree");

t.printTopView();
}
}
Output:
FollowingarenodesintopviewofBinaryTree
1236
=======================================================
KDimensionalTree
AKDTree(alsocalledasKDimensionalTree)isabinarysearchtreewheredataineachnodeisa
KDimensionalpointinspace.Inshort,itisaspacepartitioning(detailsbelow)datastructurefororganizing
pointsinaKDimensionalspace.
AnonleafnodeinKDtreedividesthespaceintotwoparts,calledashalfspaces.
Pointstotheleftofthisspacearerepresentedbytheleftsubtreeofthatnodeandpointstotherightofthe
spacearerepresentedbytherightsubtree.Wewillsoonbeexplainingtheconceptonhowthespaceis
dividedandtreeisformed.
Forthesakeofsimplicity,letusunderstanda2DTreewithanexample.
Therootwouldhaveanxalignedplane,therootschildrenwouldbothhaveyalignedplanes,theroots
grandchildrenwouldallhavexalignedplanes,andtherootsgreatgrandchildrenwouldallhaveyaligned
planesandsoon.
Generalization:
Letusnumbertheplanesas0,1,2,(K1).Fromtheaboveexample,itisquiteclearthatapoint(node)
atdepthDwillhaveAalignedplanewhereAiscalculatedas:
A=DmodK
Howtodetermineifapointwilllieintheleftsubtreeorinrightsubtree?
IftherootnodeisalignedinplaneA,thentheleftsubtreewillcontainallpointswhosecoordinatesinthat
planearesmallerthanthatofrootnode.Similarly,therightsubtreewillcontainallpointswhosecoordinates
inthatplanearegreaterequaltothatofrootnode.
Creationofa2DTree:
Considerfollowingpointsina2Dplane:
(3,6),(17,15),(13,15),(6,12),(9,1),(2,7),(10,19)
1.
Insert(3,6):Sincetreeisempty,makeittherootnode.
2.
Insert(17,15):Compareitwithrootnodepoint.SincerootnodeisXaligned,theXcoordinatevalue
willbecomparedtodetermineifitliesintherightsubtreeorintherightsubtree.Thispointwillbe
Yaligned.
3.
Insert(13,15):XvalueofthispointisgreaterthanXvalueofpointinrootnode.So,thiswillliein
therightsubtreeof(3,6).AgainCompareYvalueofthispointwiththeYvalueofpoint(17,15)
(Why?).Since,theyareequal,thispointwilllieintherightsubtreeof(17,15).Thispointwillbe
Xaligned.
4.
Insert(6,12):XvalueofthispointisgreaterthanXvalueofpointinrootnode.So,thiswilllieinthe
rightsubtreeof(3,6).AgainCompareYvalueofthispointwiththeYvalueofpoint(17,15)(Why?).
Since,12<15,thispointwilllieintheleftsubtreeof(17,15).ThispointwillbeXaligned.
5.
Insert(9,1):Similarly,thispointwilllieintherightof(6,12).
6.
Insert(2,7):Similarly,thispointwilllieintheleftof(3,6).
7.
Insert(10,19):Similarly,thispointwilllieintheleftof(13,15).
Howisspacepartitioned?
All7pointswillbeplottedintheXYplaneasfollows:
1.
Point(3,6)willdividethespaceintotwoparts:DrawlineX=3.
2.
3.
Point(2,7)willdividethespacetotheleftoflineX=3intotwopartshorizontally.
4.
DrawlineY=7totheleftoflineX=3.
5.
6.
Point(17,15)willdividethespacetotherightoflineX=3intotwopartshorizontally.
7.
DrawlineY=15totherightoflineX=3.
8.
Point(6,12)willdividethespacebelowlineY=15andtotherightoflineX=3intotwoparts.
DrawlineX=6totherightoflineX=3andbelowlineY=15.
Point(13,15)willdividethespacebelowlineY=15andtotherightoflineX=6intotwoparts.
DrawlineX=13totherightoflineX=6andbelowlineY=15.
Point(9,1)willdividethespacebetweenlinesX=3,X=6andY=15intotwoparts.
DrawlineY=1betweenlinesX=3andX=6.
Point(10,19)willdividethespacetotherightoflineX=3andabovelineY=15intotwoparts.
DrawlineY=19totherightoflineX=3andabovelineY=15.
FollowingisC++implementationofKDTreebasicoperationslikesearch,insertanddelete.
// A C++ program to demonstrate operations of KD tree
#include <iostream>
#include <cstdio>
#include <cassert>
#include <cstdlib>
// A structure to represent a point in K dimensional space

// k: K dimensional space
// coord: An array to represent position of point
struct Point
{
unsigned k;
int* coord; // Coordinate (A pointer to array of size k)
};
// A structure to represent the Input

// n: Number of points in space
// pointArray: An array to keep information of each point
struct Input
{
// n --> NUMBER OF POINTS

unsigned n;
Point* *pointArray;
};
// A structure to represent node of 2 dimensional tree

struct Node
{
Point point;
Node *left, *right;
};
// Creates and return a Point structure

Point* CreatePoint(unsigned k)
{
Point* point = new Point;
// Memory allocation failure

assert(NULL != point);
point->k = k;
point->coord = new int[k];

assert(NULL != point->coord);
return point;
}
// Creates and returns an Input structure

struct Input* CreateInput(unsigned k, unsigned n)
{
struct Input* input = new Input;

assert(NULL != input);
input->n = n;
input->pointArray = new Point*[n];

assert(NULL != input->pointArray);
return input;
}
// A method to create a node of K D tree

struct Node* CreateNode(struct Point* point)
{
struct Node* tempNode = new Node;

assert(NULL != tempNode);
// Avoid shallow copy [We could have directly use

// the below assignment, But didn't, why?]
/*tempNode->point = point;*/
(tempNode->point).k = point->k;
(tempNode->point).coord = new int[point->k];
// Copy coordinate values

for (int i=0; i<(tempNode->point).k; ++i)
(tempNode->point).coord[i] = point->coord[i];
tempNode->left = tempNode->right = NULL;

return tempNode;
}
// Root is passed as pointer to pointer so that

// The parameter depth is used to decide axis of comparison
void InsertKDTreeUtil(Node * * root, Node* newNode, unsigned depth)
{
// Tree is empty?
if (!*root)
{
*root = newNode;
return;
}
// Calculate axis of comparison to determine left/right

unsigned axisOfComparison = depth % (newNode->point).k;
// Compare the new point with root and decide the left or
// right subtree
if ((newNode->point).coord[axisOfComparison] <
((*root)->point).coord[axisOfComparison])
InsertKDTreeUtil(&((*root)->left), newNode, depth + 1);
else
InsertKDTreeUtil(&((*root)->right), newNode, depth + 1);
}
// Function to insert a new point in KD Tree. It mainly uses

// above recursive function "InsertKDTreeUtil()"
void InsertKDTree(Node* *root, Point* point)
{
Node* newNode = CreateNode(point);
unsigned zeroDepth = 0;
InsertKDTreeUtil(root, newNode, zeroDepth);
}
// A utility method to determine if two Points are same

// in K Dimensional space
int ArePointsSame(Point firstPoint, Point secondPoint)
{
if (firstPoint.k != secondPoint.k)
return 0;
// Compare individual coordinate values

for (int i = 0; i < firstPoint.k; ++i)
if (firstPoint.coord[i] != secondPoint.coord[i])
return 0;
return 1;
}
// Searches a Point in the K D tree. The parameter depth is used

// to determine current axis.
int SearchKDTreeUtil(Node* root, Point point, unsigned depth)
{
if (!root)
return 0;
if (ArePointsSame(root->point, point))
return 1;
unsigned axisOfComparison = depth % point.k;
if (point.coord[axisOfComparison] <
(root->point).coord[axisOfComparison])
return SearchKDTreeUtil(root->left, point, depth + 1);
return SearchKDTreeUtil(root->right, point, depth + 1);

}
// Searches a Point in the K D tree. It mainly uses

// SearchKDTreeUtil()
int SearchKDTree(Node* root, Point point)
{
unsigned zeroDepth = 0;
return SearchKDTreeUtil(root, point, zeroDepth);
}
// Creates a KD tree from given input points. It mainly

// uses InsertKDTree
Node* CreateKDTree(Input* input)
{
Node* root = NULL;
for (int i = 0; i < input->n; ++i)
InsertKDTree(&root, input->pointArray[i]);
return root;
}
// A utility function to print an array

void PrintArray(int* array, unsigned size)
{
for (unsigned i = 0; i < size; ++i)
cout << array[i];
cout << endl;
}
// A utility function to do inorder tree traversal

void inorderKD(Node* root)
{
if (root)
{
inorderKD(root->left);
PrintArray((root->point).coord, (root->point).k);
inorderKD(root->right);
}
}

int main()
{
// 2 Dimensional tree [For the sake of simplicity]
unsigned k = 2;
// Total number of Points is 7

unsigned n = 7;
Input* input = CreateInput(k, n);
// itc --> ITERATOR for coord

// itp --> ITERATOR for POINTS
for (int itp = 0; itp < n; ++itp)
{
input->pointArray[itp] = CreatePoint(k);
for (int itc = 0; itc < k; ++itc)

input->pointArray[itp]->coord[itc] = rand() % 20;
PrintArray(input->pointArray[itp]->coord, k);
}
Node* root = CreateKDTree(input);
cout << "Inorder traversal of K-D Tree created is:\n";

inorderKD(root);
return 0;
}
Output:
17
140
94
1818
24
55
17
InordertraversalofKDTreecreatedis:
17
140
24
17
55
94
1818
=======================================================
ConvertaBinaryTreetoThreaded
binarytree
WehavediscussedThreadedBinaryTree.Theideaofthreadedbinarytreesistomakeinordertraversal
fasteranddoitwithoutstackandwithoutrecursion.Inasimplethreadedbinarytree,theNULLrightpointers
areusedtostoreinordersuccessor.WhereeverarightpointerisNULL,itisusedtostoreinorder
successor.
FollowingdiagramshowsanexampleSingleThreadedBinaryTree.Thedottedlinesrepresentthreads.
Followingisstructureofsinglethreadedbinarytree.
struct Node
{
int key;
Node *left, *right;
// Used to indicate whether the right pointer is a normal right

// pointer or a pointer to inorder successor.
bool isThreaded;
};
HowtoconvertaGivenBinaryTreetoThreadedBinaryTree?
WebasicallyneedtosetNULLrightpointerstoinordersuccessor.Wefirstdoaninordertraversalofthe
treeandstoreitinaqueue(wecanuseasimplearrayalso)sothattheinordersuccessorbecomesthenext
node.WeagaindoaninordertraversalandwheneverwefindanodewhoserightisNULL,wetakethefront
itemfromqueuueandmakeittherightofcurrentnode.WealsosetisThreadedtotruetoindicatethatthe
rightpointerisathreadedlink.
FollowingisC++implementationoftheaboveidea.
/* C++ program to convert a Binary Tree to Threaded Tree */
#include <iostream>
#include <queue>
/* Structure of a node in threaded binary tree */

struct Node
{
int key;
Node *left, *right;
// Used to indicate whether the right pointer is a normal

// right pointer or a pointer to inorder successor.
bool isThreaded;
};
// Helper function to put the Nodes in inorder into queue

void populateQueue(Node *root, std::queue <Node *> *q)
{
if (root == NULL) return;
if (root->left)
populateQueue(root->left, q);
q->push(root);
if (root->right)
populateQueue(root->right, q);
}
// Function to traverse queue, and make tree threaded

void createThreadedUtil(Node *root, std::queue <Node *> *q)
{
if (root->left)
createThreadedUtil(root->left, q);
q->pop();
if (root->right)
createThreadedUtil(root->right, q);
// If right pointer is NULL, link it to the

// inorder successor and set 'isThreaded' bit.
else
{
root->right = q->front();
root->isThreaded = true;
}
}
// This function uses populateQueue() and

// createThreadedUtil() to convert a given binary tree
// to threaded tree.
void createThreaded(Node *root)
{
// Create a queue to store inorder traversal
std::queue <Node *> q;
// Store inorder traversal in queue

populateQueue(root, &q);
// Link NULL right pointers to inorder successor

createThreadedUtil(root, &q);
}
// A utility function to find leftmost node in a binary

// tree rooted with 'root'. This function is used in inOrder()
Node *leftMost(Node *root)
{
while (root != NULL && root->left != NULL)
root = root->left;
return root;
}
// Function to do inorder traversal of a threadded binary tree
void inOrder(Node *root)

{
// Find the leftmost node in Binary Tree

Node *cur = leftMost(root);
while (cur != NULL)

{
cout << cur->key << " ";
// If this Node is a thread Node, then go to

// inorder successor
if (cur->isThreaded)
cur = cur->right;
else // Else go to the leftmost child in right subtree

cur = leftMost(cur->right);
}
}
// A utility function to create a new node

Node *newNode(int key)
{
Node *temp = new Node;
temp->key = key;
return temp;
}

int main()
{
/*
1
/\
2
/\/\
4 56 7
*/

root->left = newNode(2);
root->right = newNode(3);
createThreaded(root);
cout << "Inorder traversal of creeated threaded tree is\n";

inOrder(root);
return 0;
}
Output:
Inordertraversalofcreeatedthreadedtreeis
4251637
==================================================================================
SerializeandDeserializeanNary
Tree
GivenanNarytreewhereeverynodehasatmostNchildren.Howtoserializeanddeserialzeit?
Serializationistostoretreeinafilesothatitcanbelaterrestored.Thestructureoftreemustbemaintained.
Deserializationisreadingtreebackfromfile.
Thispostismainlyanextensionofbelowpost.
SerializeandDeserializeaBinaryTree
InanNarytree,therearenodesignatedleftandrightchildren.AnNarytreeisrepresentedbystoringan
arrayorlistofchildpointerswitheverynode.
Theideaistostoreanendofchildrenmarkerwitheverynode.Thefollowingdiagramshowsserialization
where)isusedasendofchildrenmarker.Thediagramistakenfromhere.

// A C++ Program serialize and deserialize an N-ary tree
#include<cstdio>
#define MARKER ')'
#define N 5
// A node of N-ary tree

struct Node {
char key;
Node *child[N]; // An array of pointers for N children
};
// A utility function to create a new N-ary tree node

Node *newNode(char key)
{
temp->key = key;
for (int i = 0; i < N; i++)
temp->child[i] = NULL;
return temp;
}
// This function stores the given N-ary tree in a file pointed by fp

void serialize(Node *root, FILE *fp)
{
// Base case
// Else, store current node and recur for its children

fprintf(fp, "%c ", root->key);
for (int i = 0; i < N && root->child[i]; i++)
serialize(root->child[i], fp);
// Store marker at the end of children

fprintf(fp, "%c ", MARKER);
}
// This function constructs N-ary tree from a file pointed by 'fp'.

// This functionr returns 0 to indicate that the next item is a valid
// tree key. Else returns 0
int deSerialize(Node *&root, FILE *fp)
{
// Read next item from file. If theere are no more items or next
// item is marker, then return 1 to indicate same
char val;
if ( !fscanf(fp, "%c ", &val) || val == MARKER )
return 1;
// Else create node with this item and recur for children
root = newNode(val);
for (int i = 0; i < N; i++)
if (deSerialize(root->child[i], fp))
break;
// Finally return 0 for successful finish

return 0;
}
// A utility function to create a dummy tree shown in above diagram

Node *createDummyTree()
{
Node *root = newNode('A');

root->child[0] = newNode('B');
root->child[1] = newNode('C');
root->child[2] = newNode('D');
root->child[0]->child[0] = newNode('E');
root->child[0]->child[1] = newNode('F');
root->child[2]->child[0] = newNode('G');
root->child[2]->child[1] = newNode('H');
root->child[2]->child[2] = newNode('I');
root->child[2]->child[3] = newNode('J');
root->child[0]->child[1]->child[0] = newNode('K');
return root;
}
// A utlity function to traverse the constructed N-ary tree

void traverse(Node *root)
{
if (root)
{
printf("%c ", root->key);
for (int i = 0; i < N; i++)
traverse(root->child[i]);
}
}

int main()
{
// Let us create an N-ary tree shown in above diagram
Node *root = createDummyTree();
// Let us open a file and serialize the tree into the file
FILE *fp = fopen("tree.txt", "w");
if (fp == NULL)
{
puts("Could not open file");
return 0;
}
serialize(root, fp);
fclose(fp);
// Let us deserialize the storeed tree into root1

Node *root1 = NULL;
fp = fopen("tree.txt", "r");
deSerialize(root1, fp);
printf("Constructed N-Ary Tree from file is \n");

traverse(root1);
return 0;
}
Output:
ConstructedNAryTreefromfileis
ABEFKCDGHIJ
==================================================================================
SerializeandDeserializeaBinary
Tree
Serializationistostoretreeinafilesothatitcanbelaterrestored.Thestructureoftreemustbemaintained.
Deserializationisreadingtreebackfromfile.
Followingaresomesimplerversionsoftheproblem:
IfgivenTreeisBinarySearchTree?
IfthegivenBinaryTreeisBinarySearchTree,wecanstoreitbyeitherstoringpreorderorpostorder
traversal.IncaseofBinarySearchTrees,onlypreorderorpostordertraversalissufficienttostorestructure
information.
IfgivenBinaryTreeisCompleteTree?
ABinaryTreeiscompleteifalllevelsarecompletelyfilledexceptpossiblythelastlevelandallnodesoflast
levelareasleftaspossible(BinaryHeapsarecompleteBinaryTree).ForacompleteBinaryTree,level
ordertraversalissufficienttostorethetree.Weknowthatthefirstnodeisroot,nexttwonodesarenodesof
nextlevel,nextfournodesarenodesof2ndlevelandsoon.
IfgivenBinaryTreeisFullTree?
AfullBinaryisaBinaryTreewhereeverynodehaseither0or2children.Itiseasytoserializesuchtreesas
everyinternalnodehas2children.Wecansimplystorepreordertraversalandstoreabitwitheverynodeto
indicatewhetherthenodeisaninternalnodeoraleafnode.
HowtostoreageneralBinaryTree?
AsimplesolutionistostorebothInorderandPreordertraversals.Thissolutionrequiresrequiresspace
twicethesizeofBinaryTree.
WecansavespacebystoringPreordertraversalandamarkerforNULLpointers.
LetthemarkerforNULLpointersbe'1'
Input:
12
/
13
Output:121311
Input:
20
/\
822
Output:208112211
Input:
20
/
8
/\
412
/\
1014
Output:20841112101114111
Input:
20
/
8
/
10
/
5
Output:20810511111
Input:
20
\
8
\
10
\
5
Output:20181101511
Deserializationcanbedonebysimplyreadingdatafromfileonebyone.
FollowingisC++implementationoftheaboveidea.
// A C++ program to demonstrate serialization and deserialiation of
// Binary Tree
#include <stdio.h>
#define MARKER -1
/* A binary tree Node has key, pointer to left and right children */
struct Node
{
int key;
};
/* Helper function that allocates a new Node with the

given key and NULL left and right pointers. */
Node* newNode(int key)
{
Node* temp = new Node;
temp->key = key;
return (temp);
}
// This function stores a tree in a file pointed by fp

void serialize(Node *root, FILE *fp)
{
// If current node is NULL, store marker
if (root == NULL)
{
fprintf(fp, "%d ", MARKER);
return;
}
// Else, store current node and recur for its children

fprintf(fp, "%d ", root->key);
serialize(root->left, fp);
serialize(root->right, fp);
}
// This function constructs a tree from a file pointed by 'fp'

void deSerialize(Node *&root, FILE *fp)
{
// Read next item from file. If theere are no more items or next
// item is marker, then return
int val;
if ( !fscanf(fp, "%d ", &val) || val == MARKER)
return;
// Else create node with this item and recur for children
root = newNode(val);
deSerialize(root->left, fp);
deSerialize(root->right, fp);
}
// A simple inorder traversal used for testing the constructed tree

void inorder(Node *root)
{
if (root)
{
inorder(root->left);
printf("%d ", root->key);

inorder(root->right);
}
}

int main()
{
// Let us construct a tree shown in the above figure
struct Node *root
= newNode(20);
root->left
= newNode(8);
root->right
= newNode(22);
root->left->left
= newNode(4);
root->left->right
= newNode(12);
// Let us open a file and serialize the tree into the file
FILE *fp = fopen("tree.txt", "w");
if (fp == NULL)
{
puts("Could not open file");
return 0;
}
serialize(root, fp);
fclose(fp);
// Let us deserialize the storeed tree into root1

Node *root1 = NULL;
fp = fopen("tree.txt", "r");
deSerialize(root1, fp);
printf("Inorder Traversal of the tree constructed from file:\n");

inorder(root1);
return 0;
}
Output:
InorderTraversalofthetreeconstructedfromfile:
481012142022
Howmuchextraspaceisrequiredinabovesolution?
Iftherearenkeys,thentheabovesolutionrequiresn+1markerswhichmaybebetterthansimplesolution
(storingkeystwice)insituationswherekeysarebigorkeyshavebigdataitemsassociatedwiththem.
Canweoptimizeitfurther?
Theabovesolutioncanbeoptimizedinmanyways.Ifwetakeacloserlookataboveserializedtrees,we
canobserverthatallleafnodesrequiretwomarkers.Onesimpleoptimizationistostoreaseparatebitwith
everynodetoindicatethatthenodeisinternalorexternal.Thiswaywedonthavetostoretwomarkerswith
everyleafnodeasleavescanbeidentifiedbyextrabit.Westillneedmarkerforinternalnodeswithone
child.Forexampleinthefollowingdiagramisusedtoindicateaninternalnodesetbit,and/isusedas
NULLmarker.Thediagramistakenfromhere.
PleasenotethattherearealwaysmoreleafnodesthaninternalnodesinaBinaryTree(Numberofleaf
nodesisnumberofinternalnodesplus1,sothisoptimizationmakessense.
Howtoserializenarytree?
Inannarytree,thereisnodesignatedleftorrightchild.Wecanstoreanendofchildrenmarkerwithevery
node.Thefollowingdiagramshowsserializationwhere)isusedasendofchildrenmarker.Wewillsoonbe
coveringimplementationfornarytree.Thediagramistakenfromhere.

==================================================================================
Printnodesbetweentwogivenlevel
numbersofabinarytree
Givenabinarytreeandtwolevelnumberslowandhigh,printnodesfromlevellowtolevelhigh.
Forexampleconsiderthebinarytreegiveninbelowdiagram.
Input:Rootofbelowtree,low=2,high=4
Output:
822
412
1014<>

ASimpleMethodistofirstwritearecursivefunctionthatprintsnodesofagivenlevelnumber.Thencall
recursivefunctioninaloopfromlowtohigh.TimecomplexityofthismethodisO(n2)
WecanprintnodesinO(n)timeusingqueuebasediterativelevelordertraversal.Theideaistodosimple
queuebasedlevelordertraversal.Whiledoinginordertraversal,addamarkernodeattheend.Whenever
weseeamarkernode,weincreaselevelnumber.Iflevelnumberisbetweenlowandhigh,thenprintnodes.
ThefollowingisC++implementationofaboveidea.
// A C++ program to print Nodes level by level berween given two levels.
#include <iostream>
#include <queue>
/* A binary tree Node has key, pointer to left and right children */
struct Node
{
int key;
};
/* Given a binary tree, print nodes from level number 'low' to level
number 'high'*/
void printLevels(Node* root, int low, int high)
{
queue <Node *> Q;
Node *marker = new Node; // Marker node to indicate end of level
int level = 1;
// Initialize level number
// Enqueue the only first level node and marker node for end of level
Q.push(root);
Q.push(marker);
// Simple level order traversal loop

while (Q.empty() == false)
{
// Remove the front item from queue
Node *n = Q.front();
Q.pop();
// Check if end of level is reached

if (n == marker)
{
// print a new line and increment level number
cout << endl;
level++;
// Check if marker node was last node in queue or

// level number is beyond the given upper limit
if (Q.empty() == true || level > high) break;
// Enqueue the marker for end of next level

Q.push(marker);
// If this is marker, then we don't need print it

// and enqueue its children
continue;
}
// If level is equal to or greater than given lower level,

// print it
if (level >= low)
cout << n->key << " ";
// Enqueue children of non-marker node

if (n->left != NULL) Q.push(n->left);
if (n->right != NULL) Q.push(n->right);
}
}

given key and NULL left and right pointers. */
{
temp->key = key;
return (temp);
}

int main()
{
struct Node *root
= newNode(20);
root->left
= newNode(8);
root->right
= newNode(22);
root->left->left
= newNode(4);
root->left->right
= newNode(12);
cout << "Level Order traversal between given two levels is";
printLevels(root, 2, 3);
return 0;
}
LevelOrdertraversalbetweengiventwolevelsis
822
412
==================================================================================
FindHeightofBinaryTree
representedbyParentarray
Agivenarrayrepresentsatreeinsuchawaythatthearrayvaluegivestheparentnodeofthatparticular
index.Thevalueoftherootnodeindexwouldalwaysbe1.Findtheheightofthetree.
HeightofaBinaryTreeisnumberofnodesonthepathfromroottothedeepestleafnode,thenumber
includesbothrootandleaf.
Input:parent[]={1552213}
Output:4
ThegivenarrayrepresentsfollowingBinaryTree
5
/\
12
//\
034
/
6
Input:parent[]={1,0,0,1,1,3,5}
Output:5
ThegivenarrayrepresentsfollowingBinaryTree
0
/\
12
/\
34
/
5
/
Source:AmazonInterviewexperience|Set128(ForSDET)
Asimplesolutionistofirstconstructthetreeandthenfindheightoftheconstructedbinarytree.Thetree
canbeconstructedrecursivelybyfirstsearchingthecurrentroot,thenrecurringforthefoundindexesand
makingthemleftandrightsubtreesofroot.ThissolutiontakesO(n2)aswehavetolinearlysearchforevery
node.
AnefficientsolutioncansolvetheaboveprobleminO(n)time.Theideaistofirstcalculatedepthofevery
nodeandstoreinanarraydepth[].Oncewehavedepthsofallnodes,wereturnmaximumofalldepths.
1)Finddepthofallnodesandfillinanauxiliaryarraydepth[].
2)Returnmaximumvalueindepth[].
Followingarestepstofinddepthofanodeatindexi.
1)Ifitisroot,depth[i]is1.
2)Ifdepthofparent[i]isevaluated,depth[i]isdepth[parent[i]]+1.
3)Ifdepthofparent[i]isnotevaluated,recurforparentandassigndepth[i]asdepth[parent[i]]+1(sameas
above).
// C++ program to find height using parent array
#include <iostream>
// This function fills depth of i'th element in parent[]. The depth is

// filled in depth[i].
void fillDepth(int parent[], int i, int depth[])
{
// If depth[i] is already filled
if (depth[i])
return;
// If node at index i is root

if (parent[i] == -1)
{
depth[i] = 1;
return;
}
// If depth of parent is not evaluated before, then evaluate

// depth of parent first
if (depth[parent[i]] == 0)
fillDepth(parent, parent[i], depth);
// Depth of this node is depth of parent plus 1

depth[i] = depth[parent[i]] + 1;
}
// This function returns height of binary tree represented by

// parent array
int findHeight(int parent[], int n)
{
// Create an array to store depth of all nodes/ and
// initialize depth of every node as 0 (an invalid
// value). Depth of root is 1
int depth[n];
for (int i = 0; i < n; i++)
depth[i] = 0;
// fill depth of all nodes

for (int i = 0; i < n; i++)
fillDepth(parent, i, depth);
// The height of binary tree is maximum of all depths.

// Find the maximum value in depth[] and assign it to ht.
int ht = depth[0];
if (ht < depth[i])
ht = depth[i];
return ht;
}

int main()
{
// int parent[] = {1, 5, 5, 2, 2, -1, 3};
int parent[] = {-1, 0, 0, 1, 1, 3, 5};
int n = sizeof(parent)/sizeof(parent[0]);
cout << "Height is " << findHeight(parent, n);
return 0;
}
Output:
Heightis5
==================================================================================
Minimumno.ofiterationstopass
informationtoallnodesinthetree
Givenaverylargenarytree.Wheretherootnodehassomeinformationwhichitwantstopasstoallofits
childrendowntotheleaveswiththeconstraintthatitcanonlypasstheinformationtooneofitschildrenata
time(takeitasoneiteration).
Nowinthenextiterationthechildnodecantransferthatinformationtoonlyoneofitschildrenandatthe
sametimeinstancethechildsparenti.e.rootcanpasstheinfotooneofitsremainingchildren.Continuing
inthiswaywehavetofindtheminimumnoofiterationsrequiredtopasstheinformationtoallnodesinthe
tree.
Minimumnoofiterationsfortreebelowis6.TherootAfirstpassesinformationtoB.Innextiteration,A
passesinformationtoEandBpassesinformationtoHandsoon.

ThiscanbedoneusingPostOrderTraversal.Theideaistoconsiderheightandchildrencountoneachand
everynode.
Ifachildnodeitakesciiterationstopassinfobelowitssubtree,thenitsparentwilltake(ci+1)iterationsto
passinfotosubtreerootedatthatchildi.
Ifparenthasmorechildren,itwillpassinfototheminsubsequentiterations.Letssaychildrenofaparent
takesc1,c2,c3,c4,,cniterationstopassinfointheirownsubtree,Nowparenthastopassinfotothese
nchildrenonebyoneinniterations.Ifparentpickschildiinithiteration,thenparentwilltake(i+ci)
iterationstopassinfotochildiandallitssubtree.
Inanyiteration,whenparentpassesinfoachildi+1,children(1toi)whichgotinfofromparentalreadyin
previousiterations,willpassinfotofurtherdowninsubsequentiterations,ifanychild(1toi)hasitsown
childfurtherdown.
Topassinfotowholetreeinminimumiterations,itneedstobemadesurethatbandwidthisutilizedas
efficientlyaspossible(i.e.maximumpassablenoofnodesshouldpassinfofurtherdowninanyiteration)
Thebestpossiblescenariowouldbethatinnthiteration,ndifferentnodespassinfototheirchild.
Nodeswithheight=0:(Trivialcase)Leafnodehasnochildren(noinformationpassingneeded),sonoof
iterationswouldbeZERO.
Nodeswithheight=1:Herenodehastopassinfotoallthechildrenonebyone(allchildrenareleafnode,
sonomoreinformationpassingfurtherdown).Sinceallchildrenareleaf,nodecanpassinfotoanychildin
anyorder(pickanychildwhodidntreceivetheinfoyet).Oneiterationneededforeachchildandsonoof
iterationswouldbenoofchildren.Sonodewithheight1withnchildrenwilltakeniterations.
TakeacounterinitializedwithZERO,loopthroughallchildrenandkeepincrementingcounter.
Nodeswithheight>1:Letsassumethattherearenchildren(1ton)ofanodeandminimumnoiterations
forallnchildrenarec1,c2,.,cn.
Tomakesuremaximumnoofnodesparticipateininfopassinginanyiteration,parentshould1stpassinfo
tothatchildwhowilltakemaximumiterationtopassinfofurtherdowninsubsequentiterations.i.e.inany
iteration,parentshouldchoosethechildwhotakesmaximumiterationlateron.Itcanbethoughtofasa
greedyapproachwhereparentchoosethatchild1st,whoneedsmaximumnoofiterationssothatall
subsequentiterationscanbeutilizedefficiently.
Ifparentgoesinanyotherfashion,thenintheend,therecouldbesomenodeswhicharedonequiteearly,
sittingidleandsobandwidthisnotutilizedefficientlyinfurtheriterations.
Iftherearetwochildreniandjwithminimumiterationsciandcjwhereci>cj,thenIfparentpickschildj1st
thennoofiterationsneededbyparenttopassinfotobothchildrenandtheirsubtreewouldbe:max(1+cj,2
+ci)=2+ci
Ifparentpickschildi1stthennoofiterationsneededbyparenttopassinfotobothchildrenandtheir
subtreewouldbe:max(1+ci,2+cj)=1+ci(Sopickingcigivesbetterresultthanpickingcj)
Thistellsthatparentshouldalwayschoosechildiwithmaxcivalueinanyiteration.
SOheregreedyapproachis:
sortallcivaluesdecreasingorder,
letssayaftersorting,valuesarec1>c2>c3>.>cn
takeacounterc,setc=1+c1(forchildwithmaximumnoofiterations)
forallchildrenifrom2ton,c=c+1+ci
thentotalnoofiterationsneededbyparentismax(n,c)
LetminItr(A)betheminimumiterationneededtopassinfofromnodeAtoitsallthesubtree.Letchild(A)
bethecountofallchildrenfornodeA.Sorecursiverelationwouldbe:
1.GetminItr(B)ofallchildren(B)ofanode(A)
2.SortallminItr(B)indescendingorder
3.GetminItrofAbasedonallminItr(B)
minItr(A)=child(A)
ForchildrenBfromi=0tochild(A)
minItr(A)=max(minItr(A),minItr(B)+i+1)
Basecaseswouldbe:
Ifnodeisleaf,minItr=0
Ifnode'sheightis1,minItr=childrencount
// C++ program to find minimum number of iterations to pass
// information from root to all nodes in an n-ary tree
#include<iostream>
#include<list>
#include<cmath>
#include <stdlib.h>
// A class to represent n-ary tree (Note that the implementation

// is similar to graph for simplicity of implementation
class NAryTree
{
int N;
// No. of nodes in Tree
// Pointer to an array containing list of children

list<int> *adj;
// A function used by getMinIter(), it basically does postorder

void getMinIterUtil(int v, int minItr[]);
public:
NAryTree(int N);
// Constructor
// function to add a child w to v

void addChild(int v, int w);
// The main function to find minimum iterations

int getMinIter();
static int compare(const void * a, const void * b);

};
NAryTree::NAryTree(int N)
{
this->N = N;
adj = new list<int>[N];
}
// To add a child w to v
void NAryTree::addChild(int v, int w)
{
adj[v].push_back(w); // Add w to vs list.

}
/* A recursive function to used by getMinIter(). This function

// mainly does postorder traversal and get minimum iteration of all children
// of node u, sort them in decreasing order and then get minimum iteration
// of node u
1. Get minItr(B) of all children (B) of a node (A)

2. Sort all minItr(B) in descending order
3. Get minItr of A based on all minItr(B)
minItr(A) = child(A) -->> child(A) is children count of node A
For children B from i = 0 to child(A)
minItr(A) = max ( minItr(A), minItr(B) + i + 1)
Base cases would be:

If node is leaf, minItr = 0
If node's height is 1, minItr = children count
*/
void NAryTree::getMinIterUtil(int u, int minItr[])

{
minItr[u] = adj[u].size();
int *minItrTemp = new int[minItr[u]];
int k = 0, tmp = 0;
// Recur for all the vertices adjacent to this vertex
list<int>::iterator i;
for (i = adj[u].begin(); i!= adj[u].end(); ++i)
{
getMinIterUtil(*i, minItr);
minItrTemp[k++] = minItr[*i];
}
qsort(minItrTemp, minItr[u], sizeof (int), compare);
for (k = 0; k < adj[u].size(); k++)
{
tmp = minItrTemp[k] + k + 1;
minItr[u] = max(minItr[u], tmp);
}
delete[] minItrTemp;
}
// The function to do PostOrder traversal. It uses

// recursive getMinIterUtil()
int NAryTree::getMinIter()
{
// Set minimum iteration all the vertices as zero
int *minItr = new int[N];
int res = -1;
for (int i = 0; i < N; i++)
minItr[i] = 0;
// Start Post Order Traversal from Root

getMinIterUtil(0, minItr);
res = minItr[0];
delete[] minItr;
return res;
}
int NAryTree::compare(const void * a, const void * b)

{
return ( *(int*)b - *(int*)a );
}
// Driver function to test above functions

int main()
{
// TestCase 1
NAryTree tree1(17);
tree1.addChild(0, 1);
cout << "TestCase 1 - Minimum Iteration: "

<< tree1.getMinIter() << endl;
// TestCase 2
NAryTree tree2(3);
// TestCase 3
NAryTree tree3(1);
// TestCase 4
NAryTree tree4(6);

// TestCase 5
NAryTree tree5(6);
// TestCase 6
NAryTree tree6(6);
// TestCase 7
NAryTree tree7(14);
// TestCase 8
NAryTree tree8(14);
// TestCase 9
NAryTree tree9(25);

return 0;
}
Output:
TestCase1MinimumIteration:6
==================================================================================
CloneaBinaryTreewithRandom
Pointers
GivenaBinaryTreewhereeverynodehasfollowingstructure.
structnode{
intkey
structnode*left,*right,*random
}
TherandompointerpointstoanyrandomnodeofthebinarytreeandcanevenpointtoNULL,clonethe
givenbinarytree.
Method1(UseHashing)
Theideaistostoremappingfromgiventreenodestoclonetrenodeinhashtable.Followingaredetailed
steps.
1)RecursivelytraversethegivenBinaryandcopykeyvalue,leftpointerandrightpointertoclonetree.
Whilecopying,storethemappingfromgiventreenodetoclonetreenodeinahashtable.Inthefollowing
pseudocode,cloneNodeiscurrentlyvisitednodeofclonetreeandtreeNodeiscurrentlyvisitednodeof
giventree.
cloneNode>key=treeNode>key
cloneNode>left=treeNode>left
cloneNode>right=treeNode>right
map[treeNode]=cloneNode
2)Recursivelytraversebothtreesandsetrandompointersusingentriesfromhashtable.
cloneNode>random=map[treeNode>random]
FollowingisC++implementationofaboveidea.ThefollowingimplementationusesmapfromC++STL.
Notethatmapdoesntimplementhashtable,itactuallyisbasedonselfbalancingbinarysearchtree.
// A hashmap based C++ program to clone a binary tree with random pointers
#include<iostream>
#include<map>
/* A binary tree node has data, pointer to left child, a pointer to right
child and a pointer to random node*/
struct Node
{
int key;
struct Node* left, *right, *random;
};

given data and NULL left, right and random pointers. */
{
temp->key = key;
temp->random = temp->right = temp->left = NULL;
return (temp);
}
/* Given a binary tree, print its Nodes in inorder*/

void printInorder(Node* node)
{
if (node == NULL)
return;
/* First recur on left sutree */

printInorder(node->left);
/* then print data of Node and its random */

cout << "[" << node->key << " ";
if (node->random == NULL)
cout << "NULL], ";
else
cout << node->random->key << "], ";
/* now recur on right subtree */

printInorder(node->right);
}
// This function creates clone by copying key and left and right pointers
// This function also stores mapping from given tree node to clone.
Node* copyLeftRightNode(Node* treeNode, map<Node *, Node *> *mymap)
{
if (treeNode == NULL)
return NULL;
Node* cloneNode = newNode(treeNode->key);
(*mymap)[treeNode] = cloneNode;
cloneNode->left = copyLeftRightNode(treeNode->left, mymap);
cloneNode->right = copyLeftRightNode(treeNode->right, mymap);
return cloneNode;
}
// This function copies random node by using the hashmap built by

// copyLeftRightNode()
void copyRandom(Node* treeNode, Node* cloneNode, map<Node *, Node *> *mymap)
{
if (cloneNode == NULL)
return;
cloneNode->random = (*mymap)[treeNode->random];
copyRandom(treeNode->left, cloneNode->left, mymap);
copyRandom(treeNode->right, cloneNode->right, mymap);
}
// This function makes the clone of given tree. It mainly uses

// copyLeftRightNode() and copyRandom()
Node* cloneTree(Node* tree)
{
if (tree == NULL)
return NULL;
map<Node *, Node *> *mymap = new map<Node *, Node *>;
Node* newTree = copyLeftRightNode(tree, mymap);
copyRandom(tree, newTree, mymap);
return newTree;
}

int main()
{
//Test No 1
Node *tree = newNode(1);

tree->left = newNode(2);
tree->right = newNode(3);
tree->left->left = newNode(4);
tree->left->right = newNode(5);
tree->random = tree->left->right;
tree->left->left->random = tree;
tree->left->right->random = tree->right;
// Test No 2
//tree = NULL;
// Test No 3
//tree = newNode(1);
// Test No 4
/*tree = newNode(1);
tree->random = tree->right;
tree->left->random = tree;
*/
cout << "Inorder traversal of original binary tree is: \n";

printInorder(tree);
Node *clone = cloneTree(tree);
cout << "\n\nInorder traversal of cloned binary tree is: \n";

printInorder(clone);
return 0;
}
Output:
Inordertraversaloforiginalbinarytreeis:
[41],[2NULL],[53],[15],[3NULL],
Inordertraversalofclonedbinarytreeis:
[41],[2NULL],[53],[15],[3NULL],
Method2(TemporarilyModifytheGivenBinaryTree)
1.Createnewnodesinclonedtreeandinserteachnewnodeinoriginaltreebetweentheleftpointeredge
ofcorrespondingnodeintheoriginaltree(Seethebelowimage).
i.e.ifcurrentnodeisAanditsleftchildisB(A>>B),thennewclonednodewithkeyAwilbecreated
(saycA)anditwillbeputasA>>cA>>B(BcanbeaNULLoranonNULLleftchild).Rightchild
pointerwillbesetcorrectlyi.e.ifforcurrentnodeA,rightchildisCinoriginaltree(A>>C)then
correspondingclonednodescAandcCwilllikecA>>cC
2.Setrandompointerinclonedtreeasperoriginaltree
i.e.ifnodeAsrandompointerpointstonodeB,theninclonedtree,cAwillpointtocB(cAandcBarenew
nodeinclonedtreecorrespondingtonodeAandBinoriginaltree)
3.Restoreleftpointerscorrectlyinbothoriginalandclonedtree
FollowingisC++implementationofabovealgorithm.
#include <iostream>
/* A binary tree node has data, pointer to left child, a pointer to right
child and a pointer to random node*/
struct Node
{
int key;
struct Node* left, *right, *random;
};

given data and NULL left, right and random pointers. */
{
temp->key = key;
temp->random = temp->right = temp->left = NULL;
return (temp);
}
/* Given a binary tree, print its Nodes in inorder*/

{
if (node == NULL)
return;
/* First recur on left sutree */

/* then print data of Node and its random */

cout << "[" << node->key << " ";
if (node->random == NULL)
cout << "NULL], ";
else
cout << node->random->key << "], ";

}
// This function creates new nodes cloned tree and puts new cloned node
// in between current node and it's left child
// i.e. if current node is A and it's left child is B ( A --- >> B ),
//
then new cloned node with key A wil be created (say cA) and
//
it will be put as
//
A --- >> cA --- >> B
// Here B can be a NULL or a non-NULL left child

// Right child pointer will be set correctly
// i.e. if for current node A, right child is C in original tree
// (A --- >> C) then corresponding cloned nodes cA and cC will like
// cA ---- >> cC
Node* copyLeftRightNode(Node* treeNode)
{
return NULL;
Node* left = treeNode->left;

treeNode->left = newNode(treeNode->key);
treeNode->left->left = left;
if(left != NULL)
left->left = copyLeftRightNode(left);
treeNode->left->right = copyLeftRightNode(treeNode->right);
return treeNode->left;
}
// This function sets random pointer in cloned tree as per original tree
// i.e. if node A's random pointer points to node B, then
// in cloned tree, cA wil point to cB (cA and cB are new node in cloned
// tree corresponding to node A and B in original tree)
void copyRandomNode(Node* treeNode, Node* cloneNode)
{
return;
if(treeNode->random != NULL)
cloneNode->random = treeNode->random->left;
else
cloneNode->random = NULL;
if(treeNode->left != NULL && cloneNode->left != NULL)

copyRandomNode(treeNode->left->left, cloneNode->left->left);
copyRandomNode(treeNode->right, cloneNode->right);
}
// This function will restore left pointers correctly in

// both original and cloned tree
void restoreTreeLeftNode(Node* treeNode, Node* cloneNode)
{
return;
if (cloneNode->left != NULL)
{
Node* cloneLeft = cloneNode->left->left;
treeNode->left = treeNode->left->left;
cloneNode->left = cloneLeft;
}
else
treeNode->left = NULL;
restoreTreeLeftNode(treeNode->left, cloneNode->left);
restoreTreeLeftNode(treeNode->right, cloneNode->right);
}
//This function makes the clone of given tree

Node* cloneTree(Node* treeNode)
{
return NULL;
Node* cloneNode = copyLeftRightNode(treeNode);
copyRandomNode(treeNode, cloneNode);
restoreTreeLeftNode(treeNode, cloneNode);
return cloneNode;
}

int main()
{
/* //Test No 1
tree->random = tree->left->right;
tree->left->left->random = tree;
tree->left->right->random = tree->right;
// Test No 2
// Node *tree = NULL;
/*
// Test No 3
// Test No 4
tree->random = tree->right;
Test No 5
tree->right->left = newNode(6);
tree->right->right = newNode(7);
tree->random = tree->left;
*/
// Test No 6
Node *n2 = newNode(6);

tree->left = n2;
tree->right = n3;
tree->random = n2;
n2->left = n4;
n2->right = n5;
n2->random = n8;
n3->left = n6;
n3->right = n7;
n3->random = n5;
n4->random = n9;
n5->left = n8;
n5->right = n9;
n5->random = tree;
n6->random = n9;
n9->random = n8;
/* Test No 7
tree->right->random = tree->left;
*/
cout << "Inorder traversal of original binary tree is: \n";
printInorder(tree);
Node *clone = cloneTree(tree);
cout << "\n\nInorder traversal of cloned binary tree is: \n";

printInorder(clone);
return 0;
}
Output:
Inordertraversaloforiginalbinarytreeis:
[59],[67],[7NULL],[810],[97],[106],[119],[128],[13NULL],
Inordertraversalofclonedbinarytreeis:
[59],[67],[7NULL],[810],[97],[106],[119],[128],[13NULL],
==================================================================================
Checkiftwonodesarecousinsina
BinaryTree
GiventhebinaryTreeandthetwonodessayaandb,determinewhetherthetwonodesarecousinsof
eachotherornot.
Twonodesarecousinsofeachotheriftheyareatsamelevelandhavedifferentparents.
Example
6
/\
35
/\/\
7813
Saytwonodebe7and1,resultisTRUE.
Saytwonodesare3and5,resultisFALSE.
Saytwonodesare7and5,resultisFALSE.
Theideaistofindlevelofoneofthenodes.Usingthefoundlevel,checkifaandbareatthislevel.Ifa
andbareatgivenlevel,thenfinallycheckiftheyarenotchildrenofsameparent.
FollowingisCimplementationoftheaboveapproach.
// C program to check if two Nodes in a binary tree are cousins
#include <stdio.h>
#include <stdlib.h>
// A Binary Tree Node

struct Node
{
int data;
struct Node *left, *right;
};
// A utility function to create a new Binary Tree Node

struct Node *newNode(int item)
{
struct Node *temp = (struct Node *)malloc(sizeof(struct Node));
temp->data = item;
return temp;
}
// Recursive function to check if two Nodes are siblings

int isSibling(struct Node *root, struct Node *a, struct Node *b)
{
// Base case
if (root==NULL) return 0;
return ((root->left==a && root->right==b)||

(root->left==b && root->right==a)||
isSibling(root->left, a, b)||
isSibling(root->right, a, b));
}
// Recursive function to find level of Node 'ptr' in a binary tree

int level(struct Node *root, struct Node *ptr, int lev)
{
// base cases
if (root == NULL) return 0;
if (root == ptr) return lev;
// Return level if Node is present in left subtree

int l = level(root->left, ptr, lev+1);
if (l != 0) return l;
// Else search in right subtree

return level(root->right, ptr, lev+1);
}
// Returns 1 if a and b are cousins, otherwise 0

int isCousin(struct Node *root, struct Node *a, struct Node *b)
{
//1. The two Nodes should be on the same level in the binary tree.
//2. The two Nodes should not be siblings (means that they should
// not have the same parent Node).
if ((level(root,a,1) == level(root,b,1)) && !(isSibling(root,a,b)))
return 1;
else return 0;
}
// Driver Program to test above functions

int main()
{
struct Node *root = newNode(1);
struct Node *Node1,*Node2;

Node1 = root->left->left;
Node2 = root->right->right;
isCousin(root,Node1,Node2)? puts("Yes"): puts("No");
return 0;
}
Ouput:
Yes
==================================================================================
Checkifabinarytreeissubtreeof
anotherbinarytree|Set2
Giventwobinarytrees,checkifthefirsttreeissubtreeofthesecondone.AsubtreeofatreeTisatreeS
consistingofanodeinTandallofitsdescendantsinT.
Thesubtreecorrespondingtotherootnodeistheentiretreethesubtreecorrespondingtoanyothernodeis
calledapropersubtree.
Forexample,inthefollowingcase,Tree1isasubtreeofTree2.
Tree1
x
/\
ab
\
c
Tree2
z
/\
xe
/\\
abk
\
c
WehavediscussedaO(n2)solutionforthisproblem.InthispostaO(n)solutionisdiscussed.Theideais
basedonthefactthatinorderandpreorder/postorderuniquelyidentifyabinarytree.TreeSisasubtreeofT
ifbothinorderandpreordertraversalsofSarewsubstringsofinorderandpreordertraversalsofT
respectively.
Followingaredetailedsteps.
1)FindinorderandpreordertraversalsofT,storethemintwoauxiliaryarraysinT[]andpreT[].
2)FindinorderandpreordertraversalsofS,storethemintwoauxiliaryarraysinS[]andpreS[].
3)IfinS[]isasubarrayofinT[]andpreS[]isasubarraypreT[],thenSisasubtreeofT.Elsenot.
Wecanalsousepostordertraversalinplaceofpreorderintheabovealgorithm.
Letusconsidertheaboveexample
InorderandPreordertraversalsofthebigtreeare.
inT[]={a,c,x,b,z,e,k}
preT[]={z,x,a,c,b,e,k}
InorderandPreordertraversalsofsmalltreeare
inS[]={a,c,x,b}
preS[]={x,a,c,b}
WecaneasilyfigureoutthatinS[]isasubarrayof
inT[]andpreS[]isasubarrayofpreT[].
EDIT
Theabovealgorithmdoesn'tworkforcaseswhereatreeispresent
inanothertree,butnotasasubtree.Considerthefollowingexample.
Tree1
x
/\
ab
/
Tree2
x
/\
ab
/\
cd
InorderandPreordertraversalsofthebigtreeorTree2are.
InorderandPreordertraversalsofsmalltreeorTree1are
TheTree2isnotasubtreeofTree1,butinS[]andpreS[]are
subarraysofinT[]andpreT[]respectively.
Theabovealgorithmcanbeextendedtohandlesuchcasesbyaddingaspecialcharacterwheneverwe
encounterNULLininorderandpreordertraversals.ThankstoShivamGoelforsuggestingthisextension.
#include <iostream>
#include <cstring>
#define MAX 100
// Structure of a tree node

struct Node
{
char key;
};
// A utility function to create a new BST node

Node *newNode(char item)
{
temp->key = item;
return temp;
}
// A utility function to store inorder traversal of tree rooted

// with root in an array arr[]. Note that i is passed as reference
void storeInorder(Node *root, char arr[], int &i)
{
if (root == NULL)
{
arr[i++] = '$';
return;
}
storeInorder(root->left, arr, i);
arr[i++] = root->key;
storeInorder(root->right, arr, i);
}
// A utility function to store preorder traversal of tree rooted

// with root in an array arr[]. Note that i is passed as reference
void storePreOrder(Node *root, char arr[], int &i)
{
if (root == NULL)
{
arr[i++] = '$';
return;
}
arr[i++] = root->key;
storePreOrder(root->left, arr, i);
storePreOrder(root->right, arr, i);
}
/* This function returns true if S is a subtree of T, otherwise false */

bool isSubtree(Node *T, Node *S)
{
/* base cases */
if (S == NULL) return true;
if (T == NULL) return false;
// Store Inorder traversals of T and S in inT[0..m-1]

// and inS[0..n-1] respectively
int m = 0, n = 0;
char inT[MAX], inS[MAX];
storeInorder(T, inT, m);
storeInorder(S, inS, n);
inT[m] = '\0', inS[n] = '\0';
// If inS[] is not a substring of preS[], return false

if (strstr(inT, inS) == NULL)
return false;
// Store Preorder traversals of T and S in inT[0..m-1]

// and inS[0..n-1] respectively
m = 0, n = 0;
char preT[MAX], preS[MAX];
storePreOrder(T, preT, m);
storePreOrder(S, preS, n);
preT[m] = '\0', preS[n] = '\0';
// If inS[] is not a substring of preS[], return false

// Else return true
return (strstr(preT, preS) != NULL);
}
// Driver program to test above function

int main()
{
Node *T = newNode('a');
T->left = newNode('b');
T->right = newNode('d');
T->left->left = newNode('c');
T->right->right = newNode('e');
Node *S = newNode('a');
S->left = newNode('b');
S->left->left = newNode('c');
S->right = newNode('d');
if (isSubtree(T, S))
cout << "Yes: S is a subtree of T";
else
cout << "No: S is NOT a subtree of T";
return 0;
}
Output:
No:SisNOTasubtreeofT
==================================================================================
Inorderpredecessorandsuccessor
foragivenkeyinBST
Irecentlyencounteredwithaquestioninaninterviewatecommercecompany.Theintervieweraskedthe
followingquestion:
ThereisBSTgivenwithrootnodewithkeypartasintegeronly.Thestructureofeachnodeisasfollows:
struct Node
{
int key;
struct Node *left, *right ;
};
Youneedtofindtheinordersuccessorandpredecessorofagivenkey.Incasethegivenkeyisnotfoundin
BST,thenreturnthetwovalueswithinwhichthiskeywilllie.
Followingisthealgorithmtoreachthedesiredresult.Itsarecursivemethod:
Input:rootnode,key
output:predecessornode,successornode
1.IfrootisNULL
thenreturn
2.ifkeyisfoundthen
a.Ifitsleftsubtreeisnotnull
Thenpredecessorwillbetherightmost
childofleftsubtreeorleftchilditself.
b.Ifitsrightsubtreeisnotnull
Thesuccessorwillbetheleftmostchild
ofrightsubtreeorrightchilditself.
return
3.Ifkeyissmallerthenrootnode
setthesuccessorasroot
searchrecursivelyintoleftsubtree
else
setthepredecessorasroot
searchrecursivelyintorightsubtree
FollowingisC++implementationoftheabovealgorithm:
// C++ program to find predecessor and successor in a BST
#include <iostream>
// BST Node
struct Node
{
int key;
};
// This function finds predecessor and successor of key in BST.

// It sets pre and suc as predecessor and successor respectively
void findPreSuc(Node* root, Node*& pre, Node*& suc, int key)
{
// Base case
if (root == NULL) return ;
// If key is present at root

if (root->key == key)
{
// the maximum value in left subtree is predecessor
{
Node* tmp = root->left;
while (tmp->right)
tmp = tmp->right;
pre = tmp ;
}
// the minimum value in right subtree is successor

if (root->right != NULL)
{
Node* tmp = root->right ;
while (tmp->left)
tmp = tmp->left ;
suc = tmp ;
}
return ;
}
// If key is smaller than root's key, go to left subtree

if (root->key > key)
{
suc = root ;
findPreSuc(root->left, pre, suc, key) ;
}
else // go to right subtree
{
pre = root ;
findPreSuc(root->right, pre, suc, key) ;
}
}

Node *newNode(int item)
{
temp->key = item;
return temp;
}
/* A utility function to insert a new node with given key in BST */

Node* insert(Node* node, int key)
{
if (node == NULL) return newNode(key);
if (key < node->key)
node->left = insert(node->left, key);
else
node->right = insert(node->right, key);
return node;
}

int main()
{
int key = 65;
//Key to be searched in BST
/* Let us create following BST

50
/\
30
70
/ \/ \
20
40 60
80 */
Node *root = NULL;

root = insert(root, 50);
insert(root, 30);
insert(root, 20);
insert(root, 40);
insert(root, 70);
insert(root, 60);
insert(root, 80);
Node* pre = NULL, *suc = NULL;
findPreSuc(root, pre, suc, key);

if (pre != NULL)
cout << "Predecessor is " << pre->key << endl;
else
cout << "No Predecessor";
if (suc != NULL)
cout << "Successor is " << suc->key;
else
cout << "No Successor";
return 0;
}
Output:
Predecessoris60
Successoris70
==================================================================================
Findthemaximumpathsum
betweentwoleavesofabinarytree
Givenabinarytreeinwhicheachnodeelementcontainsanumber.Findthemaximumpossiblesumfrom
oneleafnodetoanother.
Themaximumsumpathmayormaynotgothroughroot.Forexample,inthefollowingbinarytree,the
maximumsumis27(3+6+9+01+10).ExpectedtimecomplexityisO(n).
AsimplesolutionistotraversethetreeanddofollowingforeverytraversednodeX.
1)FindmaximumsumfromleaftorootinleftsubtreeofX(wecanusethispostforthisandnextsteps)
2)FindmaximumsumfromleaftorootinrightsubtreeofX.
3)AddtheabovetwocalculatedvaluesandX>dataandcomparethesumwiththemaximumvalue
obtainedsofarandupdatethemaximumvalue.
4)Returnthemaximumvalue.
ThetimecomplexityofabovesolutionisO(n2)
Wecanfindthemaximumsumusingsingletraversalofbinarytree.Theideaistomaintaintwovalues
inrecursivecalls
1)Maximumroottoleafpathsumforthesubtreerootedundercurrentnode.
2)Themaximumpathsumbetweenleaves(desiredoutput).
ForeveryvisitednodeX,wefindthemaximumroottoleafsuminleftandrightsubtreesofX.Weaddthe
twovalueswithX>data,andcomparethesumwithmaximumpathsumfoundsofar.
FollowingisC++implementationoftheaboveO(n)solution.
// C++ program to find maximum path sum between two leaves of
// a binary tree
#include <iostream>
// A binary tree node

struct Node
{
int data;
};
// Utility function to allocate memory for a new node

struct Node* newNode(int data)
{
struct Node* node = new(struct Node);
node->data = data;
node->left = node->right = NULL;
return (node);
}
// Utility function to find maximum of two integers

int max(int a, int b)
{
return (a >= b)? a: b; }
// A utility function to find the maximum sum between any two leaves.
// This function calculates two values:
// 1) Maximum path sum between two leaves which is stored in res.
// 2) The maximum root to leaf path sum which is returned.
int maxPathSumUtil(struct Node *root, int &res)
{
// Base case
if (root==NULL) return 0;
// Find maximum sum in left and right subtree. Also find

// maximum root to leaf sums in left and right subtrees
// and store them in lLPSum and rLPSum
int lLPSum = maxPathSumUtil(root->left, res);
int rLPSum = maxPathSumUtil(root->right, res);
// Find the maximum path sum passing through root

int curr_sum = max((lLPSum+rLPSum+root->data), max(lLPSum, rLPSum));
// Update res (or result) if needed

if (res < curr_sum)
res = curr_sum;
// Return the maximum root to leaf path sum

return max(lLPSum, rLPSum)+root->data;
}
// The main function which returns sum of the maximum

// sum path between two leaves. This function mainly uses
// maxPathSumUtil()
int maxPathSum(struct Node *root)
{
int res = 0;
maxPathSumUtil(root, res);
return res;
}
// driver program to test above function

int main()
{
struct Node *root = newNode(-15);
root->left->left = newNode(-8);
root->right->right->right= newNode(0);
root->right->right->right->left= newNode(4);
root->right->right->right->right= newNode(-1);
root->right->right->right->right->left= newNode(10);
cout << "Max pathSum of the given binary tree is " << maxPathSum(root);
return 0;
}
Output:
MaxpathSumofthegivenbinarytreeis27.
==================================================================================
Reversealternatelevelsofaperfect
binarytree
GivenaPerfectBinaryTree,reversethealternatelevelnodesofthebinarytree.
Giventree:
a
/\
bc
/\/\
defg
/\/\/\/\
hijklmno
Modifiedtree:
a
/\
cb
/\/\
defg
/\/\/\/\
onmlkjih
Asimplesolutionistodofollowingsteps.
1)Accessnodeslevelbylevel.
2)Ifcurrentlevelisodd,thenstorenodesofthislevelinanarray.
3)Reversethearrayandstoreelementsbackintree.
Atrickysolutionistodotwoinordertraversals.Followingarestepstobefollowed.
1)Traversethegiventreeininorderfashionandstorealloddlevelnodesinanauxiliaryarray.Forthe
aboveexamplegiventree,contentsofarraybecome{h,i,b,j,k,l,m,c,n,o}
2)Reversethearray.Thearraynowbecomes{o,n,c,m,l,k,j,b,i,h}
3)Traversethetreeagaininorderfashion.Whiletraversingthetree,onebyonetakeelementsfromarray
andstoreelementsfromarraytoeveryoddleveltraversednode.
Fortheaboveexample,wetraversehfirstinabovearrayandreplacehwitho.Thenwetraverseiand
replaceitwithn.
FollowingisC++implementationoftheabovealgorithm.
// C++ program to reverse alternate levels of a binary tree
#include<iostream>
#define MAX 100
// A Binary Tree node

struct Node
{
char data;
};

struct Node *newNode(char item)
{
struct Node *temp = new Node;
temp->data = item;
return temp;
}
// Function to store nodes of alternate levels in an array

void storeAlternate(Node *root, char arr[], int *index, int l)
{
// Base case
// Store elements of left subtree

storeAlternate(root->left, arr, index, l+1);
// Store this node only if this is a odd level node

if (l%2 != 0)
{
arr[*index] = root->data;
(*index)++;
}
// Store elements of right subtree

storeAlternate(root->right, arr, index, l+1);
}
// Function to modify Binary Tree (All odd level nodes are

// updated by taking elements from array in inorder fashion)
void modifyTree(Node *root, char arr[], int *index, int l)
{
// Base case
// Update nodes in left subtree

modifyTree(root->left, arr, index, l+1);
// Update this node only if this is an odd level node

if (l%2 != 0)
{
root->data = arr[*index];
(*index)++;
}
// Update nodes in right subtree

modifyTree(root->right, arr, index, l+1);
}
// A utility function to reverse an array from index

// 0 to n-1
void reverse(char arr[], int n)
{
int l = 0, r = n-1;
while (l < r)
{
int temp = arr[l];
arr[l] = arr[r];
arr[r] = temp;
l++; r--;
}
}
// The main function to reverse alternate nodes of a binary tree

void reverseAlternate(struct Node *root)
{
// Create an auxiliary array to store nodes of alternate levels
char *arr = new char[MAX];
int index = 0;
// First store nodes of alternate levels

storeAlternate(root, arr, &index, 0);
// Reverse the array

reverse(arr, index);
// Update tree by taking elements from array

index = 0;
modifyTree(root, arr, &index, 0);
}
// A utility function to print indorder traversal of a
// binary tree
void printInorder(struct Node *root)
{
printInorder(root->left);
cout << root->data << " ";
printInorder(root->right);
}

int main()
{
struct Node *root = newNode('a');
root->left = newNode('b');
root->right = newNode('c');
root->left->left = newNode('d');
root->left->right = newNode('e');
root->right->left = newNode('f');
root->right->right = newNode('g');
root->left->left->left = newNode('h');
root->left->left->right = newNode('i');
root->left->right->left = newNode('j');
root->left->right->right = newNode('k');
root->right->left->left = newNode('l');
root->right->left->right = newNode('m');
root->right->right->left = newNode('n');
root->right->right->right = newNode('o');
cout << "Inorder Traversal of given tree\n";

printInorder(root);
reverseAlternate(root);
cout << "\n\nInorder Traversal of modified tree\n";

printInorder(root);
return 0;
}
Output:
InorderTraversalofgiventree
hdibjekalfmcngo
InorderTraversalofmodifiedtree
odncmelakfjbigh
TimecomplexityoftheabovesolutionisO(n)asitdoestwoinordertraversalsofbinarytree.
==================================================================================
TransformaBSTtogreatersumtree
GivenaBST,transformitintogreatersumtreewhereeachnodecontainssumofallnodesgreaterthanthat
node.
Westronglyrecommendtominimizethegbrowserandtrythisyourselffirst.
Method1(Naive):
ThismethoddoesntrequirethetreetobeaBST.Followingaresteps.
1.Traversenodebynode(Inorder,preorder,etc.)
2.Foreachnodefindallthenodesgreaterthanthatofthecurrentnode,sumthevalues.Storeallthese
sums.
3.ReplaceeachnodevaluewiththeircorrespondingsumbytraversinginthesameorderasinStep1.
ThistakesO(n^2)TimeComplexity.
Method2(Usingonlyonetraversal)
ByleveragingthefactthatthetreeisaBST,wecanfindanO(n)solution.TheideaistotraverseBSTin
reverseinorder.ReverseinordertraversalofaBSTgivesuskeysindecreasingorder.Beforevisitinga
node,wevisitallgreaternodesofthatnode.Whiletraversingwekeeptrackofsumofkeyswhichisthesum
ofallthekeysgreaterthanthekeyofcurrentnode.
// C++ program to transform a BST to sum tree
#include<iostream>
// A BST node
struct Node
{
int data;
};

{
struct Node *temp = new Node;
temp->data = item;
return temp;
}
// Recursive function to transform a BST to sum tree.

// This function traverses the tree in reverse inorder so
// that we have visited all greater key nodes of the currently
// visited node
void transformTreeUtil(struct Node *root, int *sum)
{
// Base case
// Recur for right subtree

transformTreeUtil(root->right, sum);
// Update sum
*sum = *sum + root->data;
// Store old sum in current node

root->data = *sum - root->data;
// Recur for left subtree

transformTreeUtil(root->left, sum);
}
// A wrapper over transformTreeUtil()

void transformTree(struct Node *root)
{
int sum = 0; // Initialize sum
transformTreeUtil(root, &sum);
}
// A utility function to print indorder traversal of a

// binary tree
void printInorder(struct Node *root)
{
printInorder(root->left);
printInorder(root->right);
}

int main()
{
cout << "Inorder Traversal of given tree\n";

printInorder(root);
transformTree(root);
cout << "\n\nInorder Traversal of transformed tree\n";

printInorder(root);
return 0;
}
Output:
InorderTraversalofgiventree
1271115293540
InorderTraversaloftransformedtree
13913713011910475400
TimecomplexityofthismethodisO(n)asitdoesasimpletraversaloftree.
==================================================================================
PrintaBinaryTreeinVertical
Order|Set2(Hashmapbased
Method)
Givenabinarytree,printitvertically.Thefollowingexampleillustratesverticalordertraversal.
1
/\
23
/\/\
4567
\\
89

Theoutputofprintthistreeverticallywillbe:
4
2
156
38
7
9
WehavediscussedaO(n2)solutioninthepreviouspost.Inthispost,anefficientsolutionbasedonhash
mapisdiscussed.WeneedtochecktheHorizontalDistancesfromrootforallnodes.Iftwonodeshavethe
sameHorizontalDistance(HD),thentheyareonsameverticalline.TheideaofHDissimple.HDforrootis
0,arightedge(edgeconnectingtorightsubtree)isconsideredas+1horizontaldistanceandaleftedgeis
consideredas1horizontaldistance.Forexample,intheabovetree,HDforNode4isat2,HDforNode2
is1,HDfor5and6is0andHDfornode7is+2.
WecandoinordertraversalofthegivenBinaryTree.Whiletraversingthetree,wecanrecursivelycalculate
HDs.Weinitiallypassthehorizontaldistanceas0forroot.Forleftsubtree,wepasstheHorizontalDistance
asHorizontaldistanceofrootminus1.Forrightsubtree,wepasstheHorizontalDistanceasHorizontal
Distanceofrootplus1.ForeveryHDvalue,wemaintainalistofnodesinahaspmap.Wheneverweseea
nodeintraversal,wegotothehashmapentryandaddthenodetothehashmapusingHDasakeyin
map.
FollowingisC++implementationoftheabovemethod.ThankstoChiragforprovidingthebelowC++
implementation.
// C++ program for printing vertical order of a given binary tree
#include <iostream>
#include <vector>
#include <map>
// Structure for a binary tree node

struct Node
{
int key;
Node *left, *right;
};
// A utility function to create a new node
struct Node* newNode(int key)

{
struct Node* node = new Node;
node->key = key;
return node;
}
// Utility function to store vertical order in map 'm'

// 'hd' is horigontal distance of current node from root.
// 'hd' is initally passed as 0
void getVerticalOrder(Node* root, int hd, map<int, vector<int>> &m)
{
// Base case
if (root == NULL)
return;
// Store current node in map 'm'

m[hd].push_back(root->key);
// Store nodes in left subtree

getVerticalOrder(root->left, hd-1, m);
// Store nodes in right subtree

getVerticalOrder(root->right, hd+1, m);
}
// The main function to print vertical oder of a binary tree

// with given root
void printVerticalOrder(Node* root)
{
// Create a map and store vertical oder in map using
// function getVerticalOrder()
map < int,vector<int> > m;
int hd = 0;
getVerticalOrder(root, hd,m);
// Traverse the map and print nodes at every horigontal

// distance (hd)
map< int,vector<int> > :: iterator it;
for (it=m.begin(); it!=m.end(); it++)
{
for (int i=0; i<it->second.size(); ++i)
cout << it->second[i] << " ";
cout << endl;
}
}

int main()
{
cout << "Vertical order traversal is \n";
printVerticalOrder(root);
return 0;
}
Output:
Verticalordertraversalis
4
2
156
38
7
9
TimeComplexityofhashingbasedsolutioncanbeconsideredasO(n)undertheassumptionthatwehave
goodhashingfunctionthatallowsinsertionandretrievaloperationsinO(1)time.IntheaboveC++
implementation,mapofSTLisused.mapinSTListypicallyimplementedusingaSelfBalancingBinary
SearchTreewherealloperationstakeO(Logn)time.Thereforetimecomplexityofaboveimplementationis
O(nLogn).
==================================================================================
PrintRightViewofaBinaryTree
GivenaBinaryTree,printRightviewofit.RightviewofaBinaryTreeissetofnodesvisiblewhentreeis
visitedfromRightside.
Rightviewoffollowingtreeis1378
1
/\
23
/\/\
4567
\
8
TheRightviewcontainsallnodesthatarelastnodesintheirlevels.Asimplesolutionistodolevelorder
traversalandprintthelastnodeineverylevel.
Theproblemcanalsobesolvedusingsimplerecursivetraversal.Wecankeeptrackoflevelofanodeby
passingaparametertoallrecursivecalls.Theideaistokeeptrackofmaximumlevelalso.Andtraversethe
treeinamannerthatrightsubtreeisvisitedbeforeleftsubtree.Wheneverweseeanodewhoselevelis
morethanmaximumlevelsofar,weprintthenodebecausethisisthelastnodeinitslevel(Notethatwe
traversetherightsubtreebeforeleftsubtree).FollowingisCimplementationofthisapproach.
// C program to print right view of Binary Tree
#include<stdio.h>
#include<stdlib.h>
struct Node
{
int data;
};

{
struct Node *temp = (struct Node *)malloc(sizeof(struct Node));
temp->data = item;
return temp;
}
// Recursive function to print right view of a binary tree.

void rightViewUtil(struct Node *root, int level, int *max_level)
{
// Base Case
if (root==NULL) return;
// If this is the last Node of its level

if (*max_level < level)
{
printf("%d\t", root->data);
*max_level = level;
}
// Recur for right subtree first, then left subtree

rightViewUtil(root->right, level+1, max_level);
rightViewUtil(root->left, level+1, max_level);
}
// A wrapper over rightViewUtil()

void rightView(struct Node *root)
{
int max_level = 0;
rightViewUtil(root, 1, &max_level);
}

int main()
{
rightView(root);
return 0;
}
Output:
1378
TimeComplexity:Thefunctiondoesasimpletraversalofthetree,sothecomplexityisO(n).
==================================================================================
RedBlackTree|Set3(Delete)
WehavediscussedfollowingtopicsonRedBlacktreeinpreviousposts.Westronglyrecommendtorefer
followingpostasprerequisiteofthispost.
RedBlackTreeIntroduction
RedBlackTreeInsert
InsertionVsDeletion:
LikeInsertion,recoloringandrotationsareusedtomaintaintheRedBlackproperties.
Ininsertoperation,wecheckcolorofuncletodecidetheappropriatecase.Indeleteoperation,wecheck
colorofsiblingtodecidetheappropriatecase.
Themainpropertythatviolatesafterinsertionistwoconsecutivereds.Indelete,themainviolatedproperty
is,changeofblackheightinsubtreesasdeletionofablacknodemaycausereducedblackheightinone
roottoleafpath.
Deletionisfairlycomplexprocess.Tounderstanddeletion,notionofdoubleblackisused.Whenablack
nodeisdeletedandreplacedbyablackchild,thechildismarkedasdoubleblack.Themaintasknow
becomestoconvertthisdoubleblacktosingleblack.
DeletionSteps
Followingaredetailedstepsfordeletion.
1)PerformstandardBSTdelete.WhenweperformstandarddeleteoperationinBST,wealwaysendup
deletinganodewhichiseitherleaforhasonlyonechild(Foraninternalnode,wecopythesuccessorand
thenrecursivelycalldeleteforsuccessor,successorisalwaysaleafnodeoranodewithonechild).Sowe
onlyneedtohandlecaseswhereanodeisleaforhasonechild.Letvbethenodedeletedandubethe
childthatreplacesv.
2)SimpleCase:Ifeitheruorvisred,wemarkthereplacedchildasblack(Nochangeinblackheight).
Notethatbothuandvcannotberedasvisparentofuandtwoconsecutiveredsarenotallowedin
redblacktree.
3)IfBothuandvareBlack.
3.1)Coloruasdoubleblack.Nowourtaskreducestoconvertthisdoubleblacktosingleblack.NotethatIf
visleaf,thenuisNULLandcolorofNULLisconsideredasblack.Sothedeletionofablackleafalso
causesadoubleblack.
3.2)Dofollowingwhilethecurrentnodeuisdoubleblackoritisnotroot.Letsiblingofnodebes.
.(a):Ifsiblingsisblackandatleastoneofsiblingschildrenisred,performrotation(s).Letthered
childofsber.Thiscasecanbedividedinfoursubcasesdependinguponpositionsofsandr.
..(i)LeftLeftCase(sisleftchildofitsparentandrisleftchildofsorbothchildrenofsarered).
Thisismirrorofrightrightcaseshowninbelowdiagram.
..(ii)LeftRightCase(sisleftchildofitsparentandrisrightchild).Thisismirrorofrightleftcase
showninbelowdiagram.
..(iii)RightRightCase(sisrightchildofitsparentandrisrightchildofsorbothchildrenofsare
red)
..(iv)RightLeftCase(sisrightchildofitsparentandrisleftchildofs)
..(b):Ifsiblingisblackanditsbothchildrenareblack,performrecoloring,andrecurfortheparentif
parentisblack.

Inthiscase,ifparentwasred,thenwedidntneedtorecurforprent,wecansimplymakeitblack(red+
doubleblack=singleblack)
..(c):Ifsiblingisred,performarotationtomoveoldsiblingup,recolortheoldsiblingandparent.The
newsiblingisalwaysblack(Seethebelowdiagram).Thismainlyconvertsthetreetoblacksiblingcase(by
rotation)andleadstocase(a)or(b).Thiscasecanbedividedintwosubcases.
..(i)LeftCase(sisleftchildofitsparent).Thisismirrorofrightrightcaseshowninbelowdiagram.
Werightrotatetheparentp.
..(iii)RightCase(sisrightchildofitsparent).Weleftrotatetheparentp.

3.3)Ifuisroot,makeitsingleblackandreturn(Blackheightofcompletetreereducesby1).
==================================================================================
ConstructatreefromInorderand
Levelordertraversals
GiveninorderandlevelordertraversalsofaBinaryTree,constructtheBinaryTree.Followingisanexample
toillustratetheproblem.
Input:TwoarraysthatrepresentInorder
andlevelordertraversalsofa
BinaryTree
in[]={4,8,10,12,14,20,22}
level[]={20,8,22,4,12,10,14}
Output:Constructthetreerepresented
bythetwoarrays.
Fortheabovetwoarrays,the
constructedtreeisshownin
thediagramonrightside
Thefollowingpostcanbeconsideredasaprerequisiteforthis.
ConstructTreefromgivenInorderandPreordertraversals
Letusconsidertheaboveexample.
in[]={4,8,10,12,14,20,22}
level[]={20,8,22,4,12,10,14}
InaLevelordersequence,thefirstelementistherootofthetree.Soweknow20isrootforgiven
sequences.Bysearching20inInordersequence,wecanfindoutallelementsonleftsideof20areinleft
subtreeandelementsonrightareinrightsubtree.Soweknowbelowstructurenow.
20
/\
/\
{4,8,10,12,14}{22}
Letuscall{4,8,10,12,14}asleftsubarrayinInordertraversaland{22}asrightsubarrayinInordertraversal.
Inlevelordertraversal,keysofleftandrightsubtreesarenotconsecutive.Soweextractallnodesfromlevel
ordertraversalwhichareinleftsubarrayofInordertraversal.Toconstructtheleftsubtreeofroot,werecur
fortheextractedelementsfromlevelordertraversalandleftsubarrayofinordertraversal.Intheabove
example,werecurforfollowingtwoarrays.
//Recurforfollowingarraystoconstructtheleftsubtree
In[]={4,8,10,12,14}
level[]={8,4,12,10,14}
Similarly,werecurforfollowingtwoarraysandconstructtherightsubtree.
//Recurforfollowingarraystoconstructtherightsubtree
In[]={22}
level[]={22}
FollowingisC++implementationoftheaboveapproach.
/* program to construct tree using inorder and levelorder traversals */
#include <iostream>
/* A binary tree node */

struct Node
{
int key;
};
/* Function to find index of value in arr[start...end] */

int search(int arr[], int strt, int end, int value)
{
for (int i = strt; i <= end; i++)
if (arr[i] == value)
return i;
return -1;
}
// n is size of level[], m is size of in[] and m < n. This

// function extracts keys from level[] which are present in
// in[]. The order of extracted keys must be maintained

int *extrackKeys(int in[], int level[], int m, int n)
{
int *newlevel = new int[m], j = 0;
for (int i = 0; i < n; i++)
if (search(in, 0, m-1, level[i]) != -1)
newlevel[j] = level[i], j++;
return newlevel;
}
/* function that allocates a new node with the given key */

{
node->key = key;
return (node);
}
/* Recursive function to construct binary tree of size n from

Inorder traversal in[] and Level Order traversal level[].
inSrt and inEnd are start and end indexes of array in[]
Initial values of inStrt and inEnd should be 0 and n -1.
The function doesn't do any error checking for cases
where inorder and levelorder do not form a tree */
Node* buildTree(int in[], int level[], int inStrt, int inEnd, int n)
{
// If start index is more than the end index

if (inStrt > inEnd)
return NULL;
/* The first node in level order traversal is root */

Node *root = newNode(level[0]);
/* If this node has no children then return */

if (inStrt == inEnd)
return root;
/* Else find the index of this node in Inorder traversal */

int inIndex = search(in, inStrt, inEnd, root->key);
// Extract left subtree keys from level order traversal

int *llevel = extrackKeys(in, level, inIndex, n);
// Extract right subtree keys from level order traversal

int *rlevel = extrackKeys(in + inIndex + 1, level, n-inIndex-1, n);
/* construct left and right subtress */

root->left = buildTree(in, llevel, inStrt, inIndex-1, n);
root->right = buildTree(in, rlevel, inIndex+1, inEnd, n);
// Free memory to avoid memory leak

delete [] llevel;
delete [] rlevel;
return root;
}
/* Uti;ity function to print inorder traversal of binary tree */

{
if (node == NULL)
return;
cout << node->key << " ";
}
/* Driver program to test above functions */

int main()
{
int in[] = {4, 8, 10, 12, 14, 20, 22};
int level[] = {20, 8, 22, 4, 12, 10, 14};
int n = sizeof(in)/sizeof(in[0]);
Node *root = buildTree(in, level, 0, n - 1, n);
/* Let us test the built tree by printing Insorder traversal */

cout << "Inorder traversal of the constructed tree is \n";
printInorder(root);
return 0;
}
Output:
Inordertraversaloftheconstructedtreeis
481012142022
AnupperboundontimecomplexityofabovemethodisO(n3).Inthemainrecursivefunction,extractNodes()
iscalledwhichtakesO(n2)time.
==================================================================================
Printallnodesatdistancekfroma
givennode
Givenabinarytree,atargetnodeinthebinarytree,andanintegervaluek,printallthenodesthatareat
distancekfromthegiventargetnode.Noparentpointersareavailable.
Considerthetreeshownindiagram
Input:target=pointertonodewithdata8.
root=pointertonodewithdata20.
k=2.
Output:101422
Iftargetis14andkis3,thenoutput
shouldbe"420"
Therearetwotypesofnodestobeconsidered.
1)Nodesinthesubtreerootedwithtargetnode.Forexampleifthetargetnodeis8andkis2,thensuch
nodesare10and14.
2)Othernodes,maybeanancestoroftarget,oranodeinsomeothersubtree.Fortargetnode8andkis2,
thenode22comesinthiscategory.
Findingthefirsttypeofnodesiseasytoimplement.Justtraversesubtreesrootedwiththetargetnodeand
decrementkinrecursivecall.Whenthekbecomes0,printthenodecurrentlybeingtraversed(Seethisfor
moredetails).HerewecallthefunctionasprintkdistanceNodeDown().
Howtofindnodesofsecondtype?Fortheoutputnodesnotlyinginthesubtreewiththetargetnodeasthe
root,wemustgothroughallancestors.Foreveryancestor,wefinditsdistancefromtargetnode,letthe
distancebed,nowwegotoothersubtree(iftargetwasfoundinleftsubtree,thenwegotorightsubtreeand
viceversa)oftheancestorandfindallnodesatkddistancefromtheancestor.
#include <iostream>
// A binary Tree node

struct node
{
int data;
struct node *left, *right;
};
/* Recursive function to print all the nodes at distance k in the
tree (or subtree) rooted with given root. See */

void printkdistanceNodeDown(node *root, int k)
{
// Base Case
if (root == NULL || k < 0) return;
// If we reach a k distant node, print it

if (k==0)
{
cout << root->data << endl;
return;
}
// Recur for left and right subtrees

printkdistanceNodeDown(root->left, k-1);
printkdistanceNodeDown(root->right, k-1);
}
// Prints all nodes at distance k from a given target node.

// The k distant nodes may be upward or downward. This function
// Returns distance of root from target node, it returns -1 if target
// node is not present in tree rooted with root.
int printkdistanceNode(node* root, node* target , int k)
{
// Base Case 1: If tree is empty, return -1
if (root == NULL) return -1;
// If target is same as root. Use the downward function

// to print all nodes at distance k in subtree rooted with
// target or root
if (root == target)
{
printkdistanceNodeDown(root, k);
return 0;
}
int dl = printkdistanceNode(root->left, target, k);
// Check if target node was found in left subtree

if (dl != -1)
{
// If root is at distance k from target, print root
// Note that dl is Distance of root's left child from target
if (dl + 1 == k)
// Else go to right subtree and print all k-dl-2 distant nodes

// Note that the right child is 2 edges away from left child
else
printkdistanceNodeDown(root->right, k-dl-2);
// Add 1 to the distance and return value for parent calls

return 1 + dl;
}
// MIRROR OF ABOVE CODE FOR RIGHT SUBTREE

// Note that we reach here only when node was not found in left subtree
int dr = printkdistanceNode(root->right, target, k);
if (dr != -1)
{
if (dr + 1 == k)
else
printkdistanceNodeDown(root->left, k-dr-2);
return 1 + dr;
}
// If target was neither present in left nor in right subtree

return -1;
}
// A utility function to create a new binary tree node

node *newnode(int data)
{
node *temp = new node;
temp->data = data;
return temp;
}

int main()
{
/* Let us construct the tree shown in above diagram */
node * root = newnode(20);
root->left = newnode(8);
root->right = newnode(22);
root->left->left = newnode(4);
root->left->right = newnode(12);
root->left->right->left = newnode(10);
root->left->right->right = newnode(14);
node * target = root->left->right;
printkdistanceNode(root, target, 2);
return 0;
}
Output:
4
20
TimeComplexity:AtfirstlookthetimecomplexitylooksmorethanO(n),butifwetakeacloserlook,wecan
observethatnonodeistraversedmorethantwice.ThereforethetimecomplexityisO(n).
==================================================================================
PrintaBinaryTreeinVertical
Order|Set1
Givenabinarytree,printitvertically.Thefollowingexampleillustratesverticalordertraversal.
1
/\
23
/\/\
4567
\\
89
Theoutputofprintthistreeverticallywillbe:
4
2
156
38
7
9
Theideaistotraversethetreeonceandgettheminimumandmaximumhorizontaldistancewithrespectto
root.Forthetreeshownabove,minimumdistanceis2(fornodewithvalue4)andmaximumdistanceis3
(Fornodewithvalue9).
Oncewehavemaximumandminimumdistancesfromroot,weiterateforeachverticallineatdistance
minimumtomaximumfromroot,andforeachverticallinetraversethetreeandprintthenodeswhichlieon
thatverticalline.
Algorithm:
//min>Minimumhorizontaldistancefromroot
//max>Maximumhorizontaldistancefromroot
//hd>Horizontaldistanceofcurrentnodefromroot
findMinMax(tree,min,max,hd)
iftreeisNULLthenreturn
ifhdislessthanminthen
min=hd
elseifhdisgreaterthanmaxthen
*max=hd
findMinMax(tree>left,min,max,hd1)
findMinMax(tree>right,min,max,hd+1)
printVerticalLine(tree,line_no,hd)
ifhdisequaltoline_no,then
print(tree>data)
printVerticalLine(tree>left,line_no,hd1)
printVerticalLine(tree>right,line_no,hd+1)
Implementation:
#include <iostream>
// A node of binary tree

struct Node
{
int data;
};
// A utility function to create a new Binary Tree node

Node* newNode(int data)
{
temp->data = data;
return temp;
}
// A utility function to find min and max distances with respect

// to root.
void findMinMax(Node *node, int *min, int *max, int hd)
{
// Base case
if (node == NULL) return;
// Update min and max

if (hd < *min) *min = hd;
else if (hd > *max) *max = hd;

findMinMax(node->left, min, max, hd-1);
findMinMax(node->right, min, max, hd+1);
}
// A utility function to print all nodes on a given line_no.

// hd is horizontal distance of current node with respect to root.
void printVerticalLine(Node *node, int line_no, int hd)
{
// Base case
// If this node is on the given line number

if (hd == line_no)
cout << node->data << " ";

printVerticalLine(node->left, line_no, hd-1);
printVerticalLine(node->right, line_no, hd+1);
}
// The main function that prints a given binary tree in

// vertical order
void verticalOrder(Node *root)
{
// Find min and max distances with resepect to root
int min = 0, max = 0;
findMinMax(root, &min, &max, 0);
// Iterate through all possible vertical lines starting

// from the leftmost line and print nodes line by line
for (int line_no = min; line_no <= max; line_no++)
{
printVerticalLine(root, line_no, 0);
cout << endl;
}
}

int main()
{
// Create binary tree shown in above figure
cout << "Vertical order traversal is \n";

verticalOrder(root);
return 0;
}
Output:
Verticalordertraversalis
4
2
156
38
7
9
TimeComplexity:TimecomplexityofabovealgorithmisO(w*n)wherewiswidthofBinaryTreeandnis
numberofnodesinBinaryTree.Inworstcase,thevalueofwcanbeO(n)(consideracompletetreefor
example)andtimecomplexitycanbecomeO(n2).
=======================================================================
IntervalTree
Considerasituationwherewehaveasetofintervalsandweneedfollowingoperationstobeimplemented
efficiently.
1)Addaninterval
2)Removeaninterval
3)Givenanintervalx,findifxoverlapswithanyoftheexistingintervals.
IntervalTree:TheideaistoaugmentaselfbalancingBinarySearchTree(BST)likeRedBlackTree,AVL
Tree,etctomaintainsetofintervalssothatalloperationscanbedoneinO(Logn)time.
EverynodeofIntervalTreestoresfollowinginformation.
a)i:Anintervalwhichisrepresentedasapair[low,high]
b)max:Maximumhighvalueinsubtreerootedwiththisnode.
ThelowvalueofanintervalisusedaskeytomaintainorderinBST.Theinsertanddeleteoperationsare
sameasinsertanddeleteinselfbalancingBSTused.
Themainoperationistosearchforanoverlappinginterval.Followingisalgorithmforsearchingan
overlappingintervalxinanIntervaltreerootedwithroot.
IntervaloverlappingIntervalSearch(root,x)
1)Ifxoverlapswithroot'sinterval,returntheroot'sinterval.
2)Ifleftchildofrootisnotemptyandthemaxinleftchild
isgreaterthanx'slowvalue,recurforleftchild

3)Elserecurforrightchild.
Howdoestheabovealgorithmwork?
Lettheintervaltobesearchedbex.Weneedtoprovethisinforfollowingtwocases.
Case1:Whenwegotorightsubtree,oneofthefollowingmustbetrue.
a)Thereisanoverlapinrightsubtree:Thisisfineasweneedtoreturnoneoverlappinginterval.
b)Thereisnooverlapineithersubtree:WegotorightsubtreeonlywheneitherleftisNULLormaximum
valueinleftissmallerthanx.low.Sotheintervalcannotbepresentinleftsubtree.
Case2:Whenwegotoleftsubtree,oneofthefollowingmustbetrue.
a)Thereisanoverlapinleftsubtree:Thisisfineasweneedtoreturnoneoverlappinginterval.
b)Thereisnooverlapineithersubtree:Thisisthemostimportantpart.Weneedtoconsiderfollowingfacts.
Wewenttoleftsubtreebecausex.low<=maxinleftsubtree
.maxinleftsubtreeisahighofoneoftheintervalsletussay[a,max]inleftsubtree.
.Sincexdoesntoverlapwithanynodeinleftsubtreex.lowmustbesmallerthana.
.AllnodesinBSTareorderedbylowvalue,soallnodesinrightsubtreemusthavelowvaluegreaterthan
a.
.Fromabovetwofacts,wecansayallintervalsinrightsubtreehavelowvaluegreaterthanx.low.Sox
cannotoverlapwithanyintervalinrightsubtree.
ImplementationofIntervalTree:
FollowingisC++implementationofIntervalTree.TheimplementationusesbasicinsertoperationofBSTto
keepthingssimple.IdeallyitshouldbeinsertionofAVLTreeorinsertionofRedBlackTree.Deletionfrom
BSTisleftasanexercise.
#include <iostream>
// Structure to represent an interval

struct Interval
{
int low, high;
};
// Structure to represent a node in Interval Search Tree

struct ITNode
{
Interval *i; // 'i' could also be a normal variable
int max;
ITNode *left, *right;

};
// A utility function to create a new Interval Search Tree Node

ITNode * newNode(Interval i)
{
ITNode *temp = new ITNode;
temp->i = new Interval(i);
temp->max = i.high;
};
// A utility function to insert a new Interval Search Tree Node

// This is similar to BST Insert. Here the low value of interval
// is used tomaintain BST property
ITNode *insert(ITNode *root, Interval i)
{
// Base case: Tree is empty, new node becomes root
if (root == NULL)
return newNode(i);
// Get low value of interval at root

int l = root->i->low;
// If root's low value is smaller, then new interval goes to

// left subtree
if (i.low < l)
root->left = insert(root->left, i);
// Else, new node goes to right subtree.

else
root->right = insert(root->right, i);
// Update the max value of this ancestor if needed

if (root->max < i.high)
root->max = i.high;
return root;
}
// A utility function to check if given two intervals overlap

bool doOVerlap(Interval i1, Interval i2)
{
if (i1.low <= i2.high && i2.low <= i1.high)
return true;
return false;
}
// The main function that searches a given interval i in a given

// Interval Tree.
Interval *overlapSearch(ITNode *root, Interval i)
{
// Base Case, tree is empty
// If given interval overlaps with root

if (doOVerlap(*(root->i), i))
return root->i;
// If left child of root is present and max of left child is

// greater than or equal to given interval, then i may
// overlap with an interval is left subtree
if (root->left != NULL && root->left->max >= i.low)
return overlapSearch(root->left, i);
// Else interval can only overlap with right subtree

return overlapSearch(root->right, i);
}
void inorder(ITNode *root)

{
cout << "[" << root->i->low << ", " << root->i->high << "]"
<< " max = " << root->max << endl;
}

int main()
{
// Let us create interval tree shown in above figure
Interval ints[] = {{15, 20}, {10, 30}, {17, 19},
{5, 20}, {12, 15}, {30, 40}
};
int n = sizeof(ints)/sizeof(ints[0]);
ITNode *root = NULL;
for (int i = 0; i < n; i++)
root = insert(root, ints[i]);
cout << "Inorder traversal of constructed Interval Tree is\n";

inorder(root);
Interval x = {6, 7};
cout << "\nSearching for interval [" << x.low << "," << x.high << "]";
Interval *res = overlapSearch(root, x);
if (res == NULL)
cout << "\nNo Overlapping Interval";
else
cout << "\nOverlaps with [" << res->low << ", " << res->high << "]";
return 0;
}
Output:
InordertraversalofconstructedIntervalTreeis
[5,20]max=20
[10,30]max=30
[12,15]max=15
[15,20]max=40
[17,19]max=40
[30,40]max=40
Searchingforinterval[6,7]
Overlapswith[5,20]
ApplicationsofIntervalTree:
Intervaltreeismainlyageometricdatastructureandoftenusedforwindowingqueries,forinstance,tofind
allroadsonacomputerizedmapinsidearectangularviewport,ortofindallvisibleelementsinsidea
threedimensionalscene(SourceWiki).
IntervalTreevsSegmentTree
Bothsegmentandintervaltreesstoreintervals.Segmenttreeismainlyoptimizedforqueriesforagiven
point,andintervaltreesaremainlyoptimizedforoverlappingqueriesforagiveninterval.
=======================================================================
CheckifagivenBinaryTreeis
heightbalancedlikeaRedBlack
Tree
InaRedBlackTree,themaximumheightofanodeisatmosttwicetheminimumheight(Thefour
RedBlacktreepropertiesmakesurethisisalwaysfollowed).GivenaBinarySearchTree,weneedtocheck
forfollowingproperty.
Foreverynode,lengthofthelongestleaftonodepathhasnotmorethantwicethenodesonshortestpath
fromnodetoleaf.
1240
\/\
1410100
\/\
1660150
CannotbeaRedBlackTreeItcanbeRedBlackTree
withanycolorassignment
Maxheightof12is1
Minheightof12is3
10
/\
5100
/\
50150
/
40
ItcanalsobeRedBlackTree
ExpectedtimecomplexityisO(n).Thetreeshouldbetraversedatmostonceinthesolution.
Foreverynode,weneedtogetthemaximumandminimumheightsandcomparethem.Theideaisto
traversethetreeandforeverynodecheckifitsbalanced.Weneedtowritearecursivefunctionthatreturns
threethings,abooleanvaluetoindicatethetreeisbalancedornot,minimumheightandmaximumheight.
Toreturnmultiplevalues,wecaneitheruseastructureorpassvariablesbyreference.Wehavepassed
maxhandminhbyreferencesothatthevaluescanbeusedinparentcalls.
/* Program to check if a given Binary Tree is balanced like a Red-Black Tree */
#include <iostream>
struct Node
{
int key;
Node *left, *right;
};
/* utility that allocates a new Node with the given key */

{
Node* node = new Node;
node->key = key;
return (node);
}
// Returns returns tree if the Binary tree is balanced like a Red-Black

// tree. This function also sets value in maxh and minh (passed by
// reference). maxh and minh are set as maximum and minimum heights of root.
bool isBalancedUtil(Node *root, int &maxh, int &minh)
{
// Base case
if (root == NULL)
{
maxh = minh = 0;
return true;
}
int lmxh, lmnh; // To store max and min heights of left subtree
int rmxh, rmnh; // To store max and min heights of right subtree
// Check if left subtree is balanced, also set lmxh and lmnh

if (isBalancedUtil(root->left, lmxh, lmnh) == false)
return false;
// Check if right subtree is balanced, also set rmxh and rmnh

if (isBalancedUtil(root->right, rmxh, rmnh) == false)
return false;
// Set the max and min heights of this node for the parent call
maxh = max(lmxh, rmxh) + 1;
minh = min(lmnh, rmnh) + 1;
// See if this node is balanced

if (maxh <= 2*minh)
return true;
return false;
}
// A wrapper over isBalancedUtil()

bool isBalanced(Node *root)
{
int maxh, minh;

return isBalancedUtil(root, maxh, minh);
}

int main()
{
Node * root = newNode(10);
isBalanced(root)? cout << "Balanced" : cout << "Not Balanced";
return 0;
}
Output:
Balanced
TimeComplexity:TimeComplexityofabovecodeisO(n)asthecodedoesasimpletreetraversal.
=======================================================================
Printallnodesthatareatdistancek
fromaleafnode
GivenaBinaryTreeandapositiveintegerk,printallnodesthataredistancekfromaleafnode.
Herethemeaningofdistanceisdifferentfrompreviouspost.Herekdistancefromaleafmeansklevels
higherthanaleafnode.ForexampleifkismorethanheightofBinaryTree,thennothingshouldbeprinted.
ExpectedtimecomplexityisO(n)wherenisthenumbernodesinthegivenBinaryTree.

Theideaistotraversethetree.Keepstoringallancestorstillwehitaleafnode.Whenwereachaleafnode,
weprinttheancestoratdistancek.Wealsoneedtokeeptrackofnodesthatarealreadyprintedasoutput.
Forthatweuseabooleanarrayvisited[].
/* Program to print all nodes which are at distance k from a leaf */
#include <iostream>
#define MAX_HEIGHT 10000
struct Node
{
int key;
Node *left, *right;
};
/* utility that allocates a new Node with the given key */

{
Node* node = new Node;
node->key = key;
return (node);
}
/* This function prints all nodes that are distance k from a leaf node
path[] --> Store ancestors of a node
visited[] --> Stores true if a node is printed as output. A node may be k
distance away from many leaves, we want to print it once */
void kDistantFromLeafUtil(Node* node, int path[], bool visited[],
int pathLen, int k)
{
// Base case
if (node==NULL) return;
/* append this Node to the path array */

path[pathLen] = node->key;
visited[pathLen] = false;
pathLen++;
/* it's a leaf, so print the ancestor at distance k only

if the ancestor is not already printed */
if (node->left == NULL && node->right == NULL &&
pathLen-k-1 >= 0 && visited[pathLen-k-1] == false)
{
cout << path[pathLen-k-1] << " ";
visited[pathLen-k-1] = true;
return;
}
/* If not leaf node, recur for left and right subtrees */

kDistantFromLeafUtil(node->left, path, visited, pathLen, k);
kDistantFromLeafUtil(node->right, path, visited, pathLen, k);
}
/* Given a binary tree and a nuber k, print all nodes that are k
distant from a leaf*/
void printKDistantfromLeaf(Node* node, int k)
{
int path[MAX_HEIGHT];
bool visited[MAX_HEIGHT] = {false};
kDistantFromLeafUtil(node, path, visited, 0, k);

}

int main()
{
// Let us create binary tree given in the above example
cout << "Nodes at distance 2 are: ";

printKDistantfromLeaf(root, 2);
return 0;
}
Output:
Nodesatdistance2are:31
TimeComplexity:TimeComplexityofabovecodeisO(n)asthecodedoesasimpletreetraversal.
======================================================================
Finddistancebetweentwogivenkeys
ofaBinaryTree
Findthedistancebetweentwokeysinabinarytree,noparentpointersaregiven.Distancebetweentwo
nodesistheminimumnumberofedgestobetraversedtoreachonenodefromother.

Thedistancebetweentwonodescanbeobtainedintermsoflowestcommonancestor.Followingisthe
formula.
Dist(n1,n2)=Dist(root,n1)+Dist(root,n2)2*Dist(root,lca)
'n1'and'n2'arethetwogivenkeys
'root'isrootofgivenBinaryTree.
'lca'islowestcommonancestorofn1andn2
Dist(n1,n2)isthedistancebetweenn1andn2.
FollowingisC++implementationofaboveapproach.Theimplementationisadoptedfromlastcodeprovided
inLowestCommonAncestorPost.
/* Program to find distance between n1 and n2 using one traversal */
#include <iostream>

struct Node
{
int key;
};
// Utility function to create a new tree Node

{
temp->key = key;
return temp;
}
// Returns level of key k if it is present in tree, otherwise returns -1

int findLevel(Node *root, int k, int level)
{
// Base Case
if (root == NULL)
return -1;
// If key is present at root, or in left subtree or right subtree,

// return true;
if (root->key == k)
return level;
int l = findLevel(root->left, k, level+1);

return (l != -1)? l : findLevel(root->right, k, level+1);
}
// This function returns pointer to LCA of two given values n1 and n2.
// It also sets d1, d2 and dist if one key is not ancestor of other
// d1 --> To store distance of n1 from root
// d2 --> To store distance of n2 from root
// lvl --> Level (or distance from root) of current node
// dist --> To store distance between n1 and n2
Node *findDistUtil(Node* root, int n1, int n2, int &d1, int &d2,
int &dist, int lvl)
{
// Base case
// If either n1 or n2 matches with root's key, report

// the presence by returning root (Note that if a key is
// ancestor of other, then the ancestor key becomes LCA
if (root->key == n1)
{
d1 = lvl;
return root;
}
{
d2 = lvl;
return root;
}
// Look for n1 and n2 in left and right subtrees

Node *left_lca = findDistUtil(root->left, n1, n2, d1, d2, dist, lvl+1);
Node *right_lca = findDistUtil(root->right, n1, n2, d1, d2, dist, lvl+1);
// If both of the above calls return Non-NULL, then one key

// is present in once subtree and other is present in other,
// So this node is the LCA
if (left_lca && right_lca)
{
dist = d1 + d2 - 2*lvl;
return root;
}
// Otherwise check if left subtree or right subtree is LCA

return (left_lca != NULL)? left_lca: right_lca;
}
// The main function that returns distance between n1 and n2

// This function returns -1 if either n1 or n2 is not present in
// Binary Tree.
int findDistance(Node *root, int n1, int n2)
{
// Initialize d1 (distance of n1 from root), d2 (distance of n2
// from root) and dist(distance between n1 and n2)

int d1 = -1, d2 = -1, dist;
Node *lca = findDistUtil(root, n1, n2, d1, d2, dist, 1);
// If both n1 and n2 were present in Binary Tree, return dist

if (d1 != -1 && d2 != -1)
return dist;
// If n1 is ancestor of n2, consider n1 as root and find level

// of n2 in subtree rooted with n1
if (d1 != -1)
{
dist = findLevel(lca, n2, 0);
return dist;
}
// If n2 is ancestor of n1, consider n2 as root and find level

// of n1 in subtree rooted with n2
if (d2 != -1)
{
dist = findLevel(lca, n1, 0);
return dist;
}
return -1;
}

int main()
{
cout << "Dist(4, 5) = " << findDistance(root, 4, 5);
cout << "\nDist(4, 6) = " << findDistance(root, 4, 6);
return 0;
}
Output:
Dist(4,5)=2
Dist(4,6)=4
Dist(3,4)=3
Dist(2,4)=1
Dist(8,5)=5
TimeComplexity:TimecomplexityoftheabovesolutionisO(n)asthemethoddoesasingletreetraversal.
=======================================================================
LowestCommonAncestorina
BinaryTree|Set1
Givenabinarytree(notabinarysearchtree)andtwovaluessayn1andn2,writeaprogramtofindtheleast
commonancestor.
FollowingisdefinitionofLCAfromWikipedia:
LetTbearootedtree.Thelowestcommonancestorbetweentwonodesn1andn2isdefinedasthelowest
nodeinTthathasbothn1andn2asdescendants(whereweallowanodetobeadescendantofitself).
TheLCAofn1andn2inTisthesharedancestorofn1andn2thatislocatedfarthestfromtheroot.
Computationoflowestcommonancestorsmaybeuseful,forinstance,aspartofaprocedurefor
determiningthedistancebetweenpairsofnodesinatree:thedistancefromn1ton2canbecomputedas
thedistancefromtherootton1,plusthedistancefromtherootton2,minustwicethedistancefromtheroot
totheirlowestcommonancestor.(SourceWiki)

WehavediscussedanefficientsolutiontofindLCAinBinarySearchTree.InBinarySearchTree,using
BSTproperties,wecanfindLCAinO(h)timewherehisheightoftree.Suchanimplementationisnot
possibleinBinaryTreeaskeysBinaryTreenodesdontfollowanyorder.Followingaredifferentapproaches
tofindLCAinBinaryTree.
Method1(ByStoringrootton1androotton2paths):
FollowingissimpleO(n)algorithmtofindLCAofn1andn2.
1)Findpathfromrootton1andstoreitinavectororarray.
2)Findpathfromrootton2andstoreitinanothervectororarray.
3)Traversebothpathstillthevaluesinarraysaresame.Returnthecommonelementjustbeforethe
mismatch.
// A O(n) solution to find LCA of two given values n1 and n2
#include <iostream>
#include <vector>
// A Bianry Tree node

struct Node
{
int key;
};
// Utility function creates a new binary tree node with given key
Node * newNode(int k)
{

temp->key = k;
return temp;
}
// Finds the path from root node to given root of the tree, Stores the
// path in a vector path[], returns true if path exists otherwise false
bool findPath(Node *root, vector<int> &path, int k)
{
// base case
if (root == NULL) return false;
// Store this node in path vector. The node will be removed if

// not in path from root to k
path.push_back(root->key);
// See if the k is same as root's key

if (root->key == k)
return true;
// Check if k is found in left or right sub-tree

if ( (root->left && findPath(root->left, path, k)) ||
(root->right && findPath(root->right, path, k)) )
return true;
// If not present in subtree rooted with root, remove root from

// path[] and return false
path.pop_back();
return false;
}
// Returns LCA if node n1, n2 are present in the given binary tree,
// otherwise return -1
int findLCA(Node *root, int n1, int n2)
{
// to store paths to n1 and n2 from the root
vector<int> path1, path2;
// Find paths from root to n1 and root to n1. If either n1 or n2

// is not present, return -1
if ( !findPath(root, path1, n1) || !findPath(root, path2, n2))
return -1;
/* Compare the paths to get the first different value */

int i;
for (i = 0; i < path1.size() && i < path2.size() ; i++)
if (path1[i] != path2[i])
break;
return path1[i-1];
}

int main()
{
// Let us create the Binary Tree shown in above diagram.
cout << "LCA(4, 5) = " << findLCA(root, 4, 5);
cout << "\nLCA(4, 6) = " << findLCA(root, 4, 6);
return 0;
}
Output:
LCA(4,5)=2
LCA(4,6)=1
LCA(3,4)=1
LCA(2,4)=2
TimeComplexity:TimecomplexityoftheabovesolutionisO(n).Thetreeistraversedtwice,andthenpath
arraysarecompared.
ThankstoRaviChandraEnagantiforsuggestingtheinitialsolutionbasedonthismethod.
Method2(UsingSingleTraversal)
Themethod1findsLCAinO(n)time,butrequiresthreetreetraversalsplusextraspacesforpatharrays.If
weassumethatthekeysn1andn2arepresentinBinaryTree,wecanfindLCAusingsingletraversalof
BinaryTreeandwithoutextrastorageforpatharrays.
Theideaistotraversethetreestartingfromroot.Ifanyofthegivenkeys(n1andn2)matcheswithroot,
thenrootisLCA(assumingthatbothkeysarepresent).Ifrootdoesntmatchwithanyofthekeys,werecur
forleftandrightsubtree.Thenodewhichhasonekeypresentinitsleftsubtreeandtheotherkeypresentin
rightsubtreeistheLCA.Ifbothkeyslieinleftsubtree,thenleftsubtreehasLCAalso,otherwiseLCAliesin
rightsubtree.
/* Program to find LCA of n1 and n2 using one traversal of Binary Tree */
#include <iostream>

struct Node
{
int key;
};

{
temp->key = key;
return temp;
}
// This function assumes that n1 and n2 are present in Binary Tree
struct Node *findLCA(struct Node* root, int n1, int n2)
{
// Base case
// If either n1 or n2 matches with root's key, report

// the presence by returning root (Note that if a key is
// ancestor of other, then the ancestor key becomes LCA
if (root->key == n1 || root->key == n2)
return root;
// Look for keys in left and right subtrees

Node *left_lca = findLCA(root->left, n1, n2);
Node *right_lca = findLCA(root->right, n1, n2);

if (left_lca && right_lca) return root;

}

int main()
{
cout << "LCA(4, 5) = " << findLCA(root, 4, 5)->key;
cout << "\nLCA(4, 6) = " << findLCA(root, 4, 6)->key;

return 0;
}
Output:
LCA(4,5)=2
LCA(4,6)=1
LCA(3,4)=1
LCA(2,4)=2
ThankstoAtulSinghforsuggestingthissolution.
TimeComplexity:TimecomplexityoftheabovesolutionisO(n)asthemethoddoesasimpletreetraversal
inbottomupfashion.
NotethattheabovemethodassumesthatkeysarepresentinBinaryTree.Ifonekeyispresentandotheris
absent,thenitreturnsthepresentkeyasLCA(IdeallyshouldhavereturnedNULL).
Wecanextendthismethodtohandleallcasesbypassingtwobooleanvariablesv1andv2.v1issetastrue
whenn1ispresentintreeandv2issetastrueifn2ispresentintree.
/* Program to find LCA of n1 and n2 using one traversal of Binary Tree.
It handles all cases even when n1 or n2 is not there in Binary Tree */
#include <iostream>

struct Node
{
int key;
};

{
temp->key = key;
return temp;
}
// v1 is set as true by this function if n1 is found

// v2 is set as true by this function if n2 is found
struct Node *findLCAUtil(struct Node* root, int n1, int n2, bool &v1, bool &v2)
{
// Base case
// If either n1 or n2 matches with root's key, report the presence

// by setting v1 or v2 as true and return root (Note that if a key
// is ancestor of other, then the ancestor key becomes LCA)
{
v1 = true;
return root;
}
{
v2 = true;
return root;
}
// Look for keys in left and right subtrees

Node *left_lca = findLCAUtil(root->left, n1, n2, v1, v2);
Node *right_lca = findLCAUtil(root->right, n1, n2, v1, v2);

if (left_lca && right_lca) return root;

}
// Returns true if key k is present in tree rooted with root

bool find(Node *root, int k)
{
// Base Case
if (root == NULL)
return false;
// If key is present at root, or in left subtree or right subtree,

// return true;
if (root->key == k || find(root->left, k) || find(root->right, k))
return true;
// Else return false

return false;
}
// This function returns LCA of n1 and n2 only if both n1 and n2 are present
// in tree, otherwise returns NULL;
Node *findLCA(Node *root, int n1, int n2)
{
// Initialize n1 and n2 as not visited
bool v1 = false, v2 = false;
// Find lca of n1 and n2 using the technique discussed above

Node *lca = findLCAUtil(root, n1, n2, v1, v2);
// Return LCA only if both n1 and n2 are present in tree

if (v1 && v2 || v1 && find(lca, n2) || v2 && find(lca, n1))
return lca;
// Else return NULL

return NULL;
}

int main()
{
Node *lca = findLCA(root, 4, 5);
if (lca != NULL)
cout << "LCA(4, 5) = " << lca->key;
else
cout << "Keys are not present ";
lca = findLCA(root, 4, 10);

if (lca != NULL)
cout << "\nLCA(4, 10) = " << lca->key;
else
cout << "\nKeys are not present ";
return 0;
}
Output:
LCA(4,5)=2
Keysarenotpresent
=======================================================================
Printallnodesthatdonthave
sibling
GivenaBinaryTree,printallnodesthatdonthaveasibling(asiblingisanodethathassameparent.Ina
BinaryTree,therecanbeatmostonesibling).Rootshouldnotbeprintedasrootcannothaveasibling.
Forexample,theoutputshouldbe456forthefollowingtree.

Thisisatypicaltreetraversalquestion.Westartfromrootandcheckifthenodehasonechild,ifyesthen
printtheonlychildofthatnode.Ifnodehasbothchildren,thenrecurforboththechildren.
/* Program to find singles in a given binary tree */
#include <iostream>

struct node
{
int key;
};
// Utility function to create a new tree node

node* newNode(int key)
{
temp->key = key;
return temp;
// Function to print all non-root nodes that don't have a sibling

void printSingles(struct node *root)
{
// Base case
if (root == NULL)
return;
// If this is an internal node, recur for left

// and right subtrees
if (root->left != NULL && root->right != NULL)
{
printSingles(root->left);
printSingles(root->right);
}
// If left child is NULL and right is not, print right child

// and recur for right child
else if (root->right != NULL)
{
cout << root->right->key << " ";
printSingles(root->right);
}
// If right child is NULL and left is not, print left child

// and recur for left child
else if (root->left != NULL)
{
cout << root->left->key << " ";
printSingles(root->left);
}
}

int main()
{

node *root = newNode(1);
printSingles(root);
return 0;
}
Output:
456
TimeComplexityofabovecodeisO(n)asthecodedoesasimpletreetraversal.
=======================================================================
ConvertagivenBinaryTreeto
DoublyLinkedList|Set3
GivenaBinaryTree(BT),convertittoaDoublyLinkedList(DLL)InPlace.Theleftandrightpointersin
nodesaretobeusedaspreviousandnextpointersrespectivelyinconvertedDLL.Theorderofnodesin
DLLmustbesameasInorderofthegivenBinaryTree.ThefirstnodeofInordertraversal(leftmostnodein
BT)mustbeheadnodeoftheDLL.

Followingtwodifferentsolutionshavebeendiscussedforthisproblem.
ConvertagivenBinaryTreetoDoublyLinkedList|Set1
ConvertagivenBinaryTreetoDoublyLinkedList|Set2
Inthispost,athirdsolutionisdiscussedwhichseemstobethesimplestofall.Theideaistodoinorder
traversalofthebinarytree.Whiledoinginordertraversal,keeptrackofthepreviouslyvisitednodeina
variablesayprev.Foreveryvisitednode,makeitnextofprevandpreviousofthisnodeasprev.
Thankstorahul,wishallandallotherreadersfortheirusefulcommentsontheabovetwoposts.
FollowingisC++implementationofthissolution.
// A C++ program for in-place conversion of Binary Tree to DLL
#include <iostream>
/* A binary tree node has data, and left and right pointers */
struct node
{
int data;
node* left;
node* right;
};
// A simple recursive function to convert a given Binary tree to Doubly

// Linked List
// root --> Root of Binary Tree

// head --> Pointer to head node of created doubly linked list
void BinaryTree2DoubleLinkedList(node *root, node **head)
{
// Base case
// Initialize previously visited node as NULL. This is

// static so that the same value is accessible in all recursive
// calls
static node* prev = NULL;
// Recursively convert left subtree

BinaryTree2DoubleLinkedList(root->left, head);
// Now convert this node

if (prev == NULL)
*head = root;
else
{
root->left = prev;
prev->right = root;
}
prev = root;
// Finally convert right subtree

BinaryTree2DoubleLinkedList(root->right, head);
}

node* newNode(int data)
{
node* new_node = new node;
new_node->data = data;
new_node->left = new_node->right = NULL;
return (new_node);
/* Function to print nodes in a given doubly linked list */

void printList(node *node)
{
while (node!=NULL)
{
cout << node->data << " ";
node = node->right;
}
}

int main()
{
// Let us create the tree shown in above diagram
node *root
= newNode(10);
root->left
= newNode(12);
root->right
= newNode(15);
// Convert to DLL
node *head = NULL;
BinaryTree2DoubleLinkedList(root, &head);
// Print the converted list

printList(head);
return 0;
}
Output:
251230103615
Notethatuseofstaticvariableslikeaboveisnotarecommendedpractice(wehaveusedstaticfor
simplicity).Imagineasituationwheresamefunctioniscalledfortwoormoretrees,theoldvalueofprev
wouldbeusedinnextcallforadifferenttree.Toavoidsuchproblems,wecanusedoublepointeror
referencetoapointer.
TimeComplexity:Theaboveprogramdoesasimpleinordertraversal,sotimecomplexityisO(n)wherenis
thenumberofnodesingivenbinarytree.
=======================================================================
RedBlackTree|Set2(Insert)
Inthepreviouspost,wediscussedintroductiontoRedBlackTrees.Inthispost,insertionisdiscussed.
InAVLtreeinsertion,weusedrotationasatooltodobalancingafterinsertioncausedimbalance.In
RedBlacktree,weusetwotoolstodobalancing.
1)Recoloring
2)Rotation
Wetryrecoloringfirst,ifrecoloringdoesntwork,thenwegoforrotation.Followingisdetailedalgorithm.The
algorithmshasmainlytwocasesdependinguponthecolorofuncle.Ifuncleisred,wedorecoloring.Ifuncle
isblack,wedorotationsand/orrecoloring.
ColorofaNULLnodeisconsideredasBLACK.
Letxbethenewlyinsertednode.
1)PerformstandardBSTinsertionandmakethecolorofnewlyinsertednodesasRED.
2)DofollowingifcolorofxsparentisnotBLACKorxisnotroot.
.a)IfxsuncleisRED(Grandparentmusthavebeenblackfromproperty4)
..(i)ChangecolorofparentanduncleasBLACK.
..(ii)colorofgrandparentasRED.
..(iii)Changex=xsgrandparent,repeatsteps2and3fornewx.
.b)IfxsuncleisBLACK,thentherecanbefourconfigurationsforx,xsparent(p)andxsgrandparent
(g)(ThisissimilartoAVLTree)
..i)LeftLeftCase(pisleftchildofgandxisleftchildofp)
..ii)LeftRightCase(pisleftchildofgandxisrightchildofp)
..iii)RightRightCase(Mirrorofcasea)
..iv)RightLeftCase(Mirrorofcasec)
3)Ifxisroot,changecolorofxasBLACK(Blackheightofcompletetreeincreasesby1).
FollowingareoperationstobeperformedinfoursubcaseswhenuncleisBLACK.
AllfourcaseswhenUncleisBLACK
LeftLeftCase(Seeg,pandx)
LeftRightCase(Seeg,pandx)
RightRightCase(Seeg,pandx)

RightLeftCase(Seeg,pandx)
ExamplesofInsertion
PleasereferCProgramforRedBlackTreeInsertionforcompleteimplementationofabovealgorithm.
=======================================================================
RedBlackTree|Set1(Introduction)
RedBlackTreeisaselfbalancingBinarySearchTree(BST)whereeverynodefollowsfollowingrules.

1)Everynodehasacoloreitherredorblack.
2)Rootoftreeisalwaysblack.
3)Therearenotwoadjacentrednodes(Arednodecannothavearedparentorredchild).
4)EverypathfromroottoaNULLnodehassamenumberofblacknodes.
WhyRedBlackTrees?
MostoftheBSToperations(e.g.,search,max,min,insert,delete..etc)takeO(h)timewherehistheheight
oftheBST.ThecostoftheseoperationsmaybecomeO(n)foraskewedBinarytree.Ifwemakesurethat
heightofthetreeremainsO(Logn)aftereveryinsertionanddeletion,thenwecanguaranteeanupper
boundofO(Logn)foralltheseoperations.TheheightofaRedBlacktreeisalwaysO(Logn)wherenisthe
numberofnodesinthetree.
ComparisonwithAVLTree
TheAVLtreesaremorebalancedcomparedtoRedBlackTrees,buttheymaycausemorerotationsduring
insertionanddeletion.Soifyourapplicationinvolvesmanyfrequentinsertionsanddeletions,thenRed
Blacktreesshouldbepreferred.Andiftheinsertionsanddeletionsarelessfrequentandsearchismore
frequentoperation,thenAVLtreeshouldbepreferredoverRedBlackTree.
HowdoesaRedBlackTreeensurebalance?
Asimpleexampletounderstandbalancingis,achainof3nodesisnotpossibleinredblacktree.Wecantry
anycombinationofcolorsandseeallofthemviolateRedBlacktreeproperty.
Achainof3nodesisnodesisnotpossibleinRedBlackTrees.
FollowingareNOTRedBlackTrees
303030
/\/\/\
20NIL20NIL20NIL
/\/\/\
10NIL10NIL10NIL
ViolatesViolatesViolates
Property4.Property4Property3
FollowingaredifferentpossibleRedBlackTreeswithabove3keys
2020
/\/\
10301030
/\/\/\/\
NILNILNILNILNILNILNILNIL
Fromtheaboveexamples,wegetsomeideahowRedBlacktreesensurebalance.Followingisan
importantfactaboutbalancinginRedBlackTrees.
EveryRedBlackTreewithnnodeshasheight<=
Thiscanbeprovedusingfollowingfacts:
1)ForageneralBinaryTree,letkbetheminimumnumberofnodesonallroottoNULLpaths,thenn>=2k
1(Ex.Ifkis3,thennisatleast7).Thisexpressioncanalsobewrittenask<=
2)Fromproperty4ofRedBlacktreesandaboveclaim,wecansayinaRedBlackTreewithnnodes,there
isaroottoleafpathwithatmost
blacknodes.
3)Fromproperty3ofRedBlacktrees,wecanclaimthatthenumberblacknodesinaRedBlacktreeisat
least
wherenistotalnumberofnodes.
Fromabove2points,wecanconcludethefactthatRedBlackTreewithnnodeshasheight<=
Inthispost,weintroducedRedBlacktreesanddiscussedhowbalanceisensured.Thehardpartisto
maintainbalancewhenkeysareaddedandremoved.Wewillsoonbediscussinginsertionanddeletion
operationsincomingpostsonRedBlacktree.
Exercise:
1)IsitpossibletohaveallblacknodesinaRedBlacktree?
2)DrawaRedBlackTreethatisnotanAVLtreestructurewise?
InsertionandDeletion
RedBlackTreeInsertion
RedBlackTreeDeletion
References:
IntroductiontoAlgorithms3rdEditionbyCliffordStein,ThomasH.Cormen,CharlesE.Leiserson,RonaldL.
Rivest
http://en.wikipedia.org/wiki/Red%E2%80%93black_tree
VideoLectureonRedBlackTreebyTimRoughgarden
MITVideoLectureonRedBlackTree
MITLectureNotesonRedBlackTree
=======================================================================
Sumofallthenumbersthatare
formedfromroottoleafpaths
Givenabinarytree,whereeverynodevalueisaDigitfrom19.Findthesumofallthenumberswhichare
formedfromroottoleafpaths.
ForexampleconsiderthefollowingBinaryTree.
6
/\
35
/\\
254
/\
74
Thereare4leaves,hence4roottoleafpaths:
PathNumber
6>3>2632
6>3>5>76357
6>3>5>46354
6>5>4654
Answer=632+6357+6354+654=13997
Westronglyrecommendyoutominimizethebrowserandtrythisyourselffirst.
Theideaistodoapreordertraversalofthetree.Inthepreordertraversal,keeptrackofthevaluecalculated
tillthecurrentnode,letthisvaluebeval.Foreverynode,weupdatethevalasval*10plusnodesdata.
// C program to find sum of all paths from root to leaves
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
};
// function to allocate new node with given data

struct node* newNode(int data)
{
struct node* node = (struct node*)malloc(sizeof(struct node));
node->data = data;
return (node);
}
// Returns sum of all root to leaf paths. The first parameter is root
// of current subtree, the second parameter is value of the number formed
// by nodes from root to this node
int treePathsSumUtil(struct node *root, int val)
{
// Base case
if (root == NULL) return 0;
// Update val
val = (val*10 + root->data);
// if current node is leaf, return the current value of val

if (root->left==NULL && root->right==NULL)
return val;
// recur sum of values for left and right subtree

return treePathsSumUtil(root->left, val) +
treePathsSumUtil(root->right, val);
}
// A wrapper function over treePathsSumUtil()

int treePathsSum(struct node *root)
{
// Pass the initial value as 0 as there is nothing above root

return treePathsSumUtil(root, 0);
}
// Driver function to test the above functions

int main()
{
struct node *root = newNode(6);
root->left
= newNode(3);
root->right
= newNode(5);
root->right->right= newNode(7);
printf("Sum of all paths is %d", treePathsSum(root));
return 0;
}
Output:
Sumofallpathsis13997
TimeComplexity:Theabovecodeisasimplepreordertraversalcodewhichvisitseveryexactlyonce.
Therefore,thetimecomplexityisO(n)wherenisthenumberofnodesinthegivenbinarytree.
=======================================================================
SplayTree|Set2(Insert)
Itisrecommendedtoreferfollowingpostasprerequisiteofthispost.
SplayTree|Set1(Search)
Asdiscussedinthepreviouspost,Splaytreeisaselfbalancingdatastructurewherethelastaccessedkey
isalwaysatroot.TheinsertoperationissimilartoBinarySearchTreeinsertwithadditionalstepstomake
surethatthenewlyinsertedkeybecomesthenewroot.
Followingaredifferentcasestoinsertakeykinsplaytree.
1)RootisNULL:Wesimplyallocateanewnodeandreturnitasroot.
2)Splaythegivenkeyk.Ifkisalreadypresent,thenitbecomesthenewroot.Ifnotpresent,thenlast
accessedleafnodebecomesthenewroot.
3)Ifnewrootskeyissameask,dontdoanythingaskisalreadypresent.
4)Elseallocatememoryfornewnodeandcomparerootskeywithk.
.4.a)Ifkissmallerthanrootskey,makerootasrightchildofnewnode,copyleftchildofrootasleft
childofnewnodeandmakeleftchildofrootasNULL.
.4.b)Ifkisgreaterthanrootskey,makerootasleftchildofnewnode,copyrightchildofrootasright
childofnewnodeandmakerightchildofrootasNULL.
5)Returnnewnodeasnewrootoftree.
Example:
100[20]25
/\\/\
50200502050
/insert(25)/\insert(25)/\
40======>30100========>30100
/1.Splay(25)\\2.insert25\\
304020040200
/
[20]
// This code is adopted from
http://algs4.cs.princeton.edu/33balanced/SplayBST.java.html
#include<stdio.h>
#include<stdlib.h>
// An AVL tree node

struct node
{
int key;
};
/* Helper function that allocates a new node with the given key and
NULL left and right pointers. */
struct node* newNode(int key)
{
node->key
= key;

return (node);
}
// A utility function to right rotate subtree rooted with y

// See the diagram given above.
struct node *rightRotate(struct node *x)
{
struct node *y = x->left;
x->left = y->right;
y->right = x;
return y;
}
// A utility function to left rotate subtree rooted with x

struct node *leftRotate(struct node *x)
{
struct node *y = x->right;
x->right = y->left;
y->left = x;
return y;
}
// This function brings the key at root if key is present in tree.

// If key is not present, then it brings the last accessed item at
// root. This function modifies the tree and returns the new root
struct node *splay(struct node *root, int key)
{
// Base cases: root is NULL or key is present at root
if (root == NULL || root->key == key)
return root;
// Key lies in left subtree

{
// Key is not in tree, we are done

if (root->left == NULL) return root;
// Zig-Zig (Left Left)

if (root->left->key > key)
{
// First recursively bring the key as root of left-left
root->left->left = splay(root->left->left, key);
// Do first rotation for root, second rotation is done after else

root = rightRotate(root);
}
else if (root->left->key < key) // Zig-Zag (Left Right)
{
// First recursively bring the key as root of left-right
root->left->right = splay(root->left->right, key);
// Do first rotation for root->left

if (root->left->right != NULL)
root->left = leftRotate(root->left);
}
// Do second rotation for root

return (root->left == NULL)? root: rightRotate(root);
}
else // Key lies in right subtree
{
if (root->right == NULL) return root;
// Zig-Zag (Right Left)

if (root->right->key > key)
{
// Bring the key as root of right-left
root->right->left = splay(root->right->left, key);
// Do first rotation for root->right
if (root->right->left != NULL)
root->right = rightRotate(root->right);
}
else if (root->right->key < key)// Zag-Zag (Right Right)
{
// Bring the key as root of right-right and do first rotation
root->right->right = splay(root->right->right, key);
root = leftRotate(root);
}

return (root->right == NULL)? root: leftRotate(root);
}
}
// Function to insert a new key k in splay tree with given root

struct node *insert(struct node *root, int k)
{
// Simple Case: If tree is empty
if (root == NULL) return newNode(k);
// Bring the closest leaf node to root

root = splay(root, k);
// If key is already present, then return

if (root->key == k) return root;
// Otherwise allocate memory for new node

struct node *newnode = newNode(k);
// If root's key is greater, make root as right child

// of newnode and copy the left child of root to newnode
if (root->key > k)
{
newnode->right = root;
newnode->left = root->left;
root->left = NULL;
// If root's key is smaller, make root as left child

// of newnode and copy the right child of root to newnode
else
{
newnode->left = root;
newnode->right = root->right;
root->right = NULL;
}
return newnode; // newnode becomes new root

}
// A utility function to print preorder traversal of the tree.

// The function also prints height of every node
void preOrder(struct node *root)
{
if (root != NULL)
{
preOrder(root->left);
preOrder(root->right);
}
}
/* Drier program to test above function*/

int main()
{
printf("Preorder traversal of the modified Splay tree is \n");
preOrder(root);
return 0;
}
Output:
PreordertraversalofthemodifiedSplaytreeis
2520503040100200
=======================================================================
SplayTree|Set1(Search)
TheworstcasetimecomplexityofBinarySearchTree(BST)operationslikesearch,delete,insertisO(n).
Theworstcaseoccurswhenthetreeisskewed.WecangettheworstcasetimecomplexityasO(Logn)with
AVLandRedBlackTrees.
CanwedobetterthanAVLorRedBlacktreesinpracticalsituations?
LikeAVLandRedBlackTrees,SplaytreeisalsoselfbalancingBST.Themainideaofsplaytreeistobring
therecentlyaccesseditemtorootofthetree,thismakestherecentlysearcheditemtobeaccessibleinO(1)
timeifaccessedagain.Theideaistouselocalityofreference(Inatypicalapplication,80%oftheaccess
areto20%oftheitems).Imagineasituationwherewehavemillionsorbillionsofkeysandonlyfewofthem
areaccessedfrequently,whichisverylikelyinmanypracticalapplications.
AllsplaytreeoperationsruninO(logn)timeonaverage,wherenisthenumberofentriesinthetree.Any
singleoperationcantakeTheta(n)timeintheworstcase.
SearchOperation
ThesearchoperationinSplaytreedoesthestandardBSTsearch,inadditiontosearch,italsosplays(move
anodetotheroot).Ifthesearchissuccessful,thenthenodethatisfoundissplayedandbecomesthenew
root.ElsethelastnodeaccessedpriortoreachingtheNULLissplayedandbecomesthenewroot.
Therearefollowingcasesforthenodebeingaccessed.
1)NodeisrootWesimplyreturntheroot,dontdoanythingelseastheaccessednodeisalreadyroot.
2)Zig:Nodeischildofroot(thenodehasnograndparent).Nodeiseitheraleftchildofroot(wedoaright
rotation)ornodeisarightchildofitsparent(wedoaleftrotation).
T1,T2andT3aresubtreesofthetreerootedwithy(onleftside)orx(onrightside)
yx
/\Zig(RightRotation)/\
xT3>T1y
/\</\
T1T2Zag(LeftRotation)T2T3
3)Nodehasbothparentandgrandparent.Therecanbefollowingsubcases.
........3.a)ZigZigandZagZagNodeisleftchildofparentandparentisalsoleftchildofgrandparent(Two
rightrotations)ORnodeisrightchildofitsparentandparentisalsorightchildofgrandparent(TwoLeft
Rotations).
ZigZig(LeftLeftCase):
GPX
/\/\/\
PT4rightRotate(G)XGrightRotate(P)T1P
/\============>/\/\============>/\
XT3T1T2T3T4T2G
/\/\
T1T2T3T4
ZagZag(RightRightCase):
GPX
/\/\/\
T1PleftRotate(G)GXleftRotate(P)PT4
/\============>/\/\============>/\
T2XT1T2T3T4GT3
/\/\
T3T4T1T2
........3.b)ZigZagandZagZigNodeisleftchildofparentandparentisrightchildofgrandparent(Left
Rotationfollowedbyrightrotation)ORnodeisrightchildofitsparentandparentisleftchildofgrandparent
(RightRotationfollowedbyleftrotation).
ZigZag(LeftRightCase):
GGX
/\/\/\
PT4leftRotate(P)XT4rightRotate(G)PG
/\============>/\============>/\/\
T1XPT3T1T2T3T4
/\/\
T2T3T1T2
ZagZig(RightLeftCase):
GGX
/\/\/\
T1PrightRotate(P)T1XleftRotate(P)GP
/\=============>/\============>/\/\
XT4T2PT1T2T3T4
/\/\
T2T3T3T4
Example:
100100[20]
/\/\\
502005020050
/search(20)/search(20)/\
40======>[20]========>30100
/1.ZigZig\2.ZigZig\\
30at4030at10040200
/\
[20]40
Theimportantthingtonoteis,thesearchorsplayoperationnotonlybringsthesearchedkeytoroot,but
alsobalancestheBST.Forexampleinabovecase,heightofBSTisreducedby1.
Implementation:
// The code is adopted from http://goo.gl/SDH9hH
#include<stdio.h>
#include<stdlib.h>
// An AVL tree node

struct node
{
int key;
};
{
node->key
= key;
return (node);
}

struct node *rightRotate(struct node *x)
{
struct node *y = x->left;
x->left = y->right;
y->right = x;
return y;
}

{
x->right = y->left;
y->left = x;
return y;
}
// This function brings the key at root if key is present in tree.

// If key is not present, then it brings the last accessed item at
// root. This function modifies the tree and returns the new root
struct node *splay(struct node *root, int key)
{
// Base cases: root is NULL or key is present at root
if (root == NULL || root->key == key)
return root;
// Key lies in left subtree

{
if (root->left == NULL) return root;
// Zig-Zig (Left Left)

if (root->left->key > key)
{
// First recursively bring the key as root of left-left
root->left->left = splay(root->left->left, key);
// Do first rotation for root, second rotation is done after else

root = rightRotate(root);
}
else if (root->left->key < key) // Zig-Zag (Left Right)
{
// First recursively bring the key as root of left-right
root->left->right = splay(root->left->right, key);
// Do first rotation for root->left

if (root->left->right != NULL)
}

return (root->left == NULL)? root: rightRotate(root);
}
else // Key lies in right subtree
{
if (root->right == NULL) return root;
// Zag-Zig (Right Left)

if (root->right->key > key)
{
// Bring the key as root of right-left
root->right->left = splay(root->right->left, key);
// Do first rotation for root->right

if (root->right->left != NULL)
}
else if (root->right->key < key)// Zag-Zag (Right Right)
{
// Bring the key as root of right-right and do first rotation
root->right->right = splay(root->right->right, key);
root = leftRotate(root);
}

return (root->right == NULL)? root: leftRotate(root);
}
}
// The search function for Splay tree. Note that this function
// returns the new root of Splay Tree. If key is present in tree
// then, it is moved to root.
struct node *search(struct node *root, int key)
{
return splay(root, key);
}

{
if (root != NULL)
{
}
}

int main()
{
root = search(root, 20);

printf("Preorder traversal of the modified Splay tree is \n");
preOrder(root);
return 0;
}
Output:
PreordertraversalofthemodifiedSplaytreeis
20503040100200
Summary
1)Splaytreeshaveexcellentlocalityproperties.Frequentlyaccesseditemsareeasytofind.Infrequent
itemsareoutofway.
2)AllsplaytreeoperationstakeO(Logn)timeonaverage.Splaytreescanberigorouslyshowntorunin
O(logn)averagetimeperoperation,overanysequenceofoperations(assumingwestartfromanempty
tree)
3)SplaytreesaresimplercomparedtoAVLandRedBlackTreesasnoextrafieldisrequiredineverytree
node.
4)UnlikeAVLtree,asplaytreecanchangeevenwithreadonlyoperationslikesearch.
ApplicationsofSplayTrees
Splaytreeshavebecomethemostwidelyusedbasicdatastructureinventedinthelast30years,because
they'rethefastesttypeofbalancedsearchtreeformanyapplications.
SplaytreesareusedinWindowsNT(inthevirtualmemory,networking,andfilesystemcode),thegcc
compilerandGNUC++library,thesedstringeditor,ForeSystemsnetworkrouters,themostpopular
implementationofUnixmalloc,Linuxloadablekernelmodules,andinmuchothersoftware(Source:
http://www.cs.berkeley.edu/~jrs/61b/lec/36)
=======================================================================
Findnextrightnodeofagivenkey
GivenaBinarytreeandakeyinthebinarytree,findthenoderighttothegivenkey.Ifthereisnonodeon
rightside,thenreturnNULL.ExpectedtimecomplexityisO(n)wherenisthenumberofnodesinthegiven
binarytree.
Forexample,considerthefollowingBinaryTree.Outputfor2is6,outputfor4is5.Outputfor10,6and5is
NULL.
10
/\
26
/\\
845
Solution:TheideaistodolevelordertraversalofgivenBinaryTree.Whenwefindthegivenkey,wejust
checkifthenextnodeinlevelordertraversalisofsamelevel,ifyes,wereturnthenextnode,otherwise
returnNULL.
/* Program to find next right of a given key */
#include <iostream>
#include <queue>

struct node
{
int key;
};
// Method to find next right of given key k, it returns NULL if k is

// not present in tree or k is the rightmost node of its level
node* nextRight(node *root, int k)
{
// Base Case
if (root == NULL)
return 0;
// Create an empty queue for level order tarversal

queue<node *> qn; // A queue to store node addresses
queue<int> ql;
// Another queue to store node levels
int level = 0; // Initialize level as 0
// Enqueue Root and its level

qn.push(root);
ql.push(level);
// A standard BFS loop

while (qn.size())
{
// dequeue an node from qn and its level from ql
node *node = qn.front();
level = ql.front();
qn.pop();
ql.pop();
// If the dequeued node has the given key k

if (node->key == k)
{
// If there are no more items in queue or given node is
// the rightmost node of its level, then return NULL
if (ql.size() == 0 || ql.front() != level)
return NULL;
// Otherwise return next node from queue of nodes

return qn.front();
}
// Standard BFS steps: enqueue children of this node

if (node->left != NULL)
{
qn.push(node->left);
ql.push(level+1);
}
if (node->right != NULL)
{
qn.push(node->right);
ql.push(level+1);
}
}
// We reach here if given key x doesn't exist in tree

return NULL;
}

node* newNode(int key)
{
temp->key = key;
return temp;
}
// A utility function to test above functions

void test(node *root, int k)
{
node *nr = nextRight(root, k);
if (nr != NULL)
cout << "Next Right of " << k << " is " << nr->key << endl;
else
cout << "No next right node found for " << k << endl;
}

int main()
{
test(root, 10);
test(root, 2);
test(root, 6);
test(root, 5);
test(root, 8);
test(root, 4);
return 0;
}
Output:
Nonextrightnodefoundfor10
NextRightof2is6
NextRightof8is4
NextRightof4is5
TimeComplexity:TheabovecodeisasimpleBFStraversalcodewhichvisitseveryenqueueand
dequeuesanodeatmostonce.Therefore,thetimecomplexityisO(n)wherenisthenumberofnodesin
thegivenbinarytree.
=======================================================================
Deepestleftleafnodeinabinarytree
GivenaBinaryTree,findthedeepestleafnodethatisleftchildofitsparent.Forexample,considerthe
followingtree.Thedeepestleftleafnodeisthenodewithvalue9.
1
/\
23
//\
456
\\
78
/\
910
Theideaistorecursivelytraversethegivenbinarytreeandwhiletraversing,maintainlevelwhichwillstore
thecurrentnodeslevelinthetree.Ifcurrentnodeisleftleaf,thencheckifitslevelismorethanthelevelof
deepestleftleafseensofar.Iflevelismorethenupdatetheresult.Ifcurrentnodeisnotleaf,then
recursivelyfindmaximumdepthinleftandrightsubtrees,andreturnmaximumofthetwodepths.Thanksto
Coder011forsuggestingthisapproach.
// A C++ program to find the deepest left leaf in a given binary tree
#include <stdio.h>
#include <iostream>
struct Node
{
int val;
};
Node *newNode(int data)

{
temp->val = data;
return temp;
}
// A utility function to find deepest leaf node.

// lvl: level of current node.
// maxlvl: pointer to the deepest left leaf node found so far
// isLeft: A bool indicate that this node is left child of its parent
// resPtr: Pointer to the result
void deepestLeftLeafUtil(Node *root, int lvl, int *maxlvl,
bool isLeft, Node **resPtr)
{
// Base case
if (root == NULL)
return;
// Update result if this node is left leaf and its level is more
// than the maxl level of the current result
if (isLeft && !root->left && !root->right && lvl > *maxlvl)
{
*resPtr = root;
*maxlvl = lvl;
return;
}

deepestLeftLeafUtil(root->left, lvl+1, maxlvl, true, resPtr);
deepestLeftLeafUtil(root->right, lvl+1, maxlvl, false, resPtr);
}
// A wrapper over deepestLeftLeafUtil().

Node* deepestLeftLeaf(Node *root)
{
int maxlevel = 0;
Node *result = NULL;
deepestLeftLeafUtil(root, 0, &maxlevel, false, &result);
return result;
}

int main()
{
Node* root = newNode(1);
Node *result = deepestLeftLeaf(root);

if (result)
cout << "The deepest left child is " << result->val;
else
cout << "There is no left leaf in the given tree";
return 0;
}
Output:
Thedeepestleftchildis9
=======================================================================
ExtractLeavesofaBinaryTreeina
DoublyLinkedList
GivenaBinaryTree,extractallleavesofitinaDoublyLinkedList(DLL).NotethattheDLLneedtobe
createdinplace.AssumethatthenodestructureofDLLandBinaryTreeissame,onlythemeaningofleft
andrightpointersaredifferent.InDLL,leftmeanspreviouspointerandrightmeansnextpointer.
Letthefollowingbeinputbinarytree
1
/\
23
/\\
456
/\/\
78910
Output:
DoublyLinkedList
7<>8<>5<>9<>10
ModifiedTree:
1
/\
23
/\
46
Weneedtotraverseallleavesandconnectthembychangingtheirleftandrightpointers.Wealsoneedto
removethemfromBinaryTreebychangingleftorrightpointersinparentnodes.Therecanbemanyways
tosolvethis.Inthefollowingimplementation,weaddleavesatthebeginningofcurrentlinkedlistand
updateheadofthelistusingpointertoheadpointer.Sinceweinsertatthebeginning,weneedtoprocess
leavesinreverseorder.Forreverseorder,wefirsttraversetherightsubtreethentheleftsubtree.Weuse
returnvaluestoupdateleftorrightpointersinparentnodes.
// C program to extract leaves of a Binary Tree in a Doubly Linked List
#include <stdio.h>
#include <stdlib.h>
// Structure for tree and linked list

struct Node
{
int data;
};
// Main function which extracts all leaves from given Binary Tree.
// The function returns new root of Binary Tree (Note that root may change
// if Binary Tree has only one node). The function also sets *head_ref as
// head of doubly linked list. left pointer of tree is used as prev in DLL
// and right pointer is used as next
struct Node* extractLeafList(struct Node *root, struct Node **head_ref)
{
// Base cases

{
// This node is going to be added to doubly linked list
// of leaves, set right pointer of this node as previous

// head of DLL. We don't need to set left pointer as left
// is already NULL
root->right = *head_ref;
// Change left pointer of previous head

if (*head_ref != NULL) (*head_ref)->left = root;
// Change head of linked list

*head_ref = root;
return NULL; // Return new root

}
// Recur for right and left subtrees

root->right = extractLeafList(root->right, head_ref);
root->left = extractLeafList(root->left, head_ref);
return root;
}
// Utility function for allocating node for Binary Tree.

{
struct Node* node = (struct Node*)malloc(sizeof(struct Node));
node->data = data;
return node;
}
// Utility function for printing tree in In-Order.

void print(struct Node *root)
{
if (root != NULL)
{
print(root->left);
printf("%d ",root->data);
print(root->right);
}
}
// Utility function for printing double linked list.

void printList(struct Node *head)
{
while (head)
{
printf("%d ", head->data);
head = head->right;
}
}

int main()
{
struct Node *head = NULL;
printf("Inorder Trvaersal of given Tree is:\n");

print(root);
root = extractLeafList(root, &head);
printf("\nExtracted Double Linked list is:\n");

printList(head);
printf("\nInorder traversal of modified tree is:\n");

print(root);
return 0;
}
Output:
InorderTrvaersalofgivenTreeis:
74825139610
ExtractedDoubleLinkedlistis:
785910
Inordertraversalofmodifiedtreeis:
42136
TimeComplexity:O(n),thesolutiondoesasingletraversalofgivenBinaryTree.
=======================================================================
Removeallnodeswhichdontliein
anypathwithsum>=k
Givenabinarytree,acompletepathisdefinedasapathfromroottoaleaf.Thesumofallnodesonthat
pathisdefinedasthesumofthatpath.GivenanumberK,youhavetoremove(prunethetree)allnodes
whichdontlieinanypathwithsum>=k.
Note:Anodecanbepartofmultiplepaths.Sowehavetodeleteitonlyincasewhenallpathsfromithave
sumlessthanK.
ConsiderthefollowingBinaryTree
1
/\
23
/\/\
4567
/\//
891210
/\\
131411
/
15
Forinputk=20,thetreeshouldbechangedtofollowing
(Nodeswithvalues6and8aredeleted)
1
/\
23
/\\
457
\//
91210
/\\
131411
/
15
Forinputk=45,thetreeshouldbechangedtofollowing.
1
/
2
/
4
\
9
\
14
/
15
Theideaistotraversethetreeanddeletenodesinbottomupmanner.Whiletraversingthetree,recursively
calculatethesumofnodesfromroottoleafnodeofeachpath.Foreachvisitednode,checksthetotal
calculatedsumagainstgivensumk.Ifsumislessthank,thenfree(delete)thatnode(leafnode)and
returnthesumbacktothepreviousnode.Sincethepathisfromroottoleafandnodesaredeletedin
bottomupmanner,anodeisdeletedonlywhenallofitsdescendantsaredeleted.Therefore,whenanode
isdeleted,itmustbealeafinthecurrentBinaryTree.
#include <stdio.h>
#include <stdlib.h>
// A utility function to get maximum of two integers

int max(int l, int r) { return (l > r ? l : r); }

struct Node
{
int data;
};
// A utility function to create a new Binary Tree node with given data
{
struct Node* node = (struct Node*) malloc(sizeof(struct Node));
node->data = data;
return node;
}
// print the tree in LVR (Inorder traversal) way.

void print(struct Node *root)
{
if (root != NULL)
{
print(root->left);
print(root->right);
}
}
/* Main function which truncates the binary tree. */

struct Node *pruneUtil(struct Node *root, int k, int *sum)
{
// Base Case
// Initialize left and right sums as sum from root to

// this node (including this node)
int lsum = *sum + (root->data);
int rsum = lsum;
// Recursively prune left and right subtrees

root->left = pruneUtil(root->left, k, &lsum);
root->right = pruneUtil(root->right, k, &rsum);
// Get the maximum of left and right sums

*sum = max(lsum, rsum);
// If maximum is smaller than k, then this node

// must be deleted
if (*sum < k)
{
free(root);
root = NULL;
}
return root;
}
// A wrapper over pruneUtil()

struct Node *prune(struct Node *root, int k)
{
int sum = 0;
return pruneUtil(root, k, &sum);
}

int main()
{
int k = 45;
root->right->right->left->right = newNode(11);
root->left->left->right->left = newNode(13);
root->left->left->right->right = newNode(14);
root->left->left->right->right->left = newNode(15);
printf("Tree before truncation\n");

print(root);
root = prune(root, k); // k is 45
printf("\n\nTree after truncation\n");

print(root);
return 0;
}
Output:
Treebeforetruncation
841391514212516310117
Treeaftertruncation
49151421
TimeComplexity:O(n),thesolutiondoesasingletraversalofgivenBinaryTree.
=======================================================================
Addallgreatervaluestoeverynode
inagivenBST
GivenaBinarySearchTree(BST),modifyitsothatallgreatervaluesinthegivenBSTareaddedtoevery
node.Forexample,considerthefollowingBST.
50
/\
3070
/\/\
20406080
Theabovetreeshouldbemodifiedtofollowing
260
/\
330150
/\/\
35030021080
Asimplemethodforsolvingthisistofindsumofallgreatervaluesforeverynode.Thismethodwouldtake
O(n^2)time.
Wecandoitusingasingletraversal.TheideaistousefollowingBSTproperty.IfwedoreverseInorder
traversalofBST,wegetallnodesindecreasingorder.WedoreverseInordertraversalandkeeptrackof
thesumofallnodesvisitedsofar,weaddthissumtoeverynode.
// C program to add all greater values in every node of BST
#include<stdio.h>
#include<stdlib.h>
struct node
{
int data;
};
struct node *newNode(int item)

{
struct node *temp = (struct node *)malloc(sizeof(struct node));
temp->data = item;
return temp;
}
// Recursive function to add all greater values in every node

void modifyBSTUtil(struct node *root, int *sum)
{
// Base Case
// Recur for right subtree

modifyBSTUtil(root->right, sum);
// Now *sum has sum of nodes in right subtree, add

// root->data to sum and update root->data
*sum = *sum + root->data;
root->data = *sum;

modifyBSTUtil(root->left, sum);
}
// A wrapper over modifyBSTUtil()

void modifyBST(struct node *root)
{
int sum = 0;
modifyBSTUtil(root, &sum);
}
// A utility function to do inorder traversal of BST

void inorder(struct node *root)
{
if (root != NULL)
{
printf("%d ", root->data);
}
}
/* A utility function to insert a new node with given data in BST */

struct node* insert(struct node* node, int data)
{
/* If the tree is empty, return a new node */
if (node == NULL) return newNode(data);
/* Otherwise, recur down the tree */

if (data <= node->data)
node->left = insert(node->left, data);
else
node->right = insert(node->right, data);
/* return the (unchanged) node pointer */

return node;
}

int main()
{
/* Let us create following BST
50
/\
30
70
/ \/ \
20
40 60
80 */
struct node *root = NULL;

insert(root, 30);
insert(root, 20);
insert(root, 40);
insert(root, 70);
insert(root, 60);
insert(root, 80);
modifyBST(root);
// print inoder tarversal of the modified BST

inorder(root);
return 0;
}
Output
35033030026021015080
TimeComplexity:O(n)wherenisnumberofnodesinthegivenBST.
Asasidenote,wecanalsousereverseInordertraversaltofindkthlargestelementinaBST.
=======================================================================
BTree|Set3(Delete)
Itisrecommendedtoreferfollowingpostsasprerequisiteofthispost.
BTree|Set1(Introduction)
BTree|Set2(Insert)
BTreeisatypeofamultiwaysearchtree.So,ifyouarenotfamiliarwithmultiwaysearchtreesingeneral,
itisbettertotakealookatthisvideolecturefromIITDelhi,beforeproceedingfurther.Onceyougetthe
basicsofamultiwaysearchtreeclear,BTreeoperationswillbeeasiertounderstand.
SourceofthefollowingexplanationandalgorithmisIntroductiontoAlgorithms3rdEditionbyCliffordStein,
ThomasH.Cormen,CharlesE.Leiserson,RonaldL.Rivest
Deletionprocess:
DeletionfromaBtreeismorecomplicatedthaninsertion,becausewecandeleteakeyfromanynodenot
justaleafandwhenwedeleteakeyfromaninternalnode,wewillhavetorearrangethenodeschildren.
Asininsertion,wemustmakesurethedeletiondoesntviolatetheBtreeproperties.Justaswehadto
ensurethatanodedidntgettoobigduetoinsertion,wemustensurethatanodedoesntgettoosmall
duringdeletion(exceptthattherootisallowedtohavefewerthantheminimumnumbert1ofkeys).Justas
asimpleinsertionalgorithmmighthavetobackupifanodeonthepathtowherethekeywastobeinserted
wasfull,asimpleapproachtodeletionmighthavetobackupifanode(otherthantheroot)alongthepathto
wherethekeyistobedeletedhastheminimumnumberofkeys.
Thedeletionproceduredeletesthekeykfromthesubtreerootedatx.Thisprocedureguaranteesthat
wheneveritcallsitselfrecursivelyonanodex,thenumberofkeysinxisatleasttheminimumdegreet.
NotethatthisconditionrequiresonemorekeythantheminimumrequiredbytheusualBtreeconditions,so
thatsometimesakeymayhavetobemovedintoachildnodebeforerecursiondescendstothatchild.This
strengthenedconditionallowsustodeleteakeyfromthetreeinonedownwardpasswithouthavingtoback
up(withoneexception,whichwellexplain).Youshouldinterpretthefollowingspecificationfordeletion
fromaBtreewiththeunderstandingthatiftherootnodexeverbecomesaninternalnodehavingnokeys
(thissituationcanoccurincases2cand3bthenwedeletex,andxsonlychildx.c1becomesthenewroot
ofthetree,decreasingtheheightofthetreebyoneandpreservingthepropertythattherootofthetree
containsatleastonekey(unlessthetreeisempty).
WesketchhowdeletionworkswithvariouscasesofdeletingkeysfromaBtree.
1.
Ifthekeykisinnodexandxisaleaf,deletethekeykfromx.
2.
Ifthekeykisinnodexandxisaninternalnode,dothefollowing.
a)
Ifthechildythatprecedeskinnodexhasatleasttkeys,thenfindthepredecessork0ofkinthe
subtreerootedaty.Recursivelydeletek0,andreplacekbyk0inx.(Wecanfindk0anddeleteitinasingle
downwardpass.)
b)
Ifyhasfewerthantkeys,then,symmetrically,examinethechildzthatfollowskinnodex.Ifzhas
atleasttkeys,thenfindthesuccessork0ofkinthesubtreerootedatz.Recursivelydeletek0,andreplace
kbyk0inx.(Wecanfindk0anddeleteitinasingledownwardpass.)
c)Otherwise,ifbothyandzhaveonlyt1keys,mergekandallofzintoy,sothatxlosesbothkandthe
pointertoz,andynowcontains2t1keys.Thenfreezandrecursivelydeletekfromy.
3.
Ifthekeykisnotpresentininternalnodex,determinetherootx.c(i)oftheappropriatesubtreethat
mustcontaink,ifkisinthetreeatall.Ifx.c(i)hasonlyt1keys,executestep3aor3basnecessaryto
guaranteethatwedescendtoanodecontainingatleasttkeys.Thenfinishbyrecursingontheappropriate
childofx.
a)
Ifx.c(i)hasonlyt1keysbuthasanimmediatesiblingwithatleasttkeys,givex.c(i)anextrakeyby
movingakeyfromxdownintox.c(i),movingakeyfromx.c(i)simmediateleftorrightsiblingupintox,and
movingtheappropriatechildpointerfromthesiblingintox.c(i).
b)
Ifx.c(i)andbothofx.c(i)simmediatesiblingshavet1keys,mergex.c(i)withonesibling,which
involvesmovingakeyfromxdownintothenewmergednodetobecomethemediankeyforthatnode.
SincemostofthekeysinaBtreeareintheleaves,deletionoperationsaremostoftenusedtodeletekeys
fromleaves.Therecursivedeleteprocedurethenactsinonedownwardpassthroughthetree,without
havingtobackup.Whendeletingakeyinaninternalnode,however,theproceduremakesadownward
passthroughthetreebutmayhavetoreturntothenodefromwhichthekeywasdeletedtoreplacethekey
withitspredecessororsuccessor(cases2aand2b).
ThefollowingfiguresfromCLRSbookexplainthedeletionporcess.

Implementation:
FollowingisC++implementationofdeletionprocess.
/* The following program performs deletion on a B-Tree. It contains functions
specific for deletion along with all the other functions provided in the
previous articles on B-Trees. See
http://www.geeksforgeeks.org/b-tree-set-1-introduction-2/
for previous article.
The deletion function has been compartmentalized into 8 functions for ease
of understanding and clarity
The following functions are exclusive for deletion

In class BTreeNode:
1) remove
2) removeFromLeaf
3) removeFromNonLeaf
4) getPred
5) getSucc
6) borrowFromPrev
7) borrowFromNext
8) merge
9) findKey
In class BTree:
1) remove
The removal of a key from a B-Tree is a fairly complicated process. The program
handles
all the 6 different cases that might arise while removing a key.
Testing: The code has been tested using the B-Tree provided in the CLRS book(
included
in the main function ) along with other cases.
Reference: CLRS3 - Chapter 18 - (499-502)

It is advised to read the material in CLRS before taking a look at the code. */
#include<iostream>
// A BTree node
class BTreeNode
{
int *keys; // An array of keys
int t;
// Minimum degree (defines the range for number of keys)
BTreeNode **C; // An array of child pointers

int n;
// Current number of keys
bool leaf; // Is true when node is leaf. Otherwise false
public:
BTreeNode(int _t, bool _leaf);
// Constructor
// A function to traverse all nodes in a subtree rooted with this node

void traverse();
// A function to search a key in subtree rooted with this node.

BTreeNode *search(int k);
// returns NULL if k is not present.
// A function that returns the index of the first key that is greater
// or equal to k
int findKey(int k);
// A utility function to insert a new key in the subtree rooted with

// this node. The assumption is, the node must be non-full when this
// function is called
void insertNonFull(int k);
// A utility function to split the child y of this node. i is index

// of y in child array C[]. The Child y must be full when this
void splitChild(int i, BTreeNode *y);
// A wrapper function to remove the key k in subtree rooted with

// this node.
void remove(int k);
// A function to remove the key present in idx-th position in

// this node which is a leaf
void removeFromLeaf(int idx);
// A function to remove the key present in idx-th position in

// this node which is a non-leaf node
void removeFromNonLeaf(int idx);
// A function to get the predecessor of the key- where the key

// is present in the idx-th position in the node
int getPred(int idx);
// A function to get the successor of the key- where the key
// is present in the idx-th position in the node

int getSucc(int idx);
// A function to fill up the child node present in the idx-th

// position in the C[] array if that child has less than t-1 keys
void fill(int idx);
// A function to borrow a key from the C[idx-1]-th node and place

// it in C[idx]th node
void borrowFromPrev(int idx);
// A function to borrow a key from the C[idx+1]-th node and place it

// in C[idx]th node
void borrowFromNext(int idx);
// A function to merge idx-th child of the node with (idx+1)th child of

// the node
void merge(int idx);
// Make BTree friend of this so that we can access private members of

// this class in BTree functions
friend class BTree;
};
class BTree
{
BTreeNode *root; // Pointer to root node
int t; // Minimum degree
public:
// Constructor (Initializes tree as empty)

BTree(int _t)
{
root = NULL;
t = _t;
}
void traverse()
{
if (root != NULL) root->traverse();
}
// function to search a key in this tree

BTreeNode* search(int k)
{
return (root == NULL)? NULL : root->search(k);
}
// The main function that inserts a new key in this B-Tree

void insert(int k);
// The main function that removes a new key in thie B-Tree

void remove(int k);
};
BTreeNode::BTreeNode(int t1, bool leaf1)

{
// Copy the given minimum degree and leaf property
t = t1;
leaf = leaf1;
// Allocate memory for maximum number of possible keys

// and child pointers
keys = new int[2*t-1];
C = new BTreeNode *[2*t];
// Initialize the number of keys as 0

n = 0;
}
// A utility function that returns the index of the first key that is
// greater than or equal to k
int BTreeNode::findKey(int k)
{
int idx=0;
while (idx<n && keys[idx] < k)
++idx;
return idx;
}
// A function to remove the key k from the sub-tree rooted with this node
void BTreeNode::remove(int k)
{
int idx = findKey(k);
// The key to be removed is present in this node

if (idx < n && keys[idx] == k)
{
// If the node is a leaf node - removeFromLeaf is called

// Otherwise, removeFromNonLeaf function is called
if (leaf)
removeFromLeaf(idx);
else
removeFromNonLeaf(idx);
}
else
{
// If this node is a leaf node, then the key is not present in tree
if (leaf)
{
cout << "The key "<< k <<" is does not exist in the tree\n";
return;
}
// The key to be removed is present in the sub-tree rooted with this node
// The flag indicates whether the key is present in the sub-tree rooted
// with the last child of this node
bool flag = ( (idx==n)? true : false );
// If the child where the key is supposed to exist has less that t keys,
// we fill that child
if (C[idx]->n < t)
fill(idx);
// If the last child has been merged, it must have merged with the previous
// child and so we recurse on the (idx-1)th child. Else, we recurse on the
// (idx)th child which now has atleast t keys
if (flag && idx > n)
C[idx-1]->remove(k);
else
C[idx]->remove(k);
}
return;
}
// A function to remove the idx-th key from this node - which is a leaf node
void BTreeNode::removeFromLeaf (int idx)
{
// Move all the keys after the idx-th pos one place backward
for (int i=idx+1; i<n; ++i)
keys[i-1] = keys[i];
// Reduce the count of keys

n--;
return;
}
// A function to remove the idx-th key from this node - which is a non-leaf node
void BTreeNode::removeFromNonLeaf(int idx)
{
int k = keys[idx];
// If the child that precedes k (C[idx]) has atleast t keys,

// find the predecessor 'pred' of k in the subtree rooted at
// C[idx]. Replace k by pred. Recursively delete pred
// in C[idx]
if (C[idx]->n >= t)
{
int pred = getPred(idx);
keys[idx] = pred;
C[idx]->remove(pred);
}
// If the child C[idx] has less that t keys, examine C[idx+1].

// If C[idx+1] has atleast t keys, find the successor 'succ' of k in
// the subtree rooted at C[idx+1]
// Replace k by succ
// Recursively delete succ in C[idx+1]
else if (C[idx+1]->n >= t)
{
int succ = getSucc(idx);
keys[idx] = succ;
C[idx+1]->remove(succ);
}
// If both C[idx] and C[idx+1] has less that t keys,merge k and all of C[idx+1]
// into C[idx]
// Now C[idx] contains 2t-1 keys
// Free C[idx+1] and recursively delete k from C[idx]
else
{
merge(idx);
C[idx]->remove(k);
}
return;
}
// A function to get predecessor of keys[idx]

int BTreeNode::getPred(int idx)
{
// Keep moving to the right most node until we reach a leaf
BTreeNode *cur=C[idx];
while (!cur->leaf)
cur = cur->C[cur->n];
// Return the last key of the leaf

return cur->keys[cur->n-1];
}
int BTreeNode::getSucc(int idx)

{
// Keep moving the left most node starting from C[idx+1] until we reach a leaf
BTreeNode *cur = C[idx+1];
while (!cur->leaf)
cur = cur->C[0];
// Return the first key of the leaf

return cur->keys[0];
}
// A function to fill child C[idx] which has less than t-1 keys
void BTreeNode::fill(int idx)
{
// If the previous child(C[idx-1]) has more than t-1 keys, borrow a key
// from that child
if (idx!=0 && C[idx-1]->n>=t)
borrowFromPrev(idx);
// If the next child(C[idx+1]) has more than t-1 keys, borrow a key
// from that child
else if (idx!=n && C[idx+1]->n>=t)
borrowFromNext(idx);
// Merge C[idx] with its sibling
// If C[idx] is the last child, merge it with with its previous sibling
// Otherwise merge it with its next sibling
else
{
if (idx != n)
merge(idx);
else
merge(idx-1);
}
return;
}
// A function to borrow a key from C[idx-1] and insert it

// into C[idx]
void BTreeNode::borrowFromPrev(int idx)
{
BTreeNode *child=C[idx];
BTreeNode *sibling=C[idx-1];
// The last key from C[idx-1] goes up to the parent and key[idx-1]
// from parent is inserted as the first key in C[idx]. Thus, the loses
// sibling one key and child gains one key
// Moving all key in C[idx] one step ahead

for (int i=child->n-1; i>=0; --i)
child->keys[i+1] = child->keys[i];
// If C[idx] is not a leaf, move all its child pointers one step ahead
if (!child->leaf)
{
for(int i=child->n; i>=0; --i)
child->C[i+1] = child->C[i];
}
// Setting child's first key equal to keys[idx-1] from the current node
child->keys[0] = keys[idx-1];
// Moving sibling's last child as C[idx]'s first child

if (!leaf)
child->C[0] = sibling->C[sibling->n];
// Moving the key from the sibling to the parent

// This reduces the number of keys in the sibling
keys[idx-1] = sibling->keys[sibling->n-1];
child->n += 1;
sibling->n -= 1;
return;
}
// A function to borrow a key from the C[idx+1] and place

// it in C[idx]
void BTreeNode::borrowFromNext(int idx)
{
BTreeNode *child=C[idx];
BTreeNode *sibling=C[idx+1];
// keys[idx] is inserted as the last key in C[idx]

child->keys[(child->n)] = keys[idx];
// Sibling's first child is inserted as the last child

// into C[idx]
if (!(child->leaf))
child->C[(child->n)+1] = sibling->C[0];
//The first key from sibling is inserted into keys[idx]

keys[idx] = sibling->keys[0];
// Moving all keys in sibling one step behind

for (int i=1; i<sibling->n; ++i)
sibling->keys[i-1] = sibling->keys[i];
// Moving the child pointers one step behind

if (!sibling->leaf)
{
for(int i=1; i<=sibling->n; ++i)
sibling->C[i-1] = sibling->C[i];
}
// Increasing and decreasing the key count of C[idx] and C[idx+1]

// respectively
child->n += 1;
sibling->n -= 1;
return;
}
// A function to merge C[idx] with C[idx+1]

// C[idx+1] is freed after merging
void BTreeNode::merge(int idx)
{
BTreeNode *child = C[idx];
BTreeNode *sibling = C[idx+1];
// Pulling a key from the current node and inserting it into (t-1)th
// position of C[idx]
child->keys[t-1] = keys[idx];
// Copying the keys from C[idx+1] to C[idx] at the end

for (int i=0; i<sibling->n; ++i)
child->keys[i+t] = sibling->keys[i];
// Copying the child pointers from C[idx+1] to C[idx]

if (!child->leaf)
{
for(int i=0; i<=sibling->n; ++i)
child->C[i+t] = sibling->C[i];
}
// Moving all keys after idx in the current node one step before // to fill the gap created by moving keys[idx] to C[idx]
for (int i=idx+1; i<n; ++i)
keys[i-1] = keys[i];
// Moving the child pointers after (idx+1) in the current node one
// step before
for (int i=idx+2; i<=n; ++i)
C[i-1] = C[i];
// Updating the key count of child and the current node

child->n += sibling->n+1;
n--;
// Freeing the memory occupied by sibling

delete(sibling);
return;
}

void BTree::insert(int k)
{
// If tree is empty
if (root == NULL)
{
// Allocate memory for root
root = new BTreeNode(t, true);
root->keys[0] = k; // Insert key
root->n = 1; // Update number of keys in root
}
else // If tree is not empty
{
// If root is full, then tree grows in height
if (root->n == 2*t-1)
{
// Allocate memory for new root
BTreeNode *s = new BTreeNode(t, false);
// Make old root as child of new root

s->C[0] = root;
// Split the old root and move 1 key to the new root
s->splitChild(0, root);
// New root has two children now. Decide which of the

// two children is going to have new key
int i = 0;
if (s->keys[0] < k)
i++;
s->C[i]->insertNonFull(k);
// Change root
root = s;
}
else // If root is not full, call insertNonFull for root
root->insertNonFull(k);
}
}
// A utility function to insert a new key in this node

// The assumption is, the node must be non-full when this
void BTreeNode::insertNonFull(int k)
{
// Initialize index as index of rightmost element
int i = n-1;
// If this is a leaf node

if (leaf == true)
{
// The following loop does two things
// a) Finds the location of new key to be inserted
// b) Moves all greater keys to one place ahead
while (i >= 0 && keys[i] > k)

{
keys[i+1] = keys[i];
i--;
}
// Insert the new key at found location

keys[i+1] = k;
n = n+1;
}
else // If this node is not leaf
{
// Find the child which is going to have the new key
i--;
// See if the found child is full

if (C[i+1]->n == 2*t-1)
{
// If the child is full, then split it
splitChild(i+1, C[i+1]);
// After split, the middle key of C[i] goes up and

// C[i] is splitted into two. See which of the two
// is going to have the new key
if (keys[i+1] < k)
i++;
}
C[i+1]->insertNonFull(k);
}
}
// A utility function to split the child y of this node

// Note that y must be full when this function is called
void BTreeNode::splitChild(int i, BTreeNode *y)
{
// Create a new node which is going to store (t-1) keys
// of y
BTreeNode *z = new BTreeNode(y->t, y->leaf);
z->n = t - 1;
// Copy the last (t-1) keys of y to z

for (int j = 0; j < t-1; j++)
z->keys[j] = y->keys[j+t];
// Copy the last t children of y to z

if (y->leaf == false)
{
for (int j = 0; j < t; j++)
z->C[j] = y->C[j+t];
}
// Reduce the number of keys in y

y->n = t - 1;
// Since this node is going to have a new child,

// create space of new child
for (int j = n; j >= i+1; j--)
C[j+1] = C[j];
// Link the new child to this node

C[i+1] = z;
// A key of y will move to this node. Find location of

// new key and move all greater keys one space ahead
for (int j = n-1; j >= i; j--)
keys[j+1] = keys[j];
// Copy the middle key of y to this node

keys[i] = y->keys[t-1];
// Increment count of keys in this node

n = n + 1;
}
// Function to traverse all nodes in a subtree rooted with this node

void BTreeNode::traverse()
{
// There are n keys and n+1 children, travers through n keys
// and first n children
int i;
for (i = 0; i < n; i++)
{
// If this is not leaf, then before printing key[i],
// traverse the subtree rooted with child C[i].
if (leaf == false)
C[i]->traverse();
cout << " " << keys[i];
}
// Print the subtree rooted with last child

if (leaf == false)
C[i]->traverse();
}
// Function to search key k in subtree rooted with this node

BTreeNode *BTreeNode::search(int k)
{
// Find the first key greater than or equal to k
int i = 0;
while (i < n && k > keys[i])
i++;
// If the found key is equal to k, return this node

if (keys[i] == k)
return this;
// If key is not found here and this is a leaf node

if (leaf == true)
return NULL;
// Go to the appropriate child

return C[i]->search(k);
}
void BTree::remove(int k)
{
if (!root)
{
cout << "The tree is empty\n";
return;
}
// Call the remove function for root

root->remove(k);
// If the root node has 0 keys, make its first child as the new root
// if it has a child, otherwise set root as NULL
if (root->n==0)
{
BTreeNode *tmp = root;
if (root->leaf)
root = NULL;
else
root = root->C[0];
// Free the old root

delete tmp;
}
return;
}

int main()
{
BTree t(3); // A B-Tree with minium degree 3
t.insert(1);
t.insert(3);
t.insert(7);
t.insert(10);
t.insert(11);
t.insert(13);
t.insert(14);
t.insert(15);
t.insert(18);
t.insert(16);
t.insert(19);
t.insert(24);
t.insert(25);
t.insert(26);
t.insert(21);
t.insert(4);
t.insert(5);
t.insert(20);
t.insert(22);
t.insert(2);
t.insert(17);
t.insert(12);
t.insert(6);
cout << "Traversal of tree constructed is\n";

t.traverse();
cout << endl;
t.remove(6);
cout << "Traversal of tree after removing 6\n";
t.traverse();
cout << endl;
t.remove(13);
t.traverse();
cout << endl;
t.remove(7);
t.traverse();
cout << endl;
t.remove(4);
t.traverse();
cout << endl;
t.remove(2);
t.traverse();
cout << endl;
t.remove(16);
t.traverse();
cout << endl;
return 0;
}
Output:
Traversaloftreeconstructedis
123456710111213141516171819202122242526
Traversaloftreeafterremoving6
12345710111213141516171819202122242526
123457101112141516171819202122242526
12345101112141516171819202122242526
1235101112141516171819202122242526
135101112141516171819202122242526
1351011121415171819202122242526
=======================================================================
PrintLeftViewofaBinaryTree
GivenaBinaryTree,printleftviewofit.LeftviewofaBinaryTreeissetofnodesvisiblewhentreeisvisited
fromleftside.Leftviewoffollowingtreeis12,10,25.
12
/\
1030
/\
2540
Theleftviewcontainsallnodesthatarefirstnodesintheirlevels.Asimplesolutionistodolevelorder
traversalandprintthefirstnodeineverylevel.
Theproblemcanalsobesolvedusingsimplerecursivetraversal.Wecankeeptrackoflevelofanodeby
passingaparametertoallrecursivecalls.Theideaistokeeptrackofmaximumlevelalso.Wheneverwe
seeanodewhoselevelismorethanmaximumlevelsofar,weprintthenodebecausethisisthefirstnode
initslevel(Notethatwetraversetheleftsubtreebeforerightsubtree).FollowingisCimplementationofthis
approach.
// C program to print left view of Binary Tree
#include<stdio.h>
#include<stdlib.h>
struct node
{
int data;
};
// A utility function to create a new Binary Tree node

struct node *newNode(int item)
{
struct node *temp = (struct node *)malloc(sizeof(struct node));
temp->data = item;
return temp;
}
// Recursive function to print left view of a binary tree.

void leftViewUtil(struct node *root, int level, int *max_level)
{
// Base Case
if (root==NULL) return;
// If this is the first node of its level

if (*max_level < level)
{
printf("%d\t", root->data);
*max_level = level;
}

leftViewUtil(root->left, level+1, max_level);
leftViewUtil(root->right, level+1, max_level);
}
// A wrapper over leftViewUtil()

void leftView(struct node *root)
{
int max_level = 0;
leftViewUtil(root, 1, &max_level);
}

int main()
{
leftView(root);
return 0;
}
Output:
121025
=======================================================================
Checkifallleavesareatsamelevel
GivenaBinaryTree,checkifallleavesareatsamelevelornot.
12
/\
57
/\
31
Leavesareatsamelevel
12
/\
57
LeavesareNotatsamelevel
12
/
5
/\
39
//
12
TheideaistofirstfindleveloftheleftmostleafandstoreitinavariableleafLevel.Thencomparelevelofall
otherleaveswithleafLevel,ifsame,returntrue,elsereturnfalse.WetraversethegivenBinaryTreein
Preorderfashion.Anargumentleaflevelispassedtoallcalls.ThevalueofleafLevelisinitializedas0to
indicatethatthefirstleafisnotyetseenyet.Thevalueisupdatedwhenwefindfirstleaf.Levelof
subsequentleaves(inpreorder)iscomparedwithleafLevel.
// C program to check if all leaves are at same level
#include <stdio.h>
#include <stdlib.h>
// A binary tree node

struct Node
{
int data;
};
// A utility function to allocate a new tree node

{
node->data = data;
return node;
}
/* Recursive function which checks whether all leaves are at same level */
bool checkUtil(struct Node *root, int level, int *leafLevel)
{
// Base case
if (root == NULL) return true;
// If a leaf node is encountered

{
// When a leaf node is found first time
if (*leafLevel == 0)
{
*leafLevel = level; // Set first found leaf's level

return true;
}
// If this is not first leaf node, compare its level with

// first leaf's level
return (level == *leafLevel);
}
// If this node is not leaf, recursively check left and right subtrees
return checkUtil(root->left, level+1, leafLevel) &&
checkUtil(root->right, level+1, leafLevel);
}
/* The main function to check if all leafs are at same level.

It mainly uses checkUtil() */
bool check(struct Node *root)
{
int level = 0, leafLevel = 0;
return checkUtil(root, level, &leafLevel);
}

int main()
{
// Let us create tree shown in thirdt example
if (check(root))
printf("Leaves are at same level\n");
else
printf("Leaves are not at same level\n");
getchar();
return 0;
}
Output:
=======================================================================
Finddepthofthedeepestoddlevel
leafnode
WriteaCcodetogetthedepthofthedeepestoddlevelleafnodeinabinarytree.Considerthatlevelstarts
with1.Depthofaleafnodeisnumberofnodesonthepathfromroottoleaf(includingbothleafandroot).
Forexample,considerthefollowingtree.Thedeepestoddlevelnodeisthenodewithvalue9anddepthof
thisnodeis5.
1
/\
23
//\
456
\\
78
/\
910
/
11
Theideaistorecursivelytraversethegivenbinarytreeandwhiletraversing,maintainavariablelevel
whichwillstorethecurrentnodeslevelinthetree.Ifcurrentnodeisleafthenchecklevelisoddornot.If
levelisoddthenreturnit.Ifcurrentnodeisnotleaf,thenrecursivelyfindmaximumdepthinleftandright
subtrees,andreturnmaximumofthetwodepths.
// C program to find depth of the deepest odd level leaf node
#include <stdio.h>
#include <stdlib.h>
// A utility function to find maximum of two integers
int max(int x, int y) { return (x > y)? x : y; }

struct Node
{
int data;
};
// A utility function to allocate a new tree node

{
node->data = data;
return node;
}
// A recursive function to find depth of the deepest odd level leaf

int depthOfOddLeafUtil(Node *root,int level)
{
// Base Case
if (root == NULL)
return 0;
// If this node is a leaf and its level is odd, return its level
if (root->left==NULL && root->right==NULL && level&1)
return level;
// If not leaf, return the maximum value from left and right subtrees
return max(depthOfOddLeafUtil(root->left, level+1),
depthOfOddLeafUtil(root->right, level+1));
}
/* Main function which calculates the depth of deepest odd level leaf.
This function mainly uses depthOfOddLeafUtil() */
int depthOfOddLeaf(struct Node *root)
{
int level = 1, depth = 0;
return depthOfOddLeafUtil(root, level);
}

int main()
{
struct Node* root = newNode(1);
root->right->right->right->right->left = newNode(11);
printf("%d is the required depth\n", depthOfOddLeaf(root));

getchar();
return 0;
}
Output:
5istherequireddepth
=======================================================================
PrintPostordertraversalfromgiven
InorderandPreordertraversals
GivenInorderandPreordertraversalsofabinarytree,printPostordertraversal.
Example:
Input:
Inordertraversalin[]={4,2,5,1,3,6}
Preordertraversalpre[]={1,2,4,5,3,6}
Output:
Postordertraversalis{4,5,2,6,3,1}
Trversalsintheaboveexamplerepresentsfollowingtree
1
/\
23
/\\
456
Anaivemethodistofirstconstructthetree,thenusesimplerecursivemethodtoprintpostordertraversalof
theconstructedtree.
Wecanprintpostordertraversalwithoutconstructingthetree.Theideais,rootisalwaysthefirstitem
inpreordertraversalanditmustbethelastiteminpostordertraversal.Wefirstrecursivelyprintleftsubtree,
thenrecursivelyprintrightsubtree.Finally,printroot.Tofindboundariesofleftandrightsubtreesinpre[]
andin[],wesearchrootinin[],allelementsbeforerootinin[]areelementsofleftsubtreeandallelements
afterrootareelementsofrightsubtree.Inpre[],allelementsafterindexofrootinin[]areelementsofright
subtree.Andelementsbeforeindex(includingtheelementatindexandexcludingthefirstelement)are
elementsofleftsubtree.
// C++ program to print postorder traversal from preorder and inorder traversals
#include <iostream>
// A utility function to search x in arr[] of size n

int search(int arr[], int x, int n)
{
for (int i = 0; i < n; i++)
if (arr[i] == x)
return i;
return -1;
}
// Prints postorder traversal from given inorder and preorder traversals

void printPostOrder(int in[], int pre[], int n)
{
// The first element in pre[] is always root, search it
// in in[] to find left and right subtrees
int root = search(in, pre[0], n);
// If left subtree is not empty, print left subtree

if (root != 0)
printPostOrder(in, pre+1, root);
// If right subtree is not empty, print right subtree

if (root != n-1)
printPostOrder(in+root+1, pre+root+1, n-root-1);
// Print root
cout << pre[0] << " ";
}

int main()
{
int in[] = {4, 2, 5, 1, 3, 6};
int pre[] = {1, 2, 4, 5, 3, 6};
int n = sizeof(in)/sizeof(in[0]);
cout << "Postorder traversal " << endl;
printPostOrder(in, pre, n);
return 0;
}
Output
Postordertraversal
452631
TimeComplexity:Theabovefunctionvisitseverynodeinarray.Foreveryvisit,itcallssearchwhichtakes
O(n)time.Therefore,overalltimecomplexityofthefunctionisO(n2)
=======================================================================
Differencebetweensumsofoddlevel
andevenlevelnodesofaBinaryTree
GivenaaBinaryTree,findthedifferencebetweenthesumofnodesatoddlevelandthesumofnodesat
evenlevel.Considerrootaslevel1,leftandrightchildrenofrootaslevel2andsoon.
Forexample,inthefollowingtree,sumofnodesatoddlevelis(5+1+4+8)whichis18.Andsumof
nodesatevenlevelis(2+6+3+7+9)whichis27.Theoutputforfollowingtreeshouldbe1827which
is9.
5
/\
26
/\\
148
//\
379
Astraightforwardmethodistouselevelordertraversal.Inthetraversal,checklevelofcurrentnode,ifitis
odd,incrementoddsumbydataofcurrentnode,otherwiseincrementevensum.Finallyreturndifference
betweenoddsumandevensum.Seefollowingforimplementationofthisapproach.
Cimplementationoflevelordertraversalbasedapproachtofindthedifference.
Theproblemcanalsobesolvedusingsimplerecursivetraversal.Wecanrecursivelycalculatethe
requireddifferenceas,valueofrootsdatasubtractedbythedifferenceforsubtreeunderleftchildandthe
differenceforsubtreeunderrightchild.FollowingisCimplementationofthisapproach.
// A recursive program to find difference between sum of nodes at
// odd level and sum at even level
#include <stdio.h>
#include <stdlib.h>
// Binary Tree node

struct node
{
int data;
struct node* left, *right;
};
// A utility function to allocate a new tree node with given data

{
node->data = data;
return (node);
}
// The main function that return difference between odd and even level
// nodes
int getLevelDiff(struct node *root)
{
// Base case
if (root == NULL)
return 0;
// Difference for root is root's data - difference for left subtree

// - difference for right subtree
return root->data - getLevelDiff(root->left) - getLevelDiff(root->right);
}

int main()
{
printf("%d is the required difference\n", getLevelDiff(root));
getchar();
return 0;
}
Output:
9istherequireddifference
TimecomplexityofbothmethodsisO(n),butthesecondmethodissimpleandeasytoimplement.
=======================================================================
Printancestorsofagivenbinarytree
nodewithoutrecursion
GivenaBinaryTreeandakey,writeafunctionthatprintsalltheancestorsofthekeyinthegivenbinary
tree.
Forexample,considerthefollowingBinaryTree
1
/\
23
/\/\
4567
/\/
8910
Followingaredifferentinputkeysandtheirancestorsintheabovetree
InputKeyListofAncestors
21
31
421
521
631
731
8421
9521
10731
Recursivesolutionforthisproblemisdiscussedhere.
ItisclearthatweneedtouseastackbasediterativetraversaloftheBinaryTree.Theideaistohaveall
ancestorsinstackwhenwereachthenodewithgivenkey.Oncewereachthekey,allwehavetodois,
printcontentsofstack.
Howtogetallancestorsinstackwhenwereachthegivennode?WecantraverseallnodesinPostorder
way.Ifwetakeacloserlookattherecursivepostordertraversal,wecaneasilyobservethat,whenrecursive
functioniscalledforanode,therecursioncallstackcontainsancestorsofthenode.Soideaisdoiterative
Postordertraversalandstopthetraversalwhenwereachthedesirednode.
// C program to print all ancestors of a given key
#include <stdio.h>
#include <stdlib.h>
// Maximum stack size

#define MAX_SIZE 100
// Structure for a tree node

struct Node
{
int data;
};
// Structure for Stack

struct Stack
{
int size;
int top;
struct Node* *array;
};
// A utility function to create a new tree node

{
node->data = data;
return node;
// A utility function to create a stack of given size

struct Stack* createStack(int size)
{
struct Stack* stack = (struct Stack*) malloc(sizeof(struct Stack));
stack->size = size;
stack->top = -1;
stack->array = (struct Node**) malloc(stack->size * sizeof(struct Node*));
return stack;
}
// BASIC OPERATIONS OF STACK

int isFull(struct Stack* stack)
{
return ((stack->top + 1) == stack->size);
}
int isEmpty(struct Stack* stack)
{
return stack->top == -1;
}
void push(struct Stack* stack, struct Node* node)
{
if (isFull(stack))
return;
stack->array[++stack->top] = node;
}
struct Node* pop(struct Stack* stack)
{
if (isEmpty(stack))
return NULL;
return stack->array[stack->top--];
}
struct Node* peek(struct Stack* stack)
{
if (isEmpty(stack))
return NULL;
return stack->array[stack->top];
}
// Iterative Function to print all ancestors of a given key

void printAncestors(struct Node *root, int key)
{
// Create a stack to hold ancestors

struct Stack* stack = createStack(MAX_SIZE);
// Traverse the complete tree in postorder way till we find the key
while (1)
{
// Traverse the left side. While traversing, push the nodes into
// the stack so that their right subtrees can be traversed later
while (root && root->data != key)
{
push(stack, root);
// push current node
root = root->left; // move to next node

}
// If the node whose ancestors are to be printed is found,

// then break the while loop.
if (root && root->data == key)
break;
// Check if right sub-tree exists for the node at top

// If not then pop that node because we don't need this
// node any more.
if (peek(stack)->right == NULL)
{
root = pop(stack);
// If the popped node is right child of top, then remove the top
// as well. Left child of the top must have processed before.
// Consider the following tree for example and key = 3. If we
// remove the following loop, the program will go in an

// infinite loop after reaching 5.
//
//
//
//
\
3
//
//
//
while (!isEmpty(stack) && peek(stack)->right == root)

root = pop(stack);
}
// if stack is not empty then simply set the root as right child
// of top and start traversing right sub-tree.
root = isEmpty(stack)? NULL: peek(stack)->right;
}
// If stack is not empty, print contents of stack

// Here assumption is that the key is there in tree
while (!isEmpty(stack))
printf("%d ", pop(stack)->data);
}

int main()
{
// Let us construct a binary tree
struct Node* root = newNode(1);
printf("Following are all keys and their ancestors\n");

for (int key = 1; key <= 10; key++)
{
printf("%d: ", key);
printAncestors(root, key);
printf("\n");
}
getchar();
return 0;
}
Output:
Followingareallkeysandtheirancestors
1:
2:1
3:1
4:21
5:21
6:31
7:31
8:421
9:521
10:731
Exercise
NotethattheabovesolutionassumesthatthegivenkeyispresentinthegivenBinaryTree.Itmaygoin
infiniteloopifkeyisnotpresent.Extendtheabovesolutiontoworkevenwhenthekeyisnotpresentintree.
=======================================================================
GivenaBinaryTree(BT),convertittoaDoublyLinkedList(DLL).Theleftandrightpointersinnodesareto
beusedaspreviousandnextpointersrespectivelyinconvertedDLL.TheorderofnodesinDLLmustbe
sameasInorderofthegivenBinaryTree.ThefirstnodeofInordertraversal(leftmostnodeinBT)mustbe
headnodeoftheDLL.
Asolutiontothisproblemisdiscussedinthispost.
Inthispost,anothersimpleandefficientsolutionisdiscussed.Thesolutiondiscussedherehastwosimple
steps.
1)FixLeftPointers:Inthisstep,wechangeleftpointerstopointtopreviousnodesinDLL.Theideais
simple,wedoinordertraversaloftree.Ininordertraversal,wekeeptrackofpreviousvisitednodeand
changeleftpointertothepreviousnode.SeefixPrevPtr()inbelowimplementation.
2)FixRightPointers:Theaboveisintuitiveandsimple.Howtochangerightpointerstopointtonextnode
inDLL?Theideaistouseleftpointersfixedinstep1.WestartfromtherightmostnodeinBinaryTree(BT).
TherightmostnodeisthelastnodeinDLL.SinceleftpointersarechangedtopointtopreviousnodeinDLL,
wecanlinearlytraversethecompleteDLLusingthesepointers.Thetraversalwouldbefromlasttofirst
node.WhiletraversingtheDLL,wekeeptrackofthepreviouslyvisitednodeandchangetherightpointerto
thepreviousnode.SeefixNextPtr()inbelowimplementation.
// A simple inorder traversal based program to convert a Binary Tree to DLL
#include<stdio.h>
#include<stdlib.h>
// A tree node
struct node
{
int data;

};

struct node *newNode(int data)
{
struct node *node = (struct node *)malloc(sizeof(struct node));
node->data = data;
return(node);
}
// Standard Inorder traversal of tree

{
if (root != NULL)
{
printf("\t%d",root->data);
}
}
// Changes left pointers to work as previous pointers in converted DLL

// The function simply does inorder traversal of Binary Tree and updates
// left pointer using previously visited node
void fixPrevPtr(struct node *root)
{
static struct node *pre = NULL;
if (root != NULL)
{
fixPrevPtr(root->left);
root->left = pre;
pre = root;
fixPrevPtr(root->right);
}
// Changes right pointers to work as next pointers in converted DLL

struct node *fixNextPtr(struct node *root)
{
struct node *prev = NULL;
// Find the right most node in BT or last node in DLL

while (root && root->right != NULL)
root = root->right;
// Start from the rightmost node, traverse back using left pointers.
// While traversing, change right pointer of nodes.
while (root && root->left != NULL)
{
prev = root;
root = root->left;
root->right = prev;
}
// The leftmost node is head of linked list, return it

return (root);
}
// The main function that converts BST to DLL and returns head of DLL
struct node *BTToDLL(struct node *root)
{
// Set the previous pointer
fixPrevPtr(root);
// Set the next pointer and return head of DLL

return fixNextPtr(root);
}
// Traverses the DLL from left tor right

void printList(struct node *root)
{
while (root != NULL)

{
printf("\t%d", root->data);
root = root->right;
}
}

int main(void)
{
root->left
= newNode(12);
root->right
= newNode(15);
printf("\n\t\tInorder Tree Traversal\n\n");

inorder(root);
struct node *head = BTToDLL(root);
printf("\n\n\t\tDLL Traversal\n\n");
printList(head);
return 0;
}
Output:
InorderTreeTraversal
251230103615
DLLTraversal
251230103615
TimeComplexity:O(n)wherenisthenumberofnodesingivenBinaryTree.Thesolutionsimplydoestwo
traversalsofallBinaryTreenodes.
=======================================================================
CheckforIdenticalBSTswithout
buildingthetrees
Giventwoarrayswhichrepresentasequenceofkeys.ImaginewemakeaBinarySearchTree(BST)from
eacharray.WeneedtotellwhethertwoBSTswillbeidenticalornotwithoutactuallyconstructingthetree.
Examples
Forexample,theinputarraysare{2,4,3,1}and{2,1,4,3}willconstructthesametree
Lettheinputarraysbea[]andb[]
Example1:
a[]={2,4,1,3}willconstructfollowingtree.
2
/\
14
/
3
b[]={2,4,3,1}willalsoalsoconstructthesametree.
2
/\
14
/
3
Sotheoutputis"True"
Example2:
a[]={8,3,6,1,4,7,10,14,13}
b[]={8,10,14,3,6,4,1,7,13}
TheybothconstructthesamefollowingBST,sooutputis"True"
8
/\
310
/\\
1614
/\/
4713
Solution:
AccordingtoBSTproperty,elementsofleftsubtreemustbesmallerandelementsofrightsubtreemustbe
greaterthanroot.
TwoarraysrepresentsaneBSTifforeveryelementx,theelementsinleftandrightsubtreesofxappear
afteritinbotharrays.Andsameistrueforrootsofleftandrightsubtrees.
Theideaistocheckofifnextsmallerandgreaterelementsaresameinbotharrays.Samepropertiesare
recursivelycheckedforleftandrightsubtrees.Theidealookssimple,butimplementationrequireschecking
allconditionsforallelements.Followingisaninterestingrecursiveimplementationoftheidea.
// A C program to check for Identical BSTs without building the trees
#include<stdio.h>
#include<limits.h>
#include<stdbool.h>
/* The main function that checks if two arrays a[] and b[] of size n construct
same BST. The two values 'min' and 'max' decide whether the call is made for
left subtree or right subtree of a parent element. The indexes i1 and i2 are
the indexes in (a[] and b[]) after which we search the left or right child.
Initially, the call is made for INT_MIN and INT_MAX as 'min' and 'max'
respectively, because root has no parent.
i1 and i2 are just after the indexes of the parent element in a[] and b[]. */
bool isSameBSTUtil(int a[], int b[], int n, int i1, int i2, int min, int max)
{
int j, k;
/* Search for a value satisfying the constraints of min and max in a[] and
b[]. If the parent element is a leaf node then there must be some
elements in a[] and b[] satisfying constraint. */
for (j=i1; j<n; j++)
if (a[j]>min && a[j]<max)
break;
for (k=i2; k<n; k++)
if (b[k]>min && b[k]<max)
break;
/* If the parent element is leaf in both arrays */

if (j==n && k==n)
return true;
/* Return false if any of the following is true

a) If the parent element is leaf in one array, but non-leaf in other.
b) The elements satisfying constraints are not same. We either search
for left child or right child of the parent element (decinded by min
and max values). The child found must be same in both arrays */
if (((j==n)^(k==n)) || a[j]!=b[k])
return false;
/* Make the current child as parent and recursively check for left and right
subtrees of it. Note that we can also pass a[k] in place of a[j] as they
are both are same */
return isSameBSTUtil(a, b, n, j+1, k+1, a[j], max) && // Right Subtree
isSameBSTUtil(a, b, n, j+1, k+1, min, a[j]);
}
// A wrapper over isSameBSTUtil()

bool isSameBST(int a[], int b[], int n)
{
return isSameBSTUtil(a, b, n, 0, 0, INT_MIN, INT_MAX);
}

int main()
{
int a[] = {8, 3, 6, 1, 4, 7, 10, 14, 13};
int b[] = {8, 10, 14, 3, 6, 4, 1, 7, 13};
int n=sizeof(a)/sizeof(a[0]);
printf("%s\n", isSameBST(a, b, n)?
"BSTs are same":"BSTs not same");
return 0;
}
Output:
// Left Subtree
BSTsaresame
=======================================================================
CustomTreeProblem
Youaregivenasetoflinks,e.g.
a>b
b>c
b>d
a>e
Printthetreethatwouldformwheneachpairoftheselinksthathasthesamecharacterasstartandend
pointisjoinedtogether.Youhavetomaintainfidelityw.r.t.theheightofnodes,i.e.nodesatheightnfrom
rootshouldbeprintedatsameroworcolumn.Forsetoflinksgivenabove,treeprintedshouldbe
>a
|>b
||>c
||>d
|>e
Notethattheselinksneednotformasingletreetheycouldform,ahem,aforest.Considerthefollowing
links
a>b
a>g
b>c
c>d
d>e
c>f
z>y
y>x
x>w
Theoutputwouldbefollowingforest.
>a
|>b
||>c
|||>d
||||>e
|||>f
|>g
>z
|>y
||>x
|||>w
Youcanassumethatgivenlinkscanformatreeorforestoftreesonly,andtherearenoduplicatesamong
links.
Solution:Theideaistomaintaintwoarrays,onearrayfortreenodesandotherfortreesthemselves(we
callthisarrayforest).AnelementofthenodearraycontainstheTreeNodeobjectthatcorrespondsto
respectivecharacter.AnelementoftheforestarraycontainsTreeobjectthatcorrespondstorespectiveroot
oftree.
Itshouldbeobviousthatthecrucialpartiscreatingtheforesthere,onceitiscreated,printingitoutin
requiredformatisstraightforward.Tocreatetheforest,followingprocedureisused
Dofollowingforeachinputlink,
1.Ifstartoflinkisnotpresentinnodearray
CreateTreeNodeobjectsforstartcharacter
Addentriesofstartinbotharrays.
2.Ifendoflinkisnotpresentinnodearray
CreateTreeNodeobjectsforstartcharacter
Addentryofendinnodearray.
3.Ifendoflinkispresentinnodearray.
Ifendoflinkispresentinforestarray,thenremoveit
fromthere.
4.Addanedge(intree)betweenstartandendnodesoflink.
Itshouldbeclearthatthisprocedurerunsinlineartimeinnumberofnodesaswellasoflinksitmakes
onlyonepassoverthelinks.Italsorequireslinearspaceintermsofalphabetsize.
FollowingisJavaimplementationofabovealgorithm.Inthefollowingimplementationcharactersare
assumedtobeonlylowercasecharactersfromatoz.
// Java program to create a custom tree from a given set of links.
// The main class that represents tree and has main method
public class Tree {
private TreeNode root;
/* Returns an array of trees from links input. Links are assumed to

be Strings of the form "<s> <e>" where <s> and <e> are starting
and ending points for the link. The returned array is of size 26
and has non-null values at indexes corresponding to roots of trees
in output */
public Tree[] buildFromLinks(String [] links) {
// Create two arrays for nodes and forest

TreeNode[] nodes = new TreeNode[26];
Tree[] forest = new Tree[26];
// Process each link

for (String link : links) {
// Find the two ends of current link

String[] ends = link.split(" ");
int start = (int) (ends[0].charAt(0) - 'a'); // Start node
int end
= (int) (ends[1].charAt(0) - 'a'); // End node
// If start of link not seen before, add it two both arrays

if (nodes[start] == null)
{
nodes[start] = new TreeNode((char) (start + 'a'));
// Note that it may be removed later when this character is

// last character of a link. For example, let we first see
// a--->b, then c--->a. We first add 'a' to array of trees
// and when we see link c--->a, we remove it from trees array.
forest[start] = new Tree(nodes[start]);
// If end of link is not seen before, add it to the nodes array

if (nodes[end] == null)
nodes[end] = new TreeNode((char) (end + 'a'));
// If end of link is seen before, remove it from forest if

// it exists there.
else forest[end] = null;
// Establish Parent-Child Relationship between Start and End

nodes[start].addChild(nodes[end], end);
}
return forest;
}
// Constructor
public Tree(TreeNode root) { this.root = root; }
public static void printForest(String[] links)

{
Tree t = new Tree(new TreeNode('\0'));
for (Tree t1 : t.buildFromLinks(links)) {
if (t1 != null)
{
t1.root.printTreeIdented("");
System.out.println("");
}
}
}
// Driver method to test

public static void main(String[] args) {
String [] links1 = {"a b", "b c", "b d", "a e"};
System.out.println("------------ Forest 1 ----------------");
printForest(links1);
String [] links2 = {"a b", "a g", "b c", "c d", "d e", "c f",
"z y", "y x", "x w"};
System.out.println("------------ Forest 2 ----------------");
printForest(links2);
}
// Class to represent a tree node

class TreeNode {
TreeNode []children;
char c;
// Adds a child 'n' to this node

public void addChild(TreeNode n, int index) { this.children[index] = n;}
// Constructor
public TreeNode(char c) { this.c = c; this.children = new TreeNode[26];}
// Recursive method to print indented tree rooted with this node.

public void printTreeIdented(String indent) {
System.out.println(indent + "-->" + c);
for (TreeNode child : children) {
if (child != null)
child.printTreeIdented(indent + "
}
}
}
Output:
Forest1
>a
|>b
||>c
||>d
|>e
Forest2
>a
|>b
||>c
|||>d
||||>e
|||>f
|");
|>g
>z
|>y
||>x
|||>w
=======================================================================
IterativeMethodtofindHeightof
BinaryTree
TherearetwoconventionstodefineheightofBinaryTree
1)Numberofnodesonlongestpathfromroottothedeepestnode.
2)Numberofedgesonlongestpathfromroottothedeepestnode.
Inthispost,thefirstconventionisfollowed.Forexample,heightofthebelowtreeis3.
ExampleTree
RecursivemethodtofindheightofBinaryTreeisdiscussedhere.Howtofindheightwithoutrecursion?We
canuselevelordertraversaltofindheightwithoutrecursion.Theideaistotraverselevelbylevel.Whenever
movedowntoalevel,incrementheightby1(heightisinitializedas0).Countnumberofnodesateachlevel,
stoptraversingwhencountofnodesatnextlevelis0.
Followingisdetailedalgorithmtofindlevelordertraversalusingqueue.
Createaqueue.
Pushrootintothequeue.
height=0
Loop
nodeCount=sizeofqueue
//Ifnumberofnodesatthislevelis0,returnheight
ifnodeCountis0
returnHeight
else
increaseHeight
//Removenodesofthislevelandaddnodesof
//nextlevel
while(nodeCount>0)
popnodefromfront
pushitschildrentoqueue
decreasenodeCount
//Atthispoint,queuehasnodesofnextlevel
/* Program to find height of the tree by Iterative Method */
#include <iostream>
#include <queue>

struct node
{
struct node *left;
int data;
struct node *right;
};
// Iterative method to find height of Bianry Tree

int treeHeight(node *root)
{
// Base Case
if (root == NULL)
return 0;
// Create an empty queue for level order tarversal

queue<node *> q;
// Enqueue Root and initialize height

q.push(root);
int height = 0;
while (1)
{
// nodeCount (queue size) indicates number of nodes
// at current lelvel.
int nodeCount = q.size();
if (nodeCount == 0)
return height;
height++;
// Dequeue all nodes of current level and Enqueue all

// nodes of next level
while (nodeCount > 0)
{
node *node = q.front();
q.pop();
q.push(node->left);
q.push(node->right);
nodeCount--;
}
}
}

{
temp->data = data;
temp->left = NULL;
temp->right = NULL;
return temp;
}

int main()
{
// Let us create binary tree shown in above diagram
cout << "Height of tree is " << treeHeight(root);

return 0;
}
Output:
Heightoftreeis3
TimeComplexity:O(n)wherenisnumberofnodesingivenbinarytree.
=======================================================================
Findallpossibleinterpretationsofan
arrayofdigits
Consideracodingsystemforalphabetstointegerswhereaisrepresentedas1,bas2,..zas26.Given
anarrayofdigits(1to9)asinput,writeafunctionthatprintsallvalidinterpretationsofinputarray.
Examples
Input:{1,1}
Output:("aa",'k")
[2interpretations:aa(1,1),k(11)]
Input:{1,2,1}
Output:("aba","au","la")
[3interpretations:aba(1,2,1),au(1,21),la(12,1)]
Input:{9,1,8}
Output:{"iah","ir"}
[2interpretations:iah(9,1,8),ir(9,18)]
Pleasenotewecannotchangeorderofarray.Thatmeans{1,2,1}cannotbecome{2,1,1}
Onfirstlookitlookslikeaproblemofpermutation/combination.Butoncloserlookyouwillfigureoutthatthis
isaninterestingtreeproblem.
Theideahereisstringcantakeatmosttwopaths:
1.Processingledigit
2.Processtwodigits
Thatmeanswecanusebinarytreehere.Processingwithsingledigitwillbeleftchildandtwodigitswillbe
rightchild.Ifvaluetwodigitsisgreaterthan26thenourrightchildwillbenullaswedonthavealphabetfor
greaterthan26.
Letsunderstandwithanexample.Arraya={1,2,1}.Belowdiagramshowsthathowourtreegrows.
{1,2,1}Codesusedintree
/\"a">1
/\"b">2
"a"{2,1}"l"{1}"l">12
/\/\
/\/\
"ab"{1}"au""la"null
/\
/\
"aba"null
Braces{}containarraystillpendingforprocessing.Notethatwitheverylevel,ourarraysizedecreases.If
youwillseecarefully,itisnothardtofindthattreeheightisalwaysn(arraysize)
Howtoprintallstrings(interpretations)?Outputstringsareleafnodeoftree.i.efor{1,2,1},outputis{abaau
la}.
Wecanconcludethattherearemainlytwostepstoprintallinterpretationsofgivenintegerarray.
Step1:Createabinarytreewithallpossibleinterpretationsinleafnodes.
Step2:Printallleafnodesfromthebinarytreecreatedinstep1.
FollowingisJavaimplementationofabovealgorithm.
// A Java program to print all interpretations of an integer array

import java.util.Arrays;

class Node {
String dataString;
Node left;
Node right;
Node(String dataString) {
this.dataString = dataString;
//Be default left and right child are null.
}
public String getDataString() {

return dataString;
}
}
public class arrayToAllInterpretations {
// Method to create a binary tree which stores all interpretations

// of arr[] in lead nodes
public static Node createTree(int data, String pString, int[] arr) {
// Invalid input as alphabets maps from 1 to 26

if (data > 26)
return null;
// Parent String + String for this node

String dataToStr = pString + alphabet[data];
Node root = new Node(dataToStr);
// if arr.length is 0 means we are done

if (arr.length != 0) {
data = arr[0];
// new array will be from index 1 to end as we are consuming

// first index with this node
int newArr[] = Arrays.copyOfRange(arr, 1, arr.length);
// left child
root.left = createTree(data, dataToStr, newArr);
// right child will be null if size of array is 0 or 1

if (arr.length > 1) {
data = arr[0] * 10 + arr[1];
// new array will be from index 2 to end as we

// are consuming first two index with this node
newArr = Arrays.copyOfRange(arr, 2, arr.length);
root.right = createTree(data, dataToStr, newArr);

}
}
return root;
}
// To print out leaf nodes

public static void printleaf(Node root) {
if (root == null)
return;
if (root.left == null && root.right == null)

System.out.print(root.getDataString() + " ");
printleaf(root.left);
printleaf(root.right);
}
// The main function that prints all interpretations of array
static void printAllInterpretations(int[] arr) {
// Step 1: Create Tree

Node root = createTree(0, "", arr);
// Step 2: Print Leaf nodes

printleaf(root);
System.out.println(); // Print new line

}
// For simplicity I am taking it as string array. Char Array will save space
private static final String[] alphabet = {"", "a", "b", "c", "d", "e",
"f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r",
"s", "t", "u", "v", "w", "x", "v", "z"};
// Driver method to test above methods

public static void main(String args[]) {
// aacd(1,1,3,4) amd(1,13,4) kcd(11,3,4)

// Note : 1,1,34 is not valid as we don't have values corresponding
// to 34 in alphabet
int[] arr = {1, 1, 3, 4};
printAllInterpretations(arr);
// aaa(1,1,1) ak(1,11) ka(11,1)

int[] arr2 = {1, 1, 1};
printAllInterpretations(arr2);
// bf(2,6) z(26)
int[] arr3 = {2, 6};
// ab(1,2), l(12)
int[] arr4 = {1, 2};
// a(1,0} j(10)
int[] arr5 = {1, 0};
// "" empty string output as array is empty

int[] arr6 = {};
// abba abu ava lba lu

int[] arr7 = {1, 2, 2, 1};
}
}
Output:
aacdamdkcd
aaaakka
bfz
abl
aj
abbaabuavalbalu
=======================================================================
TreeIsomorphismProblem
Writeafunctiontodetectiftwotreesareisomorphic.Twotreesarecalledisomorphicifoneofthemcanbe
obtainedfromotherbyaseriesofflips,i.e.byswappingleftandrightchildrenofanumberofnodes.Any
numberofnodesatanylevelcanhavetheirchildrenswapped.Twoemptytreesareisomorphic.
Forexample,followingtwotreesareisomorphicwithfollowingsubtreesflipped:2and3,NULLand6,7and
8.

Wesimultaneouslytraversebothtrees.Letthecurrentinternalnodesoftwotreesbeingtraversedben1and
n2respectively.Therearefollowingtwoconditionsforsubtreesrootedwithn1andn2tobeisomorphic.
1)Dataofn1andn2issame.
2)Oneofthefollowingtwoistrueforchildrenofn1andn2
a)Leftchildofn1isisomorphictoleftchildofn2andrightchildofn1isisomorphictorightchildofn2.
b)Leftchildofn1isisomorphictorightchildofn2andrightchildofn1isisomorphictoleftchildofn2.
// A C++ program to check if two given trees are isomorphic
#include <iostream>
/* A binary tree node has data, pointer to left and right children */
struct node
{
int data;
struct node* left;
struct node* right;
};
/* Given a binary tree, print its nodes in reverse level order */

bool isIsomorphic(node* n1, node *n2)
{
// Both roots are NULL, trees isomorphic by definition
if (n1 == NULL && n2 == NULL)
return true;
// Exactly one of the n1 and n2 is NULL, trees not isomorphic

if (n1 == NULL || n2 == NULL)
return false;
if (n1->data != n2->data)
return false;
// There are two possible cases for n1 and n2 to be isomorphic

// Case 1: The subtrees rooted at these nodes have NOT been "Flipped".
// Both of these subtrees have to be isomorphic, hence the &&
// Case 2: The subtrees rooted at these nodes have been "Flipped"
return
(isIsomorphic(n1->left,n2->left) && isIsomorphic(n1->right,n2->right))||
(isIsomorphic(n1->left,n2->right) && isIsomorphic(n1->right,n2->left));
}

{
node* temp = new node;
temp->data = data;
temp->left = NULL;
temp->right = NULL;
return (temp);
}

int main()
{
// Let us create trees shown in above diagram
struct node *n1 = newNode(1);
n1->left
= newNode(2);
n1->right
= newNode(3);
n1->left->left = newNode(4);
n1->left->right = newNode(5);
n1->right->left = newNode(6);
n1->left->right->left = newNode(7);
n1->left->right->right = newNode(8);
struct node *n2 = newNode(1);

n2->left
= newNode(3);
n2->right
= newNode(2);
n2->right->left
n2->right->right
n2->left->right
= newNode(4);
= newNode(5);
= newNode(6);
n2->right->right->left = newNode(8);
n2->right->right->right = newNode(7);
if (isIsomorphic(n1, n2) == true)

cout << "Yes";
else
cout << "No";
return 0;
}
Output:
Yes
TimeComplexity:Theabovesolutiondoesatraversalofbothtrees.SotimecomplexityisO(m+n)where
mandnarenumberofnodesingiventrees.
=======================================================================
LongestprefixmatchingATrie
basedsolutioninJava
Givenadictionaryofwordsandaninputstring,findthelongestprefixofthestringwhichisalsoawordin
dictionary.
Examples:
Letthedictionarycontainsthefollowingwords:
{are,area,base,cat,cater,children,basement}
Belowaresomeinput/outputexamples:
InputStringOutput
caterercater
basemexybase
child<Empty>
Solution
WebuildaTrieofalldictionarywords.OncetheTrieisbuilt,traversethroughitusingcharactersofinput
string.Ifprefixmatchesadictionaryword,storecurrentlengthandlookforalongermatch.Finally,return
thelongestmatch.
FollowingisJavaimplementationoftheabovesolutionbased.
import java.util.HashMap;
// Trie Node, which stores a character and the children in a HashMap

class TrieNode {
public TrieNode(char ch) {
value = ch;
children = new HashMap<>();
bIsEnd = false;
}
public HashMap<Character,TrieNode> getChildren() {
return children; }
public char getValue()
return value;
public void setIsEnd(boolean val)
bIsEnd = val;
public boolean isEnd()
return bIsEnd; }
private char value;

private HashMap<Character,TrieNode> children;
private boolean bIsEnd;
}
// Implements the actual Trie

class Trie {
// Constructor
public Trie()
root = new TrieNode((char)0);
// Method to insert a new word to Trie
public void insert(String word) {
// Find length of the given word

int length = word.length();
TrieNode crawl = root;
// Traverse through all characters of given word

for( int level = 0; level < length; level++)
{
HashMap<Character,TrieNode> child = crawl.getChildren();
char ch = word.charAt(level);
// If there is already a child for current character of given word

if( child.containsKey(ch))
crawl = child.get(ch);
else
// Else create a child
{
TrieNode temp = new TrieNode(ch);
child.put( ch, temp );
crawl = temp;
}
}
// Set bIsEnd true for last character

crawl.setIsEnd(true);
}
// The main method that finds out the longest string 'input'
public String getMatchingPrefix(String input) {
String result = ""; // Initialize resultant string
int length = input.length(); // Find length of the input string
// Initialize reference to traverse through Trie

TrieNode crawl = root;
// Iterate through all characters of input string 'str' and traverse

// down the Trie
int level, prevMatch = 0;

for( level = 0 ; level < length; level++ )
{
// Find current character of str
char ch = input.charAt(level);
// HashMap of current Trie node to traverse down

HashMap<Character,TrieNode> child = crawl.getChildren();
// See if there is a Trie edge for the current character

if( child.containsKey(ch) )
{
result += ch;
//Update result
crawl = child.get(ch); //Update crawl to move down in Trie
// If this is end of a word, then update prevMatch

if( crawl.isEnd() )
prevMatch = level + 1;
}
else break;
}
// If the last processed character did not match end of a word,

// return the previously matching prefix
if( !crawl.isEnd() )
return result.substring(0, prevMatch);
else return result;

}
private TrieNode root;

}
// Testing class
public class Test {
Trie dict = new Trie();

dict.insert("are");
dict.insert("area");
dict.insert("base");
dict.insert("cat");
dict.insert("cater");
dict.insert("basement");
String input = "caterer";

System.out.print(input + ":
");
System.out.println(dict.getMatchingPrefix(input));
input = "basement";
");
input = "are";
");
input = "arex";
");
input = "basemexz";
");
input = "xyz";
");
}
}
Output:
caterer:cater
basement:basement
are:are
arex:are
basemexz:base
xyz:
TimeComplexity:TimecomplexityoffindingthelongestprefixisO(n)wherenislengthoftheinputstring.
ReferthisfortimecomplexityofbuildingtheTrie.
=======================================================================
BTree|Set2(Insert)
Inthepreviouspost,weintroducedBTree.Wealsodiscussedsearch()andtraverse()functions.
Inthispost,insert()operationisdiscussed.Anewkeyisalwaysinsertedatleafnode.Letthekeytobe
insertedbek.LikeBST,westartfromrootandtraversedowntillwereachaleafnode.Oncewereachaleaf
node,weinsertthekeyinthatleafnode.UnlikeBSTs,wehaveapredefinedrangeonnumberofkeysthata
nodecancontain.Sobeforeinsertingakeytonode,wemakesurethatthenodehasextraspace.
Howtomakesurethatanodehasspaceavailableforkeybeforethekeyisinserted?Weuseanoperation
calledsplitChild()thatisusedtosplitachildofanode.Seethefollowingdiagramtounderstandsplit.Inthe
followingdiagram,childyofxisbeingsplitintotwonodesyandz.NotethatthesplitChildoperationmoves
akeyupandthisisthereasonBTreesgrowupunlikeBSTswhichgrowdown.
Asdiscussedabove,toinsertanewkey,wegodownfromroottoleaf.Beforetraversingdowntoanode,
wefirstcheckifthenodeisfull.Ifthenodeisfull,wesplitittocreatespace.Followingiscompletealgorithm.
Insertion
1)Initializexasroot.
2)Whilexisnotleaf,dofollowing
..a)Findthechildofxthatisgoingtotobetraversednext.Letthechildbey.
..b)Ifyisnotfull,changextopointtoy.
..c)Ifyisfull,splititandchangextopointtooneofthetwopartsofy.Ifkissmallerthanmidkeyiny,then
setxasfirstpartofy.Elsesecondpartofy.Whenwesplity,wemoveakeyfromytoitsparentx.
3)Theloopinstep2stopswhenxisleaf.xmusthavespacefor1extrakeyaswehavebeensplittingall
nodesinadvance.Sosimplyinsertktox.
NotethatthealgorithmfollowstheCormenbook.Itisactuallyaproactiveinsertionalgorithmwherebefore
goingdowntoanode,wesplititifitisfull.Theadvantageofsplittingbeforeis,wenevertraverseanode
twice.Ifwedontsplitanodebeforegoingdowntoitandsplititonlyifnewkeyisinserted(reactive),we
mayenduptraversingallnodesagainfromleaftoroot.Thishappensincaseswhenallnodesonthepath
fromroottoleafarefull.Sowhenwecometotheleafnode,wesplititandmoveakeyup.Movingakeyup
willcauseasplitinparentnode(becauseparentwasalreadyfull).Thiscascadingeffectneverhappensin
thisproactiveinsertionalgorithm.Thereisadisadvantageofthisproactiveinsertionthough,wemaydo
unnecessarysplits.
Letusunderstandthealgorithmwithanexampletreeofminimumdegreetas3andasequenceofintegers
10,20,30,40,50,60,70,80and90inaninitiallyemptyBTree.
InitiallyrootisNULL.Letusfirstinsert10.
Letusnowinsert20,30,40and50.Theyallwillbeinsertedinrootbecausemaximumnumberofkeysa
nodecanaccommodateis2*t1whichis5.
Letusnowinsert60.Sincerootnodeisfull,itwillfirstsplitintotwo,then60willbeinsertedintothe
appropriatechild.

Letusnowinsert70and80.Thesenewkeyswillbeinsertedintotheappropriateleafwithoutanysplit.
Letusnowinsert90.Thisinsertionwillcauseasplit.Themiddlekeywillgouptotheparent.
Seethisformoreexamples.
FollowingisC++implementationoftheaboveproactivealgorithm.
// C++ program for B-Tree insertion
#include<iostream>
// A BTree node
class BTreeNode
{
int t;

int n;

public:
// Constructor
// A utility function to insert a new key in the subtree rooted with

// this node. The assumption is, the node must be non-full when this
void insertNonFull(int k);
// A utility function to split the child y of this node. i is index of y in

// child array C[]. The Child y must be full when this function is called
void splitChild(int i, BTreeNode *y);

void traverse();

// Make BTree friend of this so that we can access private members of this
// class in BTree functions
friend class BTree;
};
// A BTree
class BTree
{
public:
BTree(int _t)
{ root = NULL; t = _t; }
// function to traverse the tree

void traverse()
{ if (root != NULL) root->traverse(); }

{ return (root == NULL)? NULL : root->search(k); }

void insert(int k);
};
// Constructor for BTreeNode class

BTreeNode::BTreeNode(int t1, bool leaf1)
{
t = t1;
leaf = leaf1;


n = 0;
}

{
int i;
for (i = 0; i < n; i++)
{

if (leaf == false)
C[i]->traverse();
cout << " " << keys[i];
}

if (leaf == false)
C[i]->traverse();
}

{
int i = 0;
i++;

if (keys[i] == k)
return this;

if (leaf == true)
return NULL;

}

void BTree::insert(int k)
{
// If tree is empty
if (root == NULL)
{
// Allocate memory for root

root = new BTreeNode(t, true);
root->keys[0] = k; // Insert key
root->n = 1; // Update number of keys in root
}
else // If tree is not empty
{
// If root is full, then tree grows in height
if (root->n == 2*t-1)
{
// Allocate memory for new root
BTreeNode *s = new BTreeNode(t, false);
// Make old root as child of new root

s->C[0] = root;
// Split the old root and move 1 key to the new root
s->splitChild(0, root);
// New root has two children now. Decide which of the

// two children is going to have new key
int i = 0;
if (s->keys[0] < k)
i++;
s->C[i]->insertNonFull(k);
// Change root
root = s;
}
else // If root is not full, call insertNonFull for root
root->insertNonFull(k);
}
}
// A utility function to insert a new key in this node

// The assumption is, the node must be non-full when this
void BTreeNode::insertNonFull(int k)
{
// Initialize index as index of rightmost element
int i = n-1;
// If this is a leaf node

if (leaf == true)
{
// The following loop does two things
// a) Finds the location of new key to be inserted
// b) Moves all greater keys to one place ahead
{
keys[i+1] = keys[i];
i--;
}
// Insert the new key at found location

keys[i+1] = k;
n = n+1;
}
else // If this node is not leaf
{
// Find the child which is going to have the new key
i--;
// See if the found child is full

if (C[i+1]->n == 2*t-1)
{
// If the child is full, then split it
splitChild(i+1, C[i+1]);
// After split, the middle key of C[i] goes up and

// C[i] is splitted into two. See which of the two
// is going to have the new key
if (keys[i+1] < k)
i++;
}
C[i+1]->insertNonFull(k);
}
}
// A utility function to split the child y of this node

// Note that y must be full when this function is called
void BTreeNode::splitChild(int i, BTreeNode *y)
{
// Create a new node which is going to store (t-1) keys
// of y
BTreeNode *z = new BTreeNode(y->t, y->leaf);
z->n = t - 1;
// Copy the last (t-1) keys of y to z

for (int j = 0; j < t-1; j++)
z->keys[j] = y->keys[j+t];
// Copy the last t children of y to z

if (y->leaf == false)
{
for (int j = 0; j < t; j++)
z->C[j] = y->C[j+t];
}
// Reduce the number of keys in y

y->n = t - 1;
// Since this node is going to have a new child,

// create space of new child
for (int j = n; j >= i+1; j--)
C[j+1] = C[j];
// Link the new child to this node

C[i+1] = z;
// A key of y will move to this node. Find location of

// new key and move all greater keys one space ahead
for (int j = n-1; j >= i; j--)
keys[j+1] = keys[j];
// Copy the middle key of y to this node

keys[i] = y->keys[t-1];
// Increment count of keys in this node

n = n + 1;
}

int main()
{
BTree t(3); // A B-Tree with minium degree 3
t.insert(10);
t.insert(20);
t.insert(5);
t.insert(6);
t.insert(12);
t.insert(30);
t.insert(7);
t.insert(17);
cout << "Traversal of the constucted tree is ";

t.traverse();
int k = 6;
(t.search(k) != NULL)? cout << "\nPresent" : cout << "\nNot Present";
k = 15;
(t.search(k) != NULL)? cout << "\nPresent" : cout << "\nNot Present";
return 0;
}
Output:
Traversaloftheconstuctedtreeis5671012172030
Present
NotPresent
=======================================================================
BTree|Set1(Introduction)
BTreeisaselfbalancingsearchtree.Inmostoftheotherselfbalancingsearchtrees(likeAVLandRed
BlackTrees),itisassumedthateverythingisinmainmemory.TounderstanduseofBTrees,wemustthink
ofhugeamountofdatathatcannotfitinmainmemory.Whenthenumberofkeysishigh,thedataisread
fromdiskintheformofblocks.Diskaccesstimeisveryhighcomparedtomainmemoryaccesstime.The
mainideaofusingBTreesistoreducethenumberofdiskaccesses.Mostofthetreeoperations(search,
insert,delete,max,min,..etc)requireO(h)diskaccesseswherehisheightofthetree.Btreeisafattree.
HeightofBTreesiskeptlowbyputtingmaximumpossiblekeysinaBTreenode.Generally,aBTreenode
sizeiskeptequaltothediskblocksize.SincehislowforBTree,totaldiskaccessesformostofthe
operationsarereducedsignificantlycomparedtobalancedBinarySearchTreeslikeAVLTree,RedBlack
Tree,..etc.
PropertiesofBTree
1)Allleavesareatsamelevel.
2)ABTreeisdefinedbythetermminimumdegreet.Thevalueoftdependsupondiskblocksize.
3)Everynodeexceptrootmustcontainatleastt1keys.Rootmaycontainminimum1key.
4)Allnodes(includingroot)maycontainatmost2t1keys.
5)Numberofchildrenofanodeisequaltothenumberofkeysinitplus1.
6)Allkeysofanodearesortedinincreasingorder.Thechildbetweentwokeysk1andk2containsallkeys
inrangefromk1andk2.
7)BTreegrowsandshrinksfromrootwhichisunlikeBinarySearchTree.BinarySearchTreesgrow
downwardandalsoshrinkfromdownward.
8)LikeotherbalancedBinarySearchTrees,timecomplexitytosearch,insertanddeleteisO(Logn).
FollowingisanexampleBTreeofminimumdegree3.NotethatinpracticalBTrees,thevalueofminimum
degreeismuchmorethan3.

Search
SearchissimilartosearchinBinarySearchTree.Letthekeytobesearchedbek.Westartfromrootand
recursivelytraversedown.Foreveryvisitednonleafnode,ifthenodehaskey,wesimplyreturnthenode.
Otherwisewerecurdowntotheappropriatechild(Thechildwhichisjustbeforethefirstgreaterkey)ofthe
node.Ifwereachaleafnodeanddontfindkintheleafnode,wereturnNULL.
Traverse
TraversalisalsosimilartoInordertraversalofBinaryTree.Westartfromtheleftmostchild,recursivelyprint
theleftmostchild,thenrepeatthesameprocessforremainingchildrenandkeys.Intheend,recursively
printtherightmostchild.
// C++ implemntation of search() and traverse() methods
#include<iostream>
// A BTree node
class BTreeNode
{
int t;

int n;

public:
// Constructor

void traverse();

// Make BTree friend of this so that we can access private members of this
// class in BTree functions
friend class BTree;
};
// A BTree
class BTree
{
public:
BTree(int _t)
{ root = NULL; t = _t; }
// function to traverse the tree

void traverse()
{ if (root != NULL) root->traverse(); }

{ return (root == NULL)? NULL : root->search(k); }
};
// Constructor for BTreeNode class

BTreeNode::BTreeNode(int _t, bool _leaf)
{
t = _t;
leaf = _leaf;


n = 0;
}

{
int i;
for (i = 0; i < n; i++)
{
if (leaf == false)
C[i]->traverse();
cout << " " << keys[i];
}

if (leaf == false)
C[i]->traverse();
}

{
int i = 0;
i++;

if (keys[i] == k)
return this;

if (leaf == true)
return NULL;

}
Theabovecodedoesntcontaindriverprogram.Wewillbecoveringthecompleteprograminournextpost
onBTreeInsertion.
TherearetwoconventionstodefineaBTree,oneistodefinebyminimumdegree(followedinCormen
book),secondisdefinebyorder.Wehavefollowedtheminimumdegreeconventionandwillbefollowing
sameincomingpostsonBTree.Thevariablenamesusedintheaboveprogramarealsokeptsameas
Cormenbookforbetterreadability.
InsertionandDeletion
BTrerInsertion
BTreeDeletion
=======================================================================
GivenaBinaryTree(Bt),convertittoaDoublyLinkedList(DLL).Theleftandrightpointersinnodesareto
beusedaspreviousandnextpointersrespectivelyinconvertedDLL.TheorderofnodesinDLLmustbe
sameasInorderofthegivenBinaryTree.ThefirstnodeofInordertraversal(leftmostnodeinBT)mustbe
headnodeoftheDLL.

Icameacrossthisinterviewduringoneofmyinterviews.Asimilarproblemisdiscussedinthispost.The
problemhereissimpleraswedontneedtocreatecircularDLL,butasimpleDLL.Theideabehindits
solutionisquitesimpleandstraight.
1.Ifleftsubtreeexists,processtheleftsubtree
..1.a)RecursivelyconverttheleftsubtreetoDLL.
..1.b)Thenfindinorderpredecessorofrootinleftsubtree(inorderpredecessorisrightmostnodeinleft
subtree).
..1.c)Makeinorderpredecessoraspreviousofrootandrootasnextofinorderpredecessor.
2.Ifrightsubtreeexists,processtherightsubtree(Below3stepsaresimilartoleftsubtree).
..2.a)RecursivelyconverttherightsubtreetoDLL.
..2.b)Thenfindinordersuccessorofrootinrightsubtree(inordersuccessorisleftmostnodeinright
subtree).
..2.c)Makeinordersuccessorasnextofrootandrootaspreviousofinordersuccessor.
3.Findtheleftmostnodeandreturnit(theleftmostnodeisalwaysheadofconvertedDLL).
Belowisthesourcecodeforabovealgorithm.
// A C++ program for in-place conversion of Binary Tree to DLL
#include <stdio.h>
/* A binary tree node has data, and left and right pointers */
struct node
{
int data;
node* left;
node* right;
};
/* This is the core function to convert Tree to list. This function follows
steps 1 and 2 of the above algorithm */
node* bintree2listUtil(node* root)
{
// Base case
if (root == NULL)
return root;
// Convert the left subtree and link to root

{
// Convert the left subtree
node* left = bintree2listUtil(root->left);
// Find inorder predecessor. After this loop, left

// will point to the inorder predecessor
for (; left->right!=NULL; left=left->right);
// Make root as next of the predecessor

left->right = root;
// Make predecssor as previous of root

root->left = left;
}
// Convert the right subtree and link to root

if (root->right!=NULL)
{
// Convert the right subtree
node* right = bintree2listUtil(root->right);
// Find inorder successor. After this loop, right

// will point to the inorder successor
for (; right->left!=NULL; right = right->left);
// Make root as previous of successor

right->left = root;
// Make successor as next of root

root->right = right;
}
return root;
}
// The main function that first calls bintree2listUtil(), then follows step 3
// of the above algorithm
node* bintree2list(node *root)
{
// Base case
if (root == NULL)
return root;
// Convert to DLL using bintree2listUtil()

root = bintree2listUtil(root);
// bintree2listUtil() returns root node of the converted

// DLL. We need pointer to the leftmost node which is
// head of the constructed DLL, so move to the leftmost node
while (root->left != NULL)
root = root->left;
return (root);
}

{
node* new_node = new node;
new_node->data = data;
new_node->left = new_node->right = NULL;
return (new_node);
}
/* Function to print nodes in a given doubly linked list */

void printList(node *node)
{
while (node!=NULL)
{
printf("%d ", node->data);
node = node->right;
}
}

int main()
{
node *root
= newNode(10);
root->left
= newNode(12);
root->right
= newNode(15);
// Convert to DLL
node *head = bintree2list(root);
// Print the converted list

printList(head);
return 0;
}
Output:
251230103615
=======================================================================
RemoveBSTkeysoutsidethegiven
range
GivenaBinarySearchTree(BST)andarange[min,max],removeallkeyswhichareoutsidethegiven
range.ThemodifiedtreeshouldalsobeBST.Forexample,considerthefollowingBSTandrange[10,13].
Thegiventreeshouldbechangedtofollowing.Notethatallkeysoutsidetherange[10,13]areremoved
andmodifiedtreeisBST.
Therearetwopossiblecasesforeverynode.
1)Nodeskeyisoutsidethegivenrange.Thiscasehastwosubcases.
.a)Nodeskeyissmallerthantheminvalue.
.b)Nodeskeyisgreaterthatthemaxvalue.
2)Nodeskeyisinrange.
Wedontneedtodoanythingforcase2.Incase1,weneedtoremovethenodeandchangerootof
subtreerootedwiththisnode.
TheideaistofixthetreeinPostorderfashion.Whenwevisitanode,wemakesurethatitsleftandright
subtreesarealreadyfixed.Incase1.a),wesimplyremoverootandreturnrightsubtreeasnewroot.In
case1.b),weremoverootandreturnleftsubtreeasnewroot.
// A C++ program to remove BST keys outside the given range
#include<stdio.h>
#include <iostream>
// A BST node has key, and left and right pointers

struct node
{
int key;
struct node *left;
struct node *right;
};
// Resmoves all nodes having value outside the given range and returns the root
// of modified tree
node* removeOutsideRange(node *root, int min, int max)
{
// Base Case
if (root == NULL)
return NULL;
// First fix the left and right subtrees of root

root->left = removeOutsideRange(root->left, min, max);
root->right = removeOutsideRange(root->right, min, max);
// Now fix the root. There are 2 possible cases for toot
// 1.a) Root's key is smaller than min value (root is not in range)
if (root->key < min)
{
node *rChild = root->right;
delete root;
return rChild;
}
// 1.b) Root's key is greater than max value (root is not in range)
if (root->key > max)
{
node *lChild = root->left;
delete root;
return lChild;
}
// 2. Root is in range
return root;
}
// A utility function to create a new BST node with key as given num
node* newNode(int num)
{
temp->key = num;
return temp;
}
// A utility function to insert a given key to BST

node* insert(node* root, int key)
{
if (root == NULL)
return newNode(key);
root->left = insert(root->left, key);
else
root->right = insert(root->right, key);
return root;
// Utility function to traverse the binary tree after conversion

void inorderTraversal(node* root)
{
if (root)
{
inorderTraversal( root->left );
cout << root->key << " ";
inorderTraversal( root->right );
}
}

int main()
{
node* root = NULL;
root = insert(root, -13);
cout << "Inorder traversal of the given tree is: ";

inorderTraversal(root);
root = removeOutsideRange(root, -10, 13);
cout << "\nInorder traversal of the modified tree is: ";

return 0;
}
Output:
Inordertraversalofthegiventreeis:13867131415
Inordertraversalofthemodifiedtreeis:86713
TimeComplexity:O(n)wherenisthenumberofnodesingivenBST.
=======================================================================
ConstructCompleteBinaryTree
fromitsLinkedListRepresentation
GivenLinkedListRepresentationofCompleteBinaryTree,constructtheBinarytree.Acompletebinarytree
canberepresentedinanarrayinthefollowingapproach.
Ifrootnodeisstoredatindexi,itsleft,andrightchildrenarestoredatindices2*i+1,2*i+2respectively.
Supposetreeisrepresentedbyalinkedlistinsameway,howdoweconvertthisintonormallinked
representationofbinarytreewhereeverynodehasdata,leftandrightpointers?Inthelinkedlist
representation,wecannotdirectlyaccessthechildrenofthecurrentnodeunlesswetraversethelist.
Wearemainlygivenlevelordertraversalinsequentialaccessform.Weknowheadoflinkedlistisalwaysis
rootofthetree.Wetakethefirstnodeasrootandwealsoknowthatthenexttwonodesareleftandright
childrenofroot.SoweknowpartialBinaryTree.TheideaistodoLevelordertraversalofthepartiallybuilt
BinaryTreeusingqueueandtraversethelinkedlistatthesametime.Ateverystep,wetaketheparent
nodefromqueue,makenexttwonodesoflinkedlistaschildrenoftheparentnode,andenqueuethenext
twonodestoqueue.
1.Createanemptyqueue.
2.Makethefirstnodeofthelistasroot,andenqueueittothequeue.
3.Untilwereachtheendofthelist,dothefollowing.
a.Dequeueonenodefromthequeue.Thisisthecurrentparent.
b.Traversetwonodesinthelist,addthemaschildrenofthecurrentparent.
c.Enqueuethetwonodesintothequeue.
BelowisthecodewhichimplementsthesameinC++.
// C++ program to create a Complete Binary tree from its Linked List
// Representation
#include <iostream>
#include <string>
#include <queue>
// Linked list node

struct ListNode
{
int data;
ListNode* next;
};
// Binary tree node structure

struct BinaryTreeNode
{
int data;
BinaryTreeNode *left, *right;
};
// Function to insert a node at the beginning of the Linked List

void push(struct ListNode** head_ref, int new_data)
{
// allocate node and assign data
struct ListNode* new_node = new ListNode;
new_node->data = new_data;
// link the old list off the new node

new_node->next = (*head_ref);
// move the head to point to the new node

(*head_ref)
= new_node;
// method to create a new binary tree node from the given data
BinaryTreeNode* newBinaryTreeNode(int data)
{
BinaryTreeNode *temp = new BinaryTreeNode;
temp->data = data;
return temp;
}
// converts a given linked list representing a complete binary tree into the
// linked representation of binary tree.
void convertList2Binary(ListNode *head, BinaryTreeNode* &root)
{
// queue to store the parent nodes
queue<BinaryTreeNode *> q;
// Base Case
if (head == NULL)
{
root = NULL; // Note that root is passed by reference
return;
}
// 1.) The first node is always the root node, and add it to the queue
root = newBinaryTreeNode(head->data);
q.push(root);
// advance the pointer to the next node

head = head->next;
// until the end of linked list is reached, do the following steps

while (head)
{
// 2.a) take the parent node from the q and remove it from q
BinaryTreeNode* parent = q.front();
q.pop();
// 2.c) take next two nodes from the linked list. We will add
// them as children of the current parent node in step 2.b. Push them
// into the queue so that they will be parents to the future nodes
BinaryTreeNode *leftChild = NULL, *rightChild = NULL;
leftChild = newBinaryTreeNode(head->data);
q.push(leftChild);
head = head->next;
if (head)
{
rightChild = newBinaryTreeNode(head->data);
q.push(rightChild);
head = head->next;
}
// 2.b) assign the left and right children of parent

parent->left = leftChild;
parent->right = rightChild;
}
}
// Utility function to traverse the binary tree after conversion

void inorderTraversal(BinaryTreeNode* root)
{
if (root)
{
inorderTraversal( root->left );
inorderTraversal( root->right );
}
}

int main()
{
// create a linked list shown in above diagram
struct ListNode* head = NULL;
push(&head, 36); /* Last node of Linked List */
push(&head, 30);
push(&head, 25);
push(&head, 15);
push(&head, 12);
push(&head, 10); /* First node of Linked List */
BinaryTreeNode *root;
convertList2Binary(head, root);
cout << "Inorder Traversal of the constructed Binary Tree is: \n";
return 0;
}
Output:
InorderTraversaloftheconstructedBinaryTreeis:
251230103615
TimeComplexity:TimecomplexityoftheabovesolutionisO(n)wherenisthenumberofnodes
=======================================================================
ReverseLevelOrderTraversal
Wehavediscussedlevelordertraversalofapostinpreviouspost.Theideaistoprintlastlevelfirst,then
secondlastlevel,andsoon.LikeLevelordertraversal,everylevelisprintedfromlefttoright.

ExampleTree
ReverseLevelordertraversaloftheabovetreeis45231.
Bothmethodsfornormallevelordertraversalcanbeeasilymodifiedtodoreverselevelordertraversal.
METHOD1(Recursivefunctiontoprintagivenlevel)
Wecaneasilymodifythemethod1ofthenormallevelordertraversal.Inmethod1,wehaveamethod
printGivenLevel()whichprintsagivenlevelnumber.Theonlythingweneedtochangeis,insteadofcalling
printGivenLevel()fromfirstleveltolastlevel,wecallitfromlastleveltofirstlevel.
// A recursive C program to print REVERSE level order traversal
#include <stdio.h>
#include <stdlib.h>
/* A binary tree node has data, pointer to left and right child */
struct node
{
int data;
struct node* left;
struct node* right;
};
/*Function protoypes*/
void printGivenLevel(struct node* root, int level);
int height(struct node* node);
struct node* newNode(int data);
/* Function to print REVERSE level order traversal a tree*/

void reverseLevelOrder(struct node* root)
{
int h = height(root);
int i;
for (i=h; i>=1; i--) //THE ONLY LINE DIFFERENT FROM NORMAL LEVEL ORDER
printGivenLevel(root, i);
}
/* Print nodes at a given level */

void printGivenLevel(struct node* root, int level)
{
if (root == NULL)
return;
if (level == 1)
else if (level > 1)
{
printGivenLevel(root->left, level-1);
printGivenLevel(root->right, level-1);
}
}
/* Compute the "height" of a tree -- the number of

nodes along the longest path from the root node
down to the farthest leaf node.*/
int height(struct node* node)
{
if (node==NULL)
return 0;
else
{
/* compute the height of each subtree */
int lheight = height(node->left);
int rheight = height(node->right);
/* use the larger one */

if (lheight > rheight)
return(lheight+1);
else return(rheight+1);
}
}

{
struct node* node = (struct node*)
malloc(sizeof(struct node));
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
printf("Level Order traversal of binary tree is \n");

reverseLevelOrder(root);
return 0;
}
Output:
LevelOrdertraversalofbinarytreeis
45231
TimeComplexity:TheworstcasetimecomplexityofthismethodisO(n^2).Foraskewedtree,
printGivenLevel()takesO(n)timewherenisthenumberofnodesintheskewedtree.Sotimecomplexityof
printLevelOrder()isO(n)+O(n1)+O(n2)+..+O(1)whichisO(n^2).
METHOD2(UsingQueueandStack)
Themethod2ofnormallevelordertraversalcanalsobeeasilymodifiedtoprintlevelordertraversalin
reverseorder.Theideaistouseastacktogetthereverselevelorder.Ifwedonormallevelordertraversal
andinsteadofprintinganode,pushthenodetoastackandthenprintcontentsofstack,weget54321
foraboveexampletree,butoutputshouldbe45231.Sotogetthecorrectsequence(lefttorightatevery
level),weprocesschildrenofanodeinreverseorder,wefirstpushtherightsubtreetostack,thenleft
subtree.
// A C++ program to print REVERSE level order traversal using stack and queue
// This approach is adopted from following link
// http://tech-queries.blogspot.in/2008/12/level-order-tree-traversal-in-reverse.html
#include <iostream>
#include <stack>
#include <queue>
/* A binary tree node has data, pointer to left and right children */
struct node
{
int data;
struct node* left;
struct node* right;
};
/* Given a binary tree, print its nodes in reverse level order */

void reverseLevelOrder(node* root)
{
stack <node *> S;
queue <node *> Q;
Q.push(root);
// Do something like normal level order traversal order. Following are the
// differences with normal level order traversal
// 1) Instead of printing a node, we push the node to stack
// 2) Right subtree is visited before left subtree
while (Q.empty() == false)
{
/* Dequeue node and make it root */
root = Q.front();
Q.pop();
S.push(root);
/* Enqueue right child */

if (root->right)
Q.push(root->right); // NOTE: RIGHT CHILD IS ENQUEUED BEFORE LEFT
/* Enqueue left child */

if (root->left)
Q.push(root->left);
}
// Now pop all items from stack one by one and print them
while (S.empty() == false)
{
root = S.top();
S.pop();
}
}

{
temp->data = data;
temp->left = NULL;
temp->right = NULL;
return (temp);
}
int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
cout << "Level Order traversal of binary tree is \n";

reverseLevelOrder(root);
return 0;
}
Output:
LevelOrdertraversalofbinarytreeis
4567231
TimeComplexity:O(n)wherenisnumberofnodesinthebinarytree.
=======================================================================
Findapairwithgivensumina
BalancedBST
GivenaBalancedBinarySearchTreeandatargetsum,writeafunctionthatreturnstrueifthereisapair
withsumequalstotargetsum,otherwisereturnfalse.ExpectedtimecomplexityisO(n)andonlyO(Logn)
extraspacecanbeused.AnymodificationtoBinarySearchTreeisnotallowed.Notethatheightofa
BalancedBSTisalwaysO(Logn).

Thisproblemismainlyextensionofthepreviouspost.HerewearenotallowedtomodifytheBST.
TheBruteForceSolutionistoconsidereachpairinBSTandcheckwhetherthesumequalstoX.Thetime
complexityofthissolutionwillbeO(n^2).
ABetterSolutionistocreateanauxiliaryarrayandstoreInordertraversalofBSTinthearray.Thearray
willbesortedasInordertraversalofBSTalwaysproducessorteddata.OncewehavetheInordertraversal,
wecanpairinO(n)time(Seethisfordetails).ThissolutionworksinO(n)time,butrequiresO(n)auxiliary
space.
Aspaceoptimizedsolutionisdiscussedinpreviouspost.TheideawastofirstinplaceconvertBSTto
DoublyLinkedList(DLL),thenfindpairinsortedDLLinO(n)time.ThissolutiontakesO(n)timeand
O(Logn)extraspace,butitmodifiesthegivenBST.
ThesolutiondiscussedbelowtakesO(n)time,O(Logn)spaceanddoesntmodifyBST.Theideais
sameasfindingthepairinsortedarray(Seemethod1ofthisfordetails).WetraverseBSTinNormal
InorderandReverseInordersimultaneously.Inreverseinorder,westartfromtherightmostnodewhichis
themaximumvaluenode.Innormalinorder,westartfromtheleftmostnodewhichisminimumvaluenode.
Weaddsumofcurrentnodesinbothtraversalsandcomparethissumwithgiventargetsum.Ifthesumis
sameastargetsum,wereturntrue.Ifthesumismorethantargetsum,wemovetonextnodeinreverse
inordertraversal,otherwisewemovetonextnodeinnormalinordertraversal.Ifanyofthetraversalsis
finishedwithoutfindingapair,wereturnfalse.FollowingisC++implementationofthisapproach.
/* In a balanced binary search tree isPairPresent two element which sums to
a given value time O(n) space O(logn) */
#include <stdio.h>
#include <stdlib.h>
// A BST node
struct node
{
int val;
};
// Stack type
struct Stack
{
int size;
int top;
struct node* *array;
};

{
struct Stack* stack =
(struct Stack*) malloc(sizeof(struct Stack));
stack->size = size;
stack->top = -1;
stack->array =
(struct node**) malloc(stack->size * sizeof(struct node*));
return stack;
}

{
return stack->top - 1 == stack->size; }

{
void push(struct Stack* stack, struct node* node)

{
if (isFull(stack))
return;
struct node* pop(struct Stack* stack)

{
if (isEmpty(stack))
return NULL;
}
// Returns true if a pair with target sum exists in BST, otherwise false
bool isPairPresent(struct node *root, int target)
{
// Create two stacks. s1 is used for normal inorder traversal
// and s2 is used for reverse inorder traversal
struct Stack* s1 = createStack(MAX_SIZE);
// Note the sizes of stacks is MAX_SIZE, we can find the tree size and
// fix stack size as O(Logn) for balanced trees like AVL and Red Black
// tree. We have used MAX_SIZE to keep the code simple
// done1, val1 and curr1 are used for normal inorder traversal using s1
// done2, val2 and curr2 are used for reverse inorder traversal using s2
bool done1 = false, done2 = false;
int val1 = 0, val2 = 0;
struct node *curr1 = root, *curr2 = root;
// The loop will break when we either find a pair or one of the two
// traversals is complete
while (1)
{
// Find next node in normal Inorder traversal. See following post
// http://www.geeksforgeeks.org/inorder-tree-traversal-without-recursion/
while (done1 == false)
{
if (curr1 != NULL)
{
push(s1, curr1);
curr1 = curr1->left;
}
else
{
if (isEmpty(s1))
done1 = 1;
else
{
curr1 = pop(s1);
val1 = curr1->val;
curr1 = curr1->right;
done1 = 1;
}
}
}
// Find next node in REVERSE Inorder traversal. The only

// difference between above and below loop is, in below loop
// right subtree is traversed before left subtree
while (done2 == false)
{
if (curr2 != NULL)
{
push(s2, curr2);
curr2 = curr2->right;
}
else
{
if (isEmpty(s2))
done2 = 1;
else
{
curr2 = pop(s2);
val2 = curr2->val;
curr2 = curr2->left;
done2 = 1;
}
}
}
// If we find a pair, then print the pair and return. The first
// condition makes sure that two same values are not added
if ((val1 != val2) && (val1 + val2) == target)
{
printf("\n Pair Found: %d + %d = %d\n", val1, val2, target);
return true;
}
// If sum of current values is smaller, then move to next node in

// normal inorder traversal
else if ((val1 + val2) < target)
done1 = false;
// If sum of current values is greater, then move to next node in

// reverse inorder traversal
else if ((val1 + val2) > target)
done2 = false;
// If any of the inorder traversals is over, then there is no pair

// so return false
if (val1 >= val2)
return false;
}
}
// A utility function to create BST node

struct node * NewNode(int val)
{
struct node *tmp = (struct node *)malloc(sizeof(struct node));
tmp->val = val;
tmp->right = tmp->left =NULL;
return tmp;
}

int main()
{
/*
15
/
10
/\
8 12
\
20
/ \
16 25*/
struct node *root = NewNode(15);

root->left = NewNode(10);
root->right = NewNode(20);
root->left->left = NewNode(8);
root->left->right = NewNode(12);
root->right->left = NewNode(16);
root->right->right = NewNode(25);
int target = 33;

if (isPairPresent(root, target) == false)
printf("\n No such values are found\n");
getchar();
return 0;
}
Output:
PairFound:8+25=33
=======================================================================
FindifthereisatripletinaBalanced
BSTthataddstozero
GivenaBalancedBinarySearchTree(BST),writeafunctionisTripletPresent()thatreturnstrueifthereisa
tripletingivenBSTwithsumequalsto0,otherwisereturnsfalse.ExpectedtimecomplexityisO(n^2)and
onlyO(Logn)extraspacecanbeused.YoucanmodifygivenBinarySearchTree.Notethatheightofa
BalancedBSTisalwaysO(Logn)
Forexample,isTripletPresent()shouldreturntrueforfollowingBSTbecausethereisatripletwithsum0,the
tripletis{13,6,7}.
TheBruteForceSolutionistoconsidereachtripletinBSTandcheckwhetherthesumaddsuptozero.
ThetimecomplexityofthissolutionwillbeO(n^3).
ABetterSolutionistocreateanauxiliaryarrayandstoreInordertraversalofBSTinthearray.Thearray
willbesortedasInordertraversalofBSTalwaysproducessorteddata.OncewehavetheInordertraversal,
wecanusemethod2ofthisposttofindthetripletwithsumequalsto0.ThissolutionworksinO(n^2)time,
butrequiresO(n)auxiliaryspace.
FollowingisthesolutionthatworksinO(n^2)timeandusesO(Logn)extraspace:
1)ConvertgivenBSTtoDoublyLinkedList(DLL)
2)NowiteratethrougheverynodeofDLLandifthekeyofnodeisnegative,thenfindapairinDLLwithsum
equaltokeyofcurrentnodemultipliedby1.Tofindthepair,wecanusetheapproachusedin
hasArrayTwoCandidates()inmethod1ofthispost.
// A C++ program to check if there is a triplet with sum equal to 0 in
// a given BST
#include<stdio.h>
// A BST node has key, and left and right pointers

struct node
{
int key;
struct node *left;
struct node *right;
};
// A function to convert given BST to Doubly Linked List. left pointer is used
// as previous pointer and right pointer is used as next pointer. The function
// sets *head to point to first and *tail to point to last node of converted DLL
void convertBSTtoDLL(node* root, node** head, node** tail)
{
// Base case
if (root == NULL)
return;
// First convert the left subtree

if (root->left)
convertBSTtoDLL(root->left, head, tail);
// Then change left of current root as last node of left subtree

root->left = *tail;
// If tail is not NULL, then set right of tail as root, else current
// node is head
if (*tail)
(*tail)->right = root;
else
*head = root;
// Update tail
*tail = root;
// Finally, convert right subtree

if (root->right)
convertBSTtoDLL(root->right, head, tail);
}
// This function returns true if there is pair in DLL with sum equal
// to given sum. The algorithm is similar to hasArrayTwoCandidates()
// in method 1 of http://tinyurl.com/dy6palr
bool isPresentInDLL(node* head, node* tail, int sum)
{
while (head != tail)
{
int curr = head->key + tail->key;
if (curr == sum)
return true;
else if (curr > sum)
tail = tail->left;
else
head = head->right;
}
return false;
}
// The main function that returns true if there is a 0 sum triplet in

// BST otherwise returns false
bool isTripletPresent(node *root)
{
// Check if the given BST is empty
if (root == NULL)
return false;
// Convert given BST to doubly linked list. head and tail store the
// pointers to first and last nodes in DLLL
node* head = NULL;
node* tail = NULL;
convertBSTtoDLL(root, &head, &tail);
// Now iterate through every node and find if there is a pair with sum
// equal to -1 * heaf->key where head is current node
while ((head->right != tail) && (head->key < 0))
{
// If there is a pair with sum equal to -1*head->key, then return
// true else move forward

if (isPresentInDLL(head->right, tail, -1*head->key))
return true;
else
head = head->right;
}
// If we reach here, then there was no 0 sum triplet

return false;
}
// A utility function to create a new BST node with key as given num
node* newNode(int num)
{
temp->key = num;
return temp;
}
// A utility function to insert a given key to BST

node* insert(node* root, int key)
{
if (root == NULL)
return newNode(key);
root->left = insert(root->left, key);
else
root->right = insert(root->right, key);
return root;
}

int main()
{
node* root = NULL;

if (isTripletPresent(root))
printf("Present");
else
printf("Not Present");
return 0;
}
Output:
Present
NotethattheabovesolutionmodifiesgivenBST.
TimeComplexity:TimetakentoconvertBSTtoDLLisO(n)andtimetakentofindtripletinDLLisO(n^2).
AuxiliarySpace:Theauxiliaryspaceisneededonlyforfunctioncallstackinrecursivefunction
convertBSTtoDLL().Sincegiventreeisbalanced(heightisO(Logn)),thenumberoffunctionsincallstack
willneverbemorethanO(Logn).
=======================================================================
IterativePostorderTraversal|Set2
(UsingOneStack)
Wehavediscussedasimpleiterativepostordertraversalusingtwostacksinthepreviouspost.Inthispost,
anapproachwithonlyonestackisdiscussed.
Theideaistomovedowntoleftmostnodeusingleftpointer.Whilemovingdown,pushrootandrootsright
childtostack.Oncewereachleftmostnode,printitifitdoesnthavearightchild.Ifithasarightchild,then
changerootsothattherightchildisprocessedbefore.
Followingisdetailedalgorithm.
1.1Createanemptystack
2.1DofollowingwhilerootisnotNULL
a)Pushroot'srightchildandthenroottostack.
b)Setrootasroot'sleftchild.
2.2Popanitemfromstackandsetitasroot.
a)Ifthepoppeditemhasarightchildandtherightchild
isattopofstack,thenremovetherightchildfromstack,
pushtherootbackandsetrootasroot'srightchild.
b)Elseprintroot'sdataandsetrootasNULL.
2.3Repeatsteps2.1and2.2whilestackisnotempty.
Letusconsiderthefollowingtree
Followingarethestepstoprintpostordertraversaloftheabovetreeusingonestack.
1.Rightchildof1exists.
Push3tostack.Push1tostack.Movetoleftchild.
Stack:3,1
2.Rightchildof2exists.
Push5tostack.Push2tostack.Movetoleftchild.
Stack:3,1,5,2
3.Rightchildof4doesn'texist.'
Push4tostack.Movetoleftchild.
Stack:3,1,5,2,4
4.CurrentnodeisNULL.
Pop4fromstack.Rightchildof4doesn'texist.
Print4.SetcurrentnodetoNULL.
Stack:3,1,5,2
5.CurrentnodeisNULL.
Pop2fromstack.Sincerightchildof2equalsstacktopelement,
pop5fromstack.Nowpush2tostack.
Movecurrentnodetorightchildof2i.e.5
Stack:3,1,2
6.Rightchildof5doesn'texist.Push5tostack.Movetoleftchild.
Stack:3,1,2,5
7.CurrentnodeisNULL.Pop5fromstack.Rightchildof5doesn'texist.
Stack:3,1,2
8.CurrentnodeisNULL.Pop2fromstack.
Rightchildof2isnotequaltostacktopelement.
Stack:3,1
9.CurrentnodeisNULL.Pop1fromstack.
Sincerightchildof1equalsstacktopelement,pop3fromstack.
Nowpush1tostack.Movecurrentnodetorightchildof1i.e.3
Stack:1
10.RepeatthesameasabovestepsandPrint6,7and3.
Pop1andPrint1.
// C program for iterative postorder traversal using one stack
#include <stdio.h>
#include <stdlib.h>

// A tree node
struct Node
{
int data;
};
// Stack type
struct Stack
{
int size;
int top;
};

{
node->data = data;
return node;
}

{
struct Stack* stack = (struct Stack*) malloc(sizeof(struct Stack));
stack->size = size;
stack->top = -1;
stack->array = (struct Node**) malloc(stack->size * sizeof(struct Node*));
return stack;
}

{ return stack->top - 1 == stack->size; }

{ return stack->top == -1; }

{
if (isFull(stack))
return;
}

{
if (isEmpty(stack))
return NULL;
}
struct Node* peek(struct Stack* stack)

{
if (isEmpty(stack))
return NULL;
return stack->array[stack->top];
}
// An iterative function to do postorder traversal of a given binary tree

void postOrderIterative(struct Node* root)
{
// Check for empty tree
if (root == NULL)
return;
struct Stack* stack = createStack(MAX_SIZE);

do
{
// Move to leftmost node
while (root)
{
// Push root's right child and then root to stack.
if (root->right)
push(stack, root->right);
push(stack, root);
// Set root as root's left child

root = root->left;
}
// Pop an item from stack and set it as root

root = pop(stack);
// If the popped item has a right child and the right child is not
// processed yet, then make sure right child is processed before root
if (root->right && peek(stack) == root->right)
{
pop(stack); // remove right child from stack
push(stack, root); // push root back to stack
root = root->right; // change root so that the right
// child is processed next
}
else // Else print root's data and set root as NULL
{
root = NULL;
}
} while (!isEmpty(stack));
}

int main()
{
// Let us construct the tree shown in above figure
struct Node* root = NULL;
root = newNode(1);
postOrderIterative(root);
return 0;
}
Output:
4526731
=======================================================================
IterativePostorderTraversal|Set1
(UsingTwoStacks)
Wehavediscussediterativeinorderanditerativepreordertraversals.Inthispost,iterativepostorder
traversalisdiscussedwhichismorecomplexthantheothertwotraversals(duetoitsnatureofnontail
recursion,thereisanextrastatementafterthefinalrecursivecalltoitself).Thepostordertraversalcan
easilybedoneusingtwostacksthough.Theideaistopushreversepostordertraversaltoastack.Oncewe
havereversepostordertraversalinastack,wecanjustpopallitemsonebyonefromthestackandprint
them,thisorderofprintingwillbeinpostorderbecauseofLIFOpropertyofstacks.Nowthequestionis,how
togetreversepostorderelementsinastacktheotherstackisusedforthispurpose.Forexample,inthe
followingtree,weneedtoget1,3,7,6,2,5,4inastack.Iftakeacloserlookatthissequence,wecan
observethatthissequenceisverysimilartopreordertraversal.Theonlydifferenceisrightchildisvisited
beforeleftchildandthereforesequenceisrootrightleftinsteadofrootleftright.Sowecandosomething
likeiterativepreordertraversalwithfollowingdifferences.
a)Insteadofprintinganitem,wepushittoastack.
b)Wepushleftsubtreebeforerightsubtree.
Followingisthecompletealgorithm.Afterstep2,wegetreversepostordertraversalinsecondstack.We
usefirststacktogetthisorder.
1.Pushroottofirststack.
2.Loopwhilefirststackisnotempty
2.1Popanodefromfirststackandpushittosecondstack
2.2Pushleftandrightchildrenofthepoppednodetofirststack
3.Printcontentsofsecondstack
Letusconsiderthefollowingtree

Followingarethestepstoprintpostordertraversaloftheabovetreeusingtwostacks.
1.Push1tofirststack.
Firststack:1
Secondstack:Empty
2.Pop1fromfirststackandpushittosecondstack.
Pushleftandrightchildrenof1tofirststack
Firststack:2,3
Secondstack:1
Firststack:2,6,7
Secondstack:1,3
Firststack:2,6
Secondstack:1,3,7
Firststack:2
Secondstack:1,3,7,6
Firststack:4,5
Secondstack:1,3,7,6,2
Firststack:4
Secondstack:1,3,7,6,2,5
Firststack:Empty
Secondstack:1,3,7,6,2,5,4
Thealgorithmstopssincethereisnomoreiteminfirststack.
Observethatcontentofsecondstackisinpostorderfashion.Printthem.
FollowingisCimplementationofiterativepostordertraversalusingtwostacks.
#include <stdio.h>
#include <stdlib.h>

// A tree node
struct Node
{
int data;
};
// Stack type
struct Stack
{
int size;
int top;
};

{
node->data = data;
return node;
}

{
struct Stack* stack =
(struct Stack*) malloc(sizeof(struct Stack));
stack->size = size;
stack->top = -1;
stack->array =
(struct Node**) malloc(stack->size * sizeof(struct Node*));
return stack;
}

{ return stack->top - 1 == stack->size; }

{ return stack->top == -1; }

{
if (isFull(stack))
return;
}

{
if (isEmpty(stack))
return NULL;
}
// An iterative function to do post order traversal of a given binary tree

void postOrderIterative(struct Node* root)
{
// Create two stacks
// push root to first stack

push(s1, root);
struct Node* node;
// Run while first stack is not empty

while (!isEmpty(s1))
{
// Pop an item from s1 and push it to s2
node = pop(s1);
push(s2, node);
// Push left and right children of removed item to s1

if (node->left)
push(s1, node->left);
if (node->right)
push(s1, node->right);
}
// Print all elements of second stack

{
node = pop(s2);
}
}
int main()
{
// Let us construct the tree shown in above figure
struct Node* root = NULL;
root = newNode(1);
postOrderIterative(root);
return 0;
}
Output:
4526731
=======================================================================
DynamicProgramming|Set26
(LargestIndependentSetProblem)
GivenaBinaryTree,findsizeoftheLargestIndependentSet(LIS)init.Asubsetofalltreenodesisan
independentsetifthereisnoedgebetweenanytwonodesofthesubset.
Forexample,considerthefollowingbinarytree.Thelargestindependentset(LIS)is{10,40,60,70,80}and
sizeoftheLISis5.

ADynamicProgrammingsolutionsolvesagivenproblemusingsolutionsofsubproblemsinbottomup
manner.Canthegivenproblembesolvedusingsolutionstosubproblems?Ifyes,thenwhatarethe
subproblems?Canwefindlargestindependentsetsize(LISS)foranodeXifweknowLISSforall
descendantsofX?IfanodeisconsideredaspartofLIS,thenitschildrencannotbepartofLIS,butits
grandchildrencanbe.Followingisoptimalsubstructureproperty.
1)OptimalSubstructure:
LetLISS(X)indicatessizeoflargestindependentsetofatreewithrootX.
LISS(X)=MAX{(1+sumofLISSforallgrandchildrenofX),
(sumofLISSforallchildrenofX)}
Theideaissimple,therearetwopossibilitiesforeverynodeX,eitherXisamemberofthesetornota
member.IfXisamember,thenthevalueofLISS(X)is1plusLISSofallgrandchildren.IfXisnota
member,thenthevalueissumofLISSofallchildren.
2)OverlappingSubproblems
Followingisrecursiveimplementationthatsimplyfollowstherecursivestructurementionedabove.
// A naive recursive implementation of Largest Independent Set problem
#include <stdio.h>
#include <stdlib.h>
// A utility function to find max of two integers

int max(int x, int y) { return (x > y)? x: y; }
/* A binary tree node has data, pointer to left child and a pointer to
right child */
struct node
{
int data;
};
// The function returns size of the largest independent set in a given

// binary tree
int LISS(struct node *root)
{
if (root == NULL)
return 0;
// Caculate size excluding the current node

int size_excl = LISS(root->left) + LISS(root->right);
// Calculate size including the current node

int size_incl = 1;
if (root->left)
size_incl += LISS(root->left->left) + LISS(root->left->right);
if (root->right)
size_incl += LISS(root->right->left) + LISS(root->right->right);
// Return the maximum of two sizes

return max(size_incl, size_excl);
}
// A utility function to create a node

struct node* newNode( int data )
{
struct node* temp = (struct node *) malloc( sizeof(struct node) );
temp->data = data;
return temp;
}

int main()
{
// Let us construct the tree given in the above diagram
struct node *root
= newNode(20);
root->left
= newNode(8);
root->left->left
= newNode(4);
root->left->right
= newNode(12);
root->left->right->left
= newNode(10);
root->right
= newNode(22);
root->right->right
= newNode(25);
printf ("Size of the Largest Independent Set is %d ", LISS(root));
return 0;
}
Output:
SizeoftheLargestIndependentSetis5
Timecomplexityoftheabovenaiverecursiveapproachisexponential.Itshouldbenotedthattheabove
functioncomputesthesamesubproblemsagainandagain.Forexample,LISSofnodewithvalue50is
evaluatedfornodewithvalues10and20as50isgrandchildof10andchildof20.
Sincesamesuproblemsarecalledagain,thisproblemhasOverlappingSubprolemsproperty.SoLISS
problemhasbothproperties(seethisandthis)ofadynamicprogrammingproblem.Likeothertypical
DynamicProgramming(DP)problems,recomputationsofsamesubproblemscanbeavoidedbystoringthe
solutionstosubproblemsandsolvingproblemsinbottomupmanner.
FollowingisCimplementationofDynamicProgrammingbasedsolution.Inthefollowingsolution,an
additionalfieldlissisaddedtotreenodes.Theinitialvalueoflississetas0forallnodes.Therecursive
functionLISS()calculateslissforanodeonlyifitisnotalreadyset.
/* Dynamic programming based program for Largest Independent Set problem */
#include <stdio.h>
#include <stdlib.h>
// A utility function to find max of two integers

int max(int x, int y) { return (x > y)? x: y; }
/* A binary tree node has data, pointer to left child and a pointer to
right child */
struct node
{
int data;
int liss;
};
// A memoization function returns size of the largest independent set in

// a given binary tree
int LISS(struct node *root)
{
if (root == NULL)
return 0;
if (root->liss)
return root->liss;

return (root->liss = 1);
// Caculate size excluding the current node

int liss_excl = LISS(root->left) + LISS(root->right);
// Calculate size including the current node

int liss_incl = 1;
if (root->left)
liss_incl += LISS(root->left->left) + LISS(root->left->right);
if (root->right)
liss_incl += LISS(root->right->left) + LISS(root->right->right);
// Return the maximum of two sizes

root->liss = max(liss_incl, liss_excl);
return root->liss;
}

{
temp->data = data;
temp->liss = 0;
return temp;
}

int main()
{
struct node *root
= newNode(20);
root->left
= newNode(8);
root->left->left
= newNode(4);
root->left->right
= newNode(12);
= newNode(10);
root->right
= newNode(22);
root->right->right
= newNode(25);
printf ("Size of the Largest Independent Set is %d ", LISS(root));
return 0;
}
Output
SizeoftheLargestIndependentSetis5
TimeComplexity:O(n)wherenisthenumberofnodesingivenBinarytree.
Followingextensionstoabovesolutioncanbetriedasanexercise.
1)Extendtheabovesolutionfornarytree.
2)Theabovesolutionmodifiesthegiventreestructurebyaddinganadditionalfieldlisstotreenodes.
Extendthesolutionsothatitdoesntmodifythetreestructure.
3)TheabovesolutiononlyreturnssizeofLIS,itdoesntprintelementsofLIS.Extendthesolutiontoprintall
nodesthatarepartofLIS.
=======================================================================
SegmentTree|Set2(Range
MinimumQuery)
Wehaveintroducedsegmenttreewithasimpleexampleinthepreviouspost.Inthispost,RangeMinimum
QueryproblemisdiscussedasanotherexamplewhereSegmentTreecanbeused.Followingisproblem
statement.
Wehaveanarrayarr[0...n1].Weshouldbeabletoefficientlyfindtheminimumvaluefromindexqs
(querystart)toqe(queryend)where0<=qs<=qe<=n1.Thearrayisstatic(elementsarenotdeletedand
insertedduringtheseriesofqueries).
Asimplesolutionistorunaloopfromqstoqeandfindminimumelementingivenrange.Thissolution
takesO(n)timeinworstcase.
Anothersolutionistocreatea2Darraywhereanentry[i,j]storestheminimumvalueinrangearr[i..j].
MinimumofagivenrangecannowbecalculatedinO(1)time,butpreprocessingtakesO(n^2)time.Also,
thisapproachneedsO(n^2)extraspacewhichmaybecomehugeforlargeinputarrays.
Segmenttreecanbeusedtodopreprocessingandqueryinmoderatetime.Withsegmenttree,
preprocessingtimeisO(n)andtimetoforrangeminimumqueryisO(Logn).Theextraspacerequiredis
O(n)tostorethesegmenttree.
RepresentationofSegmenttrees
1.LeafNodesaretheelementsoftheinputarray.
2.Eachinternalnoderepresentsminimumofallleavesunderit.
AnarrayrepresentationoftreeisusedtorepresentSegmentTrees.Foreachnodeatindexi,theleftchildis
atindex2*i+1,rightchildat2*i+2andtheparentisat

ConstructionofSegmentTreefromgivenarray
Westartwithasegmentarr[0...n1].andeverytimewedividethecurrentsegmentintotwohalves(ifithas
notyetbecomeasegmentoflength1),andthencallthesameprocedureonbothhalves,andforeachsuch
segment,westoretheminimumvalueinasegmenttreenode.
Alllevelsoftheconstructedsegmenttreewillbecompletelyfilledexceptthelastlevel.Also,thetreewillbe
aFullBinaryTreebecausewealwaysdividesegmentsintwohalvesateverylevel.Sincetheconstructed
treeisalwaysfullbinarytreewithnleaves,therewillben1internalnodes.Sototalnumberofnodeswillbe
2*n1.
Heightofthesegmenttreewillbe
.Sincethetreeisrepresentedusingarrayandrelation
betweenparentandchildindexesmustbemaintained,sizeofmemoryallocatedforsegmenttreewillbe
.
Queryforminimumvalueofgivenrange
Oncethetreeisconstructed,howtodorangeminimumqueryusingtheconstructedsegmenttree.
Followingisalgorithmtogettheminimum.
//qs>querystartindex,qe>queryendindex
intRMQ(node,qs,qe)
{
ifrangeofnodeiswithinqsandqe
returnvalueinnode
elseifrangeofnodeiscompletelyoutsideqsandqe
returnINFINITE
else
returnmin(RMQ(node'sleftchild,qs,qe),RMQ(node'srightchild,qs,qe))
}
Implementation:
// Program for range minimum query using segment tree
#include <stdio.h>
#include <math.h>
#include <limits.h>
// A utility function to get minimum of two numbers

int minVal(int x, int y) { return (x < y)? x: y; }
// A utility function to get the middle index from corner indexes.

int getMid(int s, int e) { return s + (e -s)/2; }
/* A recursive function to get the minimum value in a given range of array

indexes. The following are parameters for this function.
st--> Pointer to segment tree

index --> Index of current node in the segment tree. Initially 0 is
passed as root is always at index 0
ss & se --> Starting and ending indexes of the segment represented by
current node, i.e., st[index]
qs & qe --> Starting and ending indexes of query range */
int RMQUtil(int *st, int ss, int se, int qs, int qe, int index)
{
// If segment of this node is a part of given range, then return the
// min of the segment
if (qs <= ss && qe >= se)
return st[index];
// If segment of this node is outside the given range

if (se < qs || ss > qe)
return INT_MAX;
// If a part of this segment overlaps with the given range

int mid = getMid(ss, se);
return minVal(RMQUtil(st, ss, mid, qs, qe, 2*index+1),
RMQUtil(st, mid+1, se, qs, qe, 2*index+2));
}
// Return minimum of elements in range from index qs (quey start) to

// qe (query end). It mainly uses RMQUtil()
int RMQ(int *st, int n, int qs, int qe)
{
// Check for erroneous input values
if (qs < 0 || qe > n-1 || qs > qe)
{
printf("Invalid Input");
return -1;
}
return RMQUtil(st, 0, n-1, qs, qe, 0);

}
// A recursive function that constructs Segment Tree for array[ss..se].

// si is index of current node in segment tree st
int constructSTUtil(int arr[], int ss, int se, int *st, int si)
{
// If there is one element in array, store it in current node of
// segment tree and return
if (ss == se)
{
st[si] = arr[ss];
return arr[ss];
}
// If there are more than one elements, then recur for left and
// right subtrees and store the minimum of two values in this node
st[si] = minVal(constructSTUtil(arr, ss, mid, st, si*2+1),

constructSTUtil(arr, mid+1, se, st, si*2+2));
return st[si];
}
/* Function to construct segment tree from given array. This function

allocates memory for segment tree and calls constructSTUtil() to
fill the allocated memory */
int *constructST(int arr[], int n)
{
// Allocate memory for segment tree
int x = (int)(ceil(log2(n))); //Height of segment tree
int max_size = 2*(int)pow(2, x) - 1; //Maximum size of segment tree
int *st = new int[max_size];
// Fill the allocated memory st

constructSTUtil(arr, 0, n-1, st, 0);
// Return the constructed segment tree

return st;
}

int main()
{
int arr[] = {1, 3, 2, 7, 9, 11};
int n = sizeof(arr)/sizeof(arr[0]);
// Build segment tree from given array

int *st = constructST(arr, n);
int qs = 1; // Starting index of query range

int qe = 5; // Ending index of query range
// Print minimum value in arr[qs..qe]

printf("Minimum of values in range [%d, %d] is = %d\n",
qs, qe, RMQ(st, n, qs, qe));
return 0;
}
Output:
Minimumofvaluesinrange[1,5]is=2
TimeComplexity:
TimeComplexityfortreeconstructionisO(n).Therearetotal2n1nodes,andvalueofeverynodeis
calculatedonlyonceintreeconstruction.
TimecomplexitytoqueryisO(Logn).Toqueryarangeminimum,weprocessatmosttwonodesatevery
levelandnumberoflevelsisO(Logn).
=======================================================================
SegmentTree|Set1(Sumofgiven
range)
LetusconsiderthefollowingproblemtounderstandSegmentTrees.
Wehaveanarrayarr[0...n1].Weshouldbeableto
1Findthesumofelementsfromindexltorwhere0<=l<=r<=n1
2Changevalueofaspecifiedelementofthearrayarr[i]=xwhere0<=i<=n1.
Asimplesolutionistorunaloopfromltorandcalculatesumofelementsingivenrange.Toupdatea
value,simplydoarr[i]=x.ThefirstoperationtakesO(n)timeandsecondoperationtakesO(1)time.
Anothersolutionistocreateanotherarrayandstoresumfromstarttoiattheithindexinthisarray.Sum
ofagivenrangecannowbecalculatedinO(1)time,butupdateoperationtakesO(n)timenow.Thisworks
wellifthenumberofqueryoperationsarelargeandveryfewupdates.
Whatifthenumberofqueryandupdatesareequal?CanweperformboththeoperationsinO(logn)
timeoncegiventhearray?WecanuseaSegmentTreetodobothoperationsinO(Logn)time.
RepresentationofSegmenttrees
1.LeafNodesaretheelementsoftheinputarray.
2.Eachinternalnoderepresentssomemergingoftheleafnodes.Themergingmaybedifferentfordifferent
problems.Forthisproblem,mergingissumofleavesunderanode.
AnarrayrepresentationoftreeisusedtorepresentSegmentTrees.Foreachnodeatindexi,theleftchildis
atindex2*i+1,rightchildat2*i+2andtheparentisat
ConstructionofSegmentTreefromgivenarray
Westartwithasegmentarr[0...n1].andeverytimewedividethecurrentsegmentintotwohalves(ifithas
notyetbecomeasegmentoflength1),andthencallthesameprocedureonbothhalves,andforeachsuch
segmentwestorethesumincorrespondingnode.
Alllevelsoftheconstructedsegmenttreewillbecompletelyfilledexceptthelastlevel.Also,thetreewillbe
aFullBinaryTreebecausewealwaysdividesegmentsintwohalvesateverylevel.Sincetheconstructed
treeisalwaysfullbinarytreewithnleaves,therewillben1internalnodes.Sototalnumberofnodeswillbe
2*n1.
Heightofthesegmenttreewillbe
.Sincethetreeisrepresentedusingarrayandrelation
betweenparentandchildindexesmustbemaintained,sizeofmemoryallocatedforsegmenttreewillbe
.
QueryforSumofgivenrange
Oncethetreeisconstructed,howtogetthesumusingtheconstructedsegmenttree.Followingisalgorithm
togetthesumofelements.
intgetSum(node,l,r)
{
ifrangeofnodeiswithinlandr
returnvalueinnode
elseifrangeofnodeiscompletelyoutsidelandr
return0
else
returngetSum(node'sleftchild,l,r)+
getSum(node'srightchild,l,r)
}
Updateavalue
Liketreeconstructionandqueryoperations,updatecanalsobedonerecursively.Wearegivenanindex
whichneedstoupdated.Letdiffbethevaluetobeadded.Westartfromrootofthesegmenttree,andadd
difftoallnodeswhichhavegivenindexintheirrange.Ifanodedoesnthavegivenindexinitsrange,we
dontmakeanychangestothatnode.
Implementation:
Followingisimplementationofsegmenttree.Theprogramimplementsconstructionofsegmenttreeforany
givenarray.Italsoimplementsqueryandupdateoperations.
// Program to show segment tree operations like construction, query and update
#include <stdio.h>
#include <math.h>
// A utility function to get the middle index from corner indexes.

int getMid(int s, int e) { return s + (e -s)/2; }
/* A recursive function to get the sum of values in given range of the array.
The following are parameters for this function.
st--> Pointer to segment tree

index --> Index of current node in the segment tree. Initially 0 is
passed as root is always at index 0
ss & se --> Starting and ending indexes of the segment represented by
current node, i.e., st[index]
qs & qe --> Starting and ending indexes of query range */
int getSumUtil(int *st, int ss, int se, int qs, int qe, int index)
{
// If segment of this node is a part of given range, then return the
// sum of the segment

if (qs <= ss && qe >= se)
return st[index];
// If segment of this node is outside the given range

if (se < qs || ss > qe)
return 0;
// If a part of this segment overlaps with the given range

return getSumUtil(st, ss, mid, qs, qe, 2*index+1) +
getSumUtil(st, mid+1, se, qs, qe, 2*index+2);
}
/* A recursive function to update the nodes which have the given index in
their range. The following are parameters
st, index, ss and se are same as getSumUtil()
i --> index of the element to be updated. This index is in input array.
diff --> Value to be added to all nodes which have i in range */
void updateValueUtil(int *st, int ss, int se, int i, int diff, int index)
{
// Base Case: If the input index lies outside the range of this segment
if (i < ss || i > se)
return;
// If the input index is in range of this node, then update the value
// of the node and its children
st[index] = st[index] + diff;
if (se != ss)
{
updateValueUtil(st, ss, mid, i, diff, 2*index + 1);
updateValueUtil(st, mid+1, se, i, diff, 2*index + 2);
}
}
// The function to update a value in input array and segment tree.
// It uses updateValueUtil() to update the value in segment tree

void updateValue(int arr[], int *st, int n, int i, int new_val)
{
// Check for erroneous input index
if (i < 0 || i > n-1)
{
return;
}
// Get the difference between new value and old value

int diff = new_val - arr[i];
// Update the value in array

arr[i] = new_val;
// Update the values of nodes in segment tree

updateValueUtil(st, 0, n-1, i, diff, 0);
}
// Return sum of elements in range from index qs (quey start) to

// qe (query end). It mainly uses getSumUtil()
int getSum(int *st, int n, int qs, int qe)
{
// Check for erroneous input values
if (qs < 0 || qe > n-1 || qs > qe)
{
return -1;
}
return getSumUtil(st, 0, n-1, qs, qe, 0);

}
// A recursive function that constructs Segment Tree for array[ss..se].

// si is index of current node in segment tree st
int constructSTUtil(int arr[], int ss, int se, int *st, int si)
{
// If there is one element in array, store it in current node of
// segment tree and return
if (ss == se)
{
st[si] = arr[ss];
return arr[ss];
}
// If there are more than one elements, then recur for left and
// right subtrees and store the sum of values in this node
st[si] = constructSTUtil(arr, ss, mid, st, si*2+1) +
constructSTUtil(arr, mid+1, se, st, si*2+2);
return st[si];
}
/* Function to construct segment tree from given array. This function

allocates memory for segment tree and calls constructSTUtil() to
fill the allocated memory */
int *constructST(int arr[], int n)
{
// Allocate memory for segment tree
int x = (int)(ceil(log2(n))); //Height of segment tree
int max_size = 2*(int)pow(2, x) - 1; //Maximum size of segment tree
int *st = new int[max_size];
// Fill the allocated memory st

constructSTUtil(arr, 0, n-1, st, 0);
// Return the constructed segment tree

return st;
}

int main()
{
int arr[] = {1, 3, 5, 7, 9, 11};

// Build segment tree from given array

int *st = constructST(arr, n);
// Print sum of values in array from index 1 to 3

printf("Sum of values in given range = %d\n", getSum(st, n, 1, 3));
// Update: set arr[1] = 10 and update corresponding segment

// tree nodes
updateValue(arr, st, n, 1, 10);
// Find sum after the value is updated

printf("Updated sum of values in given range = %d\n",
getSum(st, n, 1, 3));
return 0;
}
Output:
Sumofvaluesingivenrange=15
Updatedsumofvaluesingivenrange=22
TimeComplexity:
TimeComplexityfortreeconstructionisO(n).Therearetotal2n1nodes,andvalueofeverynodeis
calculatedonlyonceintreeconstruction.
TimecomplexitytoqueryisO(Logn).Toqueryasum,weprocessatmostfournodesateveryleveland
numberoflevelsisO(Logn).
ThetimecomplexityofupdateisalsoO(Logn).Toupdatealeafvalue,weprocessonenodeateverylevel
andnumberoflevelsisO(Logn).
=======================================================================
TernarySearchTree
Aternarysearchtreeisaspecialtriedatastructurewherethechildnodesofastandardtrieareorderedas
abinarysearchtree.
Representationofternarysearchtrees:
Unliketrie(standard)datastructurewhereeachnodecontains26pointersforitschildren,eachnodeina
ternarysearchtreecontainsonly3pointers:
1.Theleftpointerpointstothenodewhosevalueislessthanthevalueinthecurrentnode.
2.Theequalpointerpointstothenodewhosevalueisequaltothevalueinthecurrentnode.
3.Therightpointerpointstothenodewhosevalueisgreaterthanthevalueinthecurrentnode.
Apartfromabovethreepointers,eachnodehasafieldtoindicatedata(characterincaseofdictionary)and
anotherfieldtomarkendofastring.
So,moreorlessitissimilartoBSTwhichstoresdatabasedonsomeorder.However,datainaternary
searchtreeisdistributedoverthenodes.e.g.Itneeds4nodestostorethewordGeek.
Belowfigureshowshowexactlythewordsinaternarysearchtreearestored?
Oneoftheadvantageofusingternarysearchtreesovertriesisthatternarysearchtreesareamorespace
efficient(involveonlythreepointerspernodeascomparedto26instandardtries).Further,ternarysearch
treescanbeusedanytimeahashtablewouldbeusedtostorestrings.
Triesaresuitablewhenthereisaproperdistributionofwordsoverthealphabetssothatspacesareutilized
mostefficiently.Otherwiseternarysearchtreesarebetter.Ternarysearchtreesareefficienttouse(interms
ofspace)whenthestringstobestoredshareacommonprefix.
Applicationsofternarysearchtrees:
1.TernarysearchtreesareefficientforquerieslikeGivenaword,findthenextwordin
dictionary(nearneighborlookups)orFindalltelephonenumbersstartingwith9342ortypingfewstarting
charactersinawebbrowserdisplaysallwebsitenameswiththisprefix(Autocompletefeature).
2.Usedinspellchecks:Ternarysearchtreescanbeusedasadictionarytostoreallthewords.Oncethe
wordistypedinaneditor,thewordcanbeparallelysearchedintheternarysearchtreetocheckforcorrect
spelling.
Implementation:
FollowingisCimplementationofternarysearchtree.Theoperationsimplementedare,search,insertand
traversal.
// C program to demonstrate Ternary Search Tree (TST) insert, travese
// and search operations
#include <stdio.h>
#include <stdlib.h>
#define MAX 50
// A node of ternary search tree

struct Node
{
char data;
// True if this character is last character of one of the words

unsigned isEndOfString: 1;
struct Node *left, *eq, *right;

};
// A utility function to create a new ternary search tree node

struct Node* newNode(char data)
{
struct Node* temp = (struct Node*) malloc(sizeof( struct Node ));
temp->data = data;
temp->isEndOfString = 0;
temp->left = temp->eq = temp->right = NULL;
return temp;
}
// Function to insert a new word in a Ternary Search Tree

void insert(struct Node** root, char *word)
{
// Base Case: Tree is empty
if (!(*root))
*root = newNode(*word);
// If current character of word is smaller than root's character,

// then insert this word in left subtree of root
if ((*word) < (*root)->data)
insert(&( (*root)->left ), word);
// If current character of word is greate than root's character,

// then insert this word in right subtree of root
else if ((*word) > (*root)->data)
insert(&( (*root)->right ), word);
// If current character of word is same as root's character,

else
{
if (*(word+1))
insert(&( (*root)->eq ), word+1);
// the last character of the word

else
(*root)->isEndOfString = 1;
}
}
// A recursive function to traverse Ternary Search Tree

void traverseTSTUtil(struct Node* root, char* buffer, int depth)
{
if (root)
{
// First traverse the left subtree
traverseTSTUtil(root->left, buffer, depth);
// Store the character of this node

buffer[depth] = root->data;
if (root->isEndOfString)
{
buffer[depth+1] = '\0';
printf( "%s\n", buffer);
}
// Traverse the subtree using equal pointer (middle subtree)

traverseTSTUtil(root->eq, buffer, depth + 1);
// Finally Traverse the right subtree

traverseTSTUtil(root->right, buffer, depth);
}
}
// The main function to traverse a Ternary Search Tree.

// It mainly uses traverseTSTUtil()
void traverseTST(struct Node* root)
{
char buffer[MAX];
traverseTSTUtil(root, buffer, 0);
}
// Function to search a given word in TST

int searchTST(struct Node *root, char *word)
{
if (!root)
return 0;
if (*word < (root)->data)

return searchTST(root->left, word);
else if (*word > (root)->data)

return searchTST(root->right, word);
else
{
if (*(word+1) == '\0')
return root->isEndOfString;
return searchTST(root->eq, word+1);

}
}

int main()
{
struct Node *root = NULL;
insert(&root, "cat");
insert(&root, "cats");
insert(&root, "up");
insert(&root, "bug");
printf("Following is traversal of ternary search tree\n");

traverseTST(root);
printf("\nFollowing are search results for cats, bu and cat respectively\n");

searchTST(root, "cats")? printf("Found\n"): printf("Not Found\n");
searchTST(root, "bu")?
printf("Found\n"): printf("Not Found\n");
searchTST(root, "cat")? printf("Found\n"): printf("Not Found\n");
return 0;
}
Output:
Followingistraversalofternarysearchtree
bug
cat
cats
up
Followingaresearchresultsforcats,buandcatrespectively
Found
NotFound
Found
TimeComplexity:Thetimecomplexityoftheternarysearchtreeoperationsissimilartothatofbinary
searchtree.i.e.theinsertion,deletionandsearchoperationstaketimeproportionaltotheheightofthe
ternarysearchtree.Thespaceisproportionaltothelengthofthestringtobestored.
=======================================================================
Linkedcompletebinarytree&its
creation
Acompletebinarytreeisabinarytreewhereeachlevellexceptthelasthas2^lnodesandthenodesat
thelastlevelareallleftaligned.Completebinarytreesaremainlyusedinheapbaseddatastructures.
Thenodesinthecompletebinarytreeareinsertedfromlefttorightinonelevelatatime.Ifalevelisfull,the
nodeisinsertedinanewlevel.
Belowaresomeofthecompletebinarytrees.
1
/\
23
1
/\
23
/\/
456
Belowbinarytreesarenotcomplete:
1
/\
23
//
45
1
/\
23
/\/
456
/
7
Completebinarytreesaregenerallyrepresentedusingarrays.Thearrayrepresentationisbetterbecauseit
doesntcontainanyemptyslot.Givenparentindexi,itsleftchildisgivenby2*i+1anditsrightchildis
givenby2*i+2.Sonoextraspaceiswastedandspacetostoreleftandrightpointersissaved.However,
itmaybeaninterestingprogrammingquestiontocreatedaCompleteBinaryTreeusinglinked
representation.HereLinkedmeananonarrayrepresentationwhereleftandrightpointers(orreferences)
areusedtoreferleftandrightchildrenrespectively.Howtowriteaninsertfunctionthatalwaysaddsanew
nodeinthelastlevelandattheleftmostavailableposition?
Tocreatealinkedcompletebinarytree,weneedtokeeptrackofthenodesinalevelorderfashionsuch
thatthenextnodetobeinsertedliesintheleftmostposition.Aqueuedatastructurecanbeusedtokeep
trackoftheinsertednodes.
FollowingarestepstoinsertanewnodeinCompleteBinaryTree.
1.Ifthetreeisempty,initializetherootwithnewnode.
2.Else,getthefrontnodeofthequeue.
.Iftheleftchildofthisfrontnodedoesntexist,settheleftchildasthenewnode.
.elseiftherightchildofthisfrontnodedoesntexist,settherightchildasthenewnode.
3.Ifthefrontnodehasboththeleftchildandrightchild,Dequeue()it.
4.Enqueue()thenewnode.
Belowistheimplementation:
// Program for linked implementation of complete binary tree
#include <stdio.h>
#include <stdlib.h>
// For Queue Size

#define SIZE 50
// A tree node
struct node
{
int data;
struct node *right,*left;
};
// A queue node
struct Queue
{
int front, rear;
int size;
struct node* *array;
};

{
struct node* temp = (struct node*) malloc(sizeof( struct node ));
temp->data = data;
return temp;
}
// A utility function to create a new Queue

struct Queue* createQueue(int size)
{
struct Queue* queue = (struct Queue*) malloc(sizeof( struct Queue ));
queue->front = queue->rear = -1;

queue->size = size;
queue->array = (struct node**) malloc(queue->size * sizeof( struct node* ));
int i;
for (i = 0; i < size; ++i)
queue->array[i] = NULL;
return queue;
}
// Standard Queue Functions

int isEmpty(struct Queue* queue)
{
return queue->front == -1;
int isFull(struct Queue* queue)

{ return queue->rear == queue->size - 1; }
int hasOnlyOneItem(struct Queue* queue)

{ return queue->front == queue->rear; }
void Enqueue(struct node *root, struct Queue* queue)

{
if (isFull(queue))
return;
queue->array[++queue->rear] = root;
if (isEmpty(queue))
++queue->front;
}
struct node* Dequeue(struct Queue* queue)

{
if (isEmpty(queue))
return NULL;
struct node* temp = queue->array[queue->front];
if (hasOnlyOneItem(queue))
queue->front = queue->rear = -1;
else
++queue->front;
return temp;
}
struct node* getFront(struct Queue* queue)

{ return queue->array[queue->front]; }
// A utility function to check if a tree node has both left and right children
int hasBothChild(struct node* temp)
{
return temp && temp->left && temp->right;
}
// Function to insert a new node in complete binary tree

void insert(struct node **root, int data, struct Queue* queue)
{
// Create a new node for given data
struct node *temp = newNode(data);
// If the tree is empty, initialize the root with new node.

if (!*root)
*root = temp;
else
{
// get the front node of the queue.
struct node* front = getFront(queue);
// If the left child of this front node doesnt exist, set the
// left child as the new node
if (!front->left)
front->left = temp;
// If the right child of this front node doesnt exist, set the
// right child as the new node
else if (!front->right)
front->right = temp;
// If the front node has both the left child and right child,
// Dequeue() it.
if (hasBothChild(front))
Dequeue(queue);
}
// Enqueue() the new node for later insertions

Enqueue(temp, queue);
}
// Standard level order traversal to test above function

void levelOrder(struct node* root)
{
struct Queue* queue = createQueue(SIZE);
Enqueue(root, queue);
while (!isEmpty(queue))
{
struct node* temp = Dequeue(queue);
printf("%d ", temp->data);
if (temp->left)
Enqueue(temp->left, queue);
if (temp->right)
Enqueue(temp->right, queue);
}
}

int main()
{
struct node* root = NULL;
struct Queue* queue = createQueue(SIZE);
int i;
for(i = 1; i <= 12; ++i)

insert(&root, i, queue);
levelOrder(root);
return 0;
}
Output:
123456789101112
=======================================================================
MorristraversalforPreorder
UsingMorrisTraversal,wecantraversethetreewithoutusingstackandrecursion.Thealgorithmfor
PreorderisalmostsimilartoMorristraversalforInorder.
1...Ifleftchildisnull,printthecurrentnodedata.Movetorightchild.
.Else,Maketherightchildoftheinorderpredecessorpointtothecurrentnode.Twocasesarise:
a)Therightchildoftheinorderpredecessoralreadypointstothecurrentnode.Setrightchildto
NULL.Movetorightchildofcurrentnode.
b)TherightchildisNULL.Setittocurrentnode.Printcurrentnodesdataandmovetoleftchildof
currentnode.
2...IterateuntilcurrentnodeisnotNULL.
FollowingisCimplementationoftheabovealgorithm.
// C program for Morris Preorder traversal
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
};

{
struct node* temp = (struct node*) malloc(sizeof(struct node));
temp->data = data;
return temp;
// Preorder traversal without recursion and without stack

void morrisTraversalPreorder(struct node* root)
{
while (root)
{
// If left child is null, print the current node data. Move to
// right child.
if (root->left == NULL)
{
printf( "%d ", root->data );
root = root->right;
}
else
{
// Find inorder predecessor
struct node* current = root->left;
while (current->right && current->right != root)
current = current->right;
// If the right child of inorder predecessor already points to

// this node
if (current->right == root)
{
current->right = NULL;
root = root->right;
}
// If right child doesn't point to this node, then print this

// node and make right child point to this node
else
{
current->right = root;
root = root->left;
}
}
}
}
// Function for sStandard preorder traversal

void preorder(struct node* root)
{
if (root)
{
printf( "%d ", root->data);
preorder(root->left);
preorder(root->right);
}
}

int main()
{
root = newNode(1);
morrisTraversalPreorder(root);
printf("\n");
preorder(root);
return 0;
}
Output:
1248951011367
1248951011367
Limitations:
Morristraversalmodifiesthetreeduringtheprocess.Itestablishestherightlinkswhilemovingdownthe
treeandresetstherightlinkswhilemovingupthetree.Sothealgorithmcannotbeappliedifwrite
operationsarenotallowed.
=======================================================================
ConvertaBSTtoaBinaryTreesuch
thatsumofallgreaterkeysisadded
toeverykey
GivenaBinarySearchTree(BST),convertittoaBinaryTreesuchthateverykeyoftheoriginalBSTis
changedtokeyplussumofallgreaterkeysinBST.
Examples:
Input:RootoffollowingBST
5
/\
213
Output:ThegivenBSTisconvertedtofollowingBinaryTree
18
/\
2013
Source:ConvertaBST
Solution:DoreverseInoordertraversal.Keeptrackofthesumofnodesvisitedsofar.Letthissumbesum.
Foreverynodecurrentlybeingvisited,firstaddthekeyofthisnodetosum,i.e.sum=sum+node>key.
Thenchangethekeyofcurrentnodetosum,i.e.,node>key=sum.
WhenaBSTisbeingtraversedinreverseInorder,foreverykeycurrentlybeingvisited,allkeysthatare
alreadyvisitedareallgreaterkeys.
// Program to change a BST to Binary Tree such that key of a node becomes
// original key plus sum of all greater keys in BST
#include <stdio.h>
#include <stdlib.h>
/* A BST node has key, left child and right child */

struct node
{
int key;
struct node* left;
struct node* right;
};
NULL left and right pointers.*/
{
node->key = key;
node->left = NULL;
node->right = NULL;
return (node);
}
// A recursive function that traverses the given BST in reverse inorder and
// for every key, adds all greater keys to it
void addGreaterUtil(struct node *root, int *sum_ptr)
{
// Base Case
if (root == NULL)
return;
// Recur for right subtree first so that sum of all greater

// nodes is stored at sum_ptr
addGreaterUtil(root->right, sum_ptr);
// Update the value at sum_ptr

*sum_ptr = *sum_ptr + root->key;
// Update key of this node

root->key = *sum_ptr;
// Recur for left subtree so that the updated sum is added

// to smaller nodes
addGreaterUtil(root->left, sum_ptr);
}
// A wrapper over addGreaterUtil(). It initializes sum and calls

// addGreaterUtil() to recursivel upodate and use value of sum
void addGreater(struct node *root)
{
int sum = 0;
addGreaterUtil(root, &sum);
}
// A utility function to print inorder traversal of Binary Tree

void printInorder(struct node* node)
{
if (node == NULL)
return;
printf("%d ", node->key);
}

int main()
{
/* Create following BST
5
/
2 13 */
printf(" Inorder traversal of the given tree\n");

printInorder(root);
addGreater(root);
printf("\n Inorder traversal of the modified tree\n");

printInorder(root);
return 0;
}
Output:
Inordertraversalofthegiventree
2513
Inordertraversalofthemodifiedtree
201813
TimeComplexity:O(n)wherenisthenumberofnodesingivenBinarySearchTree.
=======================================================================
IterativePreorderTraversal
GivenaBinaryTree,writeaniterativefunctiontoprintPreordertraversalofthegivenbinarytree.
ReferthisforrecursivepreordertraversalofBinaryTree.Toconvertaninherentlyrecursiveproceduresto
iterative,weneedanexplicitstack.FollowingisasimplestackbasediterativeprocesstoprintPreorder
traversal.
1)CreateanemptystacknodeStackandpushrootnodetostack.
2)DofollowingwhilenodeStackisnotempty.
.a)Popanitemfromstackandprintit.
.b)Pushrightchildofpoppeditemtostack
.c)Pushleftchildofpoppeditemtostack
Rightchildispushedbeforeleftchildtomakesurethatleftsubtreeisprocessedfirst.
#include <stdlib.h>
#include <stdio.h>
#include <iostream>
#include <stack>
/* A binary tree node has data, left child and right child */
struct node
{
int data;
struct node* left;
struct node* right;
};
/* Helper function that allocates a new node with the given data and
{
struct node* node = new struct node;
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
// An iterative process to print preorder traversal of Binary tree

void iterativePreorder(node *root)
{
// Base Case
if (root == NULL)
return;
// Create an empty stack and push root to it

stack<node *> nodeStack;
nodeStack.push(root);
/* Pop all items one by one. Do following for every popped item
a) print it
b) push its right child
c) push its left child
Note that right child is pushed first so that left is processed first */
while (nodeStack.empty() == false)
{
// Pop the top item from stack and print it
struct node *node = nodeStack.top();
printf ("%d ", node->data);
nodeStack.pop();
// Push right and left children of the popped node to stack

if (node->right)
nodeStack.push(node->right);
if (node->left)
nodeStack.push(node->left);
}
}

int main()
{
/* Constructed binary tree is
10
/
/ \
3 5 2
*/
root->left
= newNode(8);
root->right
= newNode(2);
iterativePreorder(root);
return 0;
}
Output:
1083522
=======================================================================
FloorandCeilfromaBST
Therearenumerousapplicationsweneedtofindfloor(ceil)valueofakeyinabinarysearchtreeorsorted
array.Forexample,considerdesigningmemorymanagementsysteminwhichfreenodesarearrangedin
BST.Findbestfitfortheinputrequest.
CeilValueNode:Nodewithsmallestdatalargerthanorequaltokeyvalue.
Imaginewearemovingdownthetree,andassumewearerootnode.Thecomparisonyieldsthree
possibilities,
A)Rootdataisequaltokey.Wearedone,rootdataisceilvalue.
B)Rootdata<keyvalue,certainlytheceilvaluecantbeinleftsubtree.Proceedtosearchonrightsubtree
asreducedprobleminstance.
C)Rootdata>keyvalue,theceilvaluemaybeinleftsubtree.Wemayfindanodewithislargerdatathan
keyvalueinleftsubtree,ifnottherootitselfwillbeceilnode.
HereiscodeinCforceilvalue.
// Program to find ceil of a given value in BST
#include <stdio.h>
#include <stdlib.h>
/* A binary tree node has key, left child and right child */
struct node
{
int key;
struct node* left;
struct node* right;
};

{
node->key = key;
node->left = NULL;
node->right = NULL;
return(node);
}
// Function to find ceil of a given input in BST. If input is more

// than the max key in BST, return -1
int Ceil(node *root, int input)
{
// Base case
if( root == NULL )
return -1;
// We found equal key

if( root->key == input )
return root->key;
// If root's key is smaller, ceil must be in right subtree

if( root->key < input )
return Ceil(root->right, input);
// Else, either left subtree or root has the ceil value

int ceil = Ceil(root->left, input);
return (ceil >= input) ? ceil : root->key;
}

int main()
{
for(int i = 0; i < 16; i++)

printf("%d %d\n", i, Ceil(root, i));
return 0;
}
Output:
02
12
22
34
44
56
66
78
88
910
1010
1112
1212
1314
1414
151
=======================================================================
ConstructBSTfromgivenpreorder
traversal|Set2
Givenpreordertraversalofabinarysearchtree,constructtheBST.
Forexample,ifthegiventraversalis{10,5,1,7,40,50},thentheoutputshouldberootoffollowingtree.
10
/\
540
/\\
1750
WehavediscussedO(n^2)andO(n)recursivesolutionsinthepreviouspost.Followingisastackbased
iterativesolutionthatworksinO(n)time.
1.Createanemptystack.
2.Makethefirstvalueasroot.Pushittothestack.
3.Keeponpoppingwhilethestackisnotemptyandthenextvalueisgreaterthanstackstopvalue.Make
thisvalueastherightchildofthelastpoppednode.Pushthenewnodetothestack.
4.Ifthenextvalueislessthanthestackstopvalue,makethisvalueastheleftchildofthestackstopnode.
Pushthenewnodetothestack.
5.Repeatsteps2and3untilthereareitemsremaininginpre[].
/* A O(n) iterative program for construction of BST from preorder traversal */
#include <stdio.h>
#include <stdlib.h>
#include <limits.h>

typedef struct Node
{
int data;
} Node;
// A Stack has array of Nodes, capacity, and top

typedef struct Stack
{
int top;
int capacity;
Node* *array;
} Stack;

Node* newNode( int data )
{
Node* temp = (Node *)malloc( sizeof( Node ) );
temp->data = data;
return temp;
}
// A utility function to create a stack of given capacity

Stack* createStack( int capacity )
{
Stack* stack = (Stack *)malloc( sizeof( Stack ) );
stack->top = -1;
stack->capacity = capacity;
stack->array = (Node **)malloc( stack->capacity * sizeof( Node* ) );
return stack;
}
// A utility function to check if stack is full

int isFull( Stack* stack )
{
return stack->top == stack->capacity - 1;
}
// A utility function to check if stack is empty

int isEmpty( Stack* stack )
{
}
// A utility function to push an item to stack
void push( Stack* stack, Node* item )

{
if( isFull( stack ) )
return;
stack->array[ ++stack->top ] = item;
}
// A utility function to remove an item from stack

Node* pop( Stack* stack )
{
if( isEmpty( stack ) )
return NULL;
return stack->array[ stack->top-- ];
}
// A utility function to get top node of stack

Node* peek( Stack* stack )
{
return stack->array[ stack->top ];
}
// The main function that constructs BST from pre[]

Node* constructTree ( int pre[], int size )
{
// Create a stack of capacity equal to size
Stack* stack = createStack( size );
// The first element of pre[] is always root

Node* root = newNode( pre[0] );
// Push root
push( stack, root );
int i;
Node* temp;
// Iterate through rest of the size-1 items of given preorder array
for ( i = 1; i < size; ++i )

{
temp = NULL;
/* Keep on popping while the next value is greater than

stack's top value. */
while ( !isEmpty( stack ) && pre[i] > peek( stack )->data )
temp = pop( stack );
// Make this greater value as the right child and push it to the stack
if ( temp != NULL)
{
temp->right = newNode( pre[i] );
push( stack, temp->right );
}
// If the next value is less than the stack's top value, make this value
// as the left child of the stack's top node. Push the new node to stack
else
{
peek( stack )->left = newNode( pre[i] );
push( stack, peek( stack )->left );
}
}
return root;
}
// A utility function to print inorder traversal of a Binary Tree

void printInorder (Node* node)
{
if (node == NULL)
return;

int main ()
{
int pre[] = {10, 5, 1, 7, 40, 50};
int size = sizeof( pre ) / sizeof( pre[0] );
Node *root = constructTree(pre, size);
printf("Inorder traversal of the constructed tree: \n");

printInorder(root);
return 0;
}
Output:
157104050
TimeComplexity:O(n).Thecomplexitylooksmorefromfirstlook.Ifwetakeacloserlook,wecanobserve
thateveryitemispushedandpoppedonlyonce.Soatmost2npush/popoperationsareperformedinthe
mainloopsofconstructTree().Therefore,timecomplexityisO(n).
=======================================================================
ConstructFullBinaryTreefrom
givenpreorderandpostorder
traversals
Giventwoarraysthatrepresentpreorderandpostordertraversalsofafullbinarytree,constructthebinary
tree.
AFullBinaryTreeisabinarytreewhereeverynodehaseither0or2children
FollowingareexamplesofFullTrees.
1
/\
23
/\/\
4567
1
/\
23
/\
45
/\
67
1
/\
23
/\/\
4567
/\
89
ItisnotpossibletoconstructageneralBinaryTreefrompreorderandpostordertraversals(Seethis).Butif
knowthattheBinaryTreeisFull,wecanconstructthetreewithoutambiguity.Letusunderstandthiswith
thehelpoffollowingexample.
Letusconsiderthetwogivenarraysaspre[]={1,2,4,8,9,5,3,6,7}andpost[]={8,9,4,5,2,6,7,3,1}
Inpre[],theleftmostelementisrootoftree.Sincethetreeisfullandarraysizeismorethan1.Thevalue
nextto1inpre[],mustbeleftchildofroot.Soweknow1isrootand2isleftchild.Howtofindtheallnodes
inleftsubtree?Weknow2isrootofallnodesinleftsubtree.Allnodesbefore2inpost[]mustbeinleft
subtree.Nowweknow1isroot,elements{8,9,4,5,2}areinleftsubtree,andtheelements{6,7,3}arein
rightsubtree.
1
/\
/\
{8,9,4,5,2}{6,7,3}
Werecursivelyfollowtheaboveapproachandgetthefollowingtree.
1
/\
23
/\/\
4567
/\
89
/* program for construction of full binary tree */

#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node *left;
struct node *right;
};

struct node* newNode (int data)
{
temp->data = data;
return temp;
}
// A recursive function to construct Full from pre[] and post[].

// preIndex is used to keep track of index in pre[].
// l is low index and h is high index for the current subarray in post[]
struct node* constructTreeUtil (int pre[], int post[], int* preIndex,
int l, int h, int size)

{
// Base case
if (*preIndex >= size || l > h)
return NULL;
// The first node in preorder traversal is root. So take the node at

// preIndex from preorder and make it root, and increment preIndex
struct node* root = newNode ( pre[*preIndex] );
++*preIndex;
// If the current subarry has only one element, no need to recur

if (l == h)
return root;
// Search the next element of pre[] in post[]

int i;
for (i = l; i <= h; ++i)
if (pre[*preIndex] == post[i])
break;
// Use the index of element found in postorder to divide postorder array in

// two parts. Left subtree and right subtree
if (i <= h)
{
root->left = constructTreeUtil (pre, post, preIndex, l, i, size);
root->right = constructTreeUtil (pre, post, preIndex, i + 1, h, size);
}
return root;
}
// The main function to construct Full Binary Tree from given preorder and
// postorder traversals. This function mainly uses constructTreeUtil()
struct node *constructTree (int pre[], int post[], int size)
{
int preIndex = 0;
return constructTreeUtil (pre, post, &preIndex, 0, size - 1, size);

}

void printInorder (struct node* node)
{
if (node == NULL)
return;
}

int main ()
{
int pre[] = {1, 2, 4, 8, 9, 5, 3, 6, 7};
int post[] = {8, 9, 4, 5, 2, 6, 7, 3, 1};
int size = sizeof( pre ) / sizeof( pre[0] );
struct node *root = constructTree(pre, post, size);
printf("Inorder traversal of the constructed tree: \n");

printInorder(root);
return 0;
}
Output:
Inordertraversaloftheconstructedtree:
849251637
=======================================================================
TwonodesofaBSTareswapped,
correcttheBST
TwoofthenodesofaBinarySearchTree(BST)areswapped.Fix(orcorrect)theBST.
InputTree:
10
/\
58
/\
220
Intheabovetree,nodes20and8mustbeswappedtofixthetree.
Followingistheoutputtree
10
/\
520
/\
28
TheinordertraversalofaBSTproducesasortedarray.Soasimplemethodistostoreinordertraversalof
theinputtreeinanauxiliaryarray.Sorttheauxiliaryarray.Finally,inserttheauxiilaryarrayelementsbackto
theBST,keepingthestructureoftheBSTsame.TimecomplexityofthismethodisO(nLogn)andauxiliary
spaceneededisO(n).
WecansolvethisinO(n)timeandwithasingletraversalofthegivenBST.Sinceinordertraversalof
BSTisalwaysasortedarray,theproblemcanbereducedtoaproblemwheretwoelementsofasorted
arrayareswapped.Therearetwocasesthatweneedtohandle:
1.TheswappednodesarenotadjacentintheinordertraversaloftheBST.
Forexample,Nodes5and25areswappedin{357810152025}.
Theinordertraversalofthegiventreeis325781015205
Ifweobservecarefully,duringinordertraversal,wefindnode7issmallerthanthepreviousvisitednode25.
Heresavethecontextofnode25(previousnode).Again,wefindthatnode5issmallerthantheprevious
node20.Thistime,wesavethecontextofnode5(currentnode).Finallyswapthetwonodesvalues.
2.TheswappednodesareadjacentintheinordertraversalofBST.
Forexample,Nodes7and8areswappedin{357810152025}.
Theinordertraversalofthegiventreeis358710152025
Unlikecase#1,hereonlyonepointexistswhereanodevalueissmallerthanpreviousnodevalue.e.g.
node7issmallerthannode8.
HowtoSolve?Wewillmaintainthreepointers,first,middleandlast.Whenwefindthefirstpointwhere
currentnodevalueissmallerthanpreviousnodevalue,weupdatethefirstwiththepreviousnode&middle
withthecurrentnode.Whenwefindthesecondpointwherecurrentnodevalueissmallerthanprevious
nodevalue,weupdatethelastwiththecurrentnode.Incase#2,wewillneverfindthesecondpoint.So,
lastpointerwillnotbeupdated.Afterprocessing,ifthelastnodevalueisnull,thentwoswappednodesof
BSTareadjacent.
FollowingisCimplementationofthegivencode.
// Two nodes in the BST's swapped, correct the BST.
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
};
// A utility function to swap two integers

void swap( int* a, int* b )
{
int t = *a;
*a = *b;
*b = t;
}

{
struct node* node = (struct node *)malloc(sizeof(struct node));
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
// This function does inorder traversal to find out the two swapped nodes.
// It sets three pointers, first, middle and last. If the swapped nodes are
// adjacent to each other, then first and middle contain the resultant nodes
// Else, first and last contain the resultant nodes
void correctBSTUtil( struct node* root, struct node** first,
struct node** middle, struct node** last,
struct node** prev )
{
if( root )
{
// Recur for the left subtree
correctBSTUtil( root->left, first, middle, last, prev );
// If this node is smaller than the previous node, it's violating

// the BST rule.
if (*prev && root->data < (*prev)->data)
{
// If this is first violation, mark these two nodes as
// 'first' and 'middle'
if ( !*first )
{
*first = *prev;
*middle = root;
}
// If this is second violation, mark this node as last

else
*last = root;
}
// Mark this node as previous

*prev = root;
// Recur for the right subtree

correctBSTUtil( root->right, first, middle, last, prev );
}
}
// A function to fix a given BST where two nodes are swapped. This
// function uses correctBSTUtil() to find out two nodes and swaps the
// nodes to fix the BST
void correctBST( struct node* root )
{
// Initialize pointers needed for correctBSTUtil()
struct node *first, *middle, *last, *prev;
first = middle = last = prev = NULL;
// Set the poiters to find out two nodes

correctBSTUtil( root, &first, &middle, &last, &prev );
// Fix (or correct) the tree

if( first && last )
swap( &(first->data), &(last->data) );
else if( first && middle ) // Adjacent nodes swapped
swap( &(first->data), &(middle->data) );
// else nodes have not been swapped, passed tree is really BST.
}
/* A utility function to print Inoder traversal */

{
if (node == NULL)
return;
}

int main()
{
/*
6
/ \
10
/\
/\
3 7 12
10 and 2 are swapped

*/

root->left
= newNode(10);
root->right
= newNode(2);
printf("Inorder Traversal of the original tree \n");

printInorder(root);
correctBST(root);
printf("\nInorder Traversal of the fixed tree \n");

printInorder(root);
return 0;
}
Output:
InorderTraversaloftheoriginaltree
110367212
InorderTraversalofthefixedtree
123671012
TimeComplexity:O(n)
=======================================================================
BoundaryTraversalofbinarytree
Givenabinarytree,printboundarynodesofthebinarytreeAntiClockwisestartingfromtheroot.For
example,boundarytraversalofthefollowingtreeis208410142522
Webreaktheproblemin3parts:
1.Printtheleftboundaryintopdownmanner.
2.Printallleafnodesfromlefttoright,whichcanagainbesubdividedintotwosubparts:
..2.1Printallleafnodesofleftsubtreefromlefttoright.
..2.2Printallleafnodesofrightsubtreefromlefttoright.
3.Printtherightboundaryinbottomupmanner.
Weneedtotakecareofonethingthatnodesarenotprintedagain.e.g.Theleftmostnodeisalsotheleaf
nodeofthetree.
Basedontheabovecases,belowistheimplementation:
/* program for boundary traversal of a binary tree */
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
};
// A simple function to print leaf nodes of a binary tree

void printLeaves(struct node* root)
{
if ( root )
{
printLeaves(root->left);
// Print it if it is a leaf node

if ( !(root->left) && !(root->right) )
printLeaves(root->right);
}
}
// A function to print all left boundry nodes, except a leaf node.

// Print the nodes in TOP DOWN manner
void printBoundaryLeft(struct node* root)
{
if (root)
{
if (root->left)
{
// to ensure top down order, print the node
// before calling itself for left subtree
printBoundaryLeft(root->left);
}
else if( root->right )
{
printBoundaryLeft(root->right);
}
// do nothing if it is a leaf node, this way we avoid
// duplicates in output
}
// A function to print all right boundry nodes, except a leaf node

// Print the nodes in BOTTOM UP manner
void printBoundaryRight(struct node* root)
{
if (root)
{
if ( root->right )
{
// to ensure bottom up order, first call for right
// subtree, then print this node
printBoundaryRight(root->right);
}
else if ( root->left )
{
printBoundaryRight(root->left);
}
// do nothing if it is a leaf node, this way we avoid
// duplicates in output
}
}
// A function to do boundary traversal of a given binary tree

void printBoundary (struct node* root)
{
if (root)
{
// Print the left boundary in top-down manner.

printBoundaryLeft(root->left);
// Print all leaf nodes
printLeaves(root->left);
printLeaves(root->right);
// Print the right boundary in bottom-up manner

printBoundaryRight(root->right);
}
}

struct node* newNode( int data )
{
temp->data = data;
return temp;
}

int main()
{
struct node *root
= newNode(20);
root->left
= newNode(8);
root->left->left
= newNode(4);
root->left->right
= newNode(12);
= newNode(10);
root->right
= newNode(22);
root->right->right
= newNode(25);
printBoundary( root );
return 0;
}
Output:
208410142522
TimeComplexity:O(n)wherenisthenumberofnodesinbinarytree.
=======================================================================
CheckwhetheragivenBinaryTree
isCompleteornot
GivenaBinaryTree,writeafunctiontocheckwhetherthegivenBinaryTreeisCompleteBinaryTreeornot.
Acompletebinarytreeisabinarytreeinwhicheverylevel,exceptpossiblythelast,iscompletelyfilled,and
allnodesareasfarleftaspossible.Seefollowingexamples.
ThefollowingtreesareexamplesofCompleteBinaryTrees
1
/\
23
1
/\
23
/
4
1
/\
23
/\/
456
ThefollowingtreesareexamplesofNonCompleteBinaryTrees
1
\
3
1
/\
23
\/\
456
1
/\
23
/\
45
Source:Writeanalgorithmtocheckifatreeiscompletebinarytreeornot
Themethod2oflevelordertraversalpostcanbeeasilymodifiedtocheckwhetheratreeisCompleteor
not.Tounderstandtheapproach,letusfirstdefineatermFullNode.AnodeisFullNodeifbothleftand
rightchildrenarenotempty(ornotNULL).
Theapproachistodoalevelordertraversalstartingfromroot.Inthetraversal,onceanodeisfoundwhich
isNOTaFullNode,allthefollowingnodesmustbeleafnodes.
Also,onemorethingneedstobecheckedtohandlethebelowcase:Ifanodehasemptyleftchild,thenthe
rightchildmustbeempty.
1
/\
23
\
4
ThankstoGudduSharmaforsuggestingthissimpleandefficientapproach.
// A program to check if a given binary tree is complete or not
#include <stdio.h>
#include <stdlib.h>
#include <stdbool.h>
#define MAX_Q_SIZE 500

struct node
{
int data;
struct node* left;
struct node* right;
};
/* frunction prototypes for functions needed for Queue data

structure. A queue is needed for level order tarversal */
struct node** createQueue(int *, int *);
void enQueue(struct node **, int *, struct node *);
struct node *deQueue(struct node **, int *);
bool isQueueEmpty(int *front, int *rear);
/* Given a binary tree, return true if the tree is complete

else false */
bool isCompleteBT(struct node* root)
{
// Base Case: An empty tree is complete Binary Tree
if (root == NULL)
return true;
// Create an empty queue

int rear, front;
struct node **queue = createQueue(&front, &rear);
// Create a flag variable which will be set true

// when a non full node is seen
bool flag = false;
// Do level order traversal using queue.

enQueue(queue, &rear, root);
while(!isQueueEmpty(&front, &rear))
{
struct node *temp_node = deQueue(queue, &front);
/* Ceck if left child is present*/

if(temp_node->left)
{
// If we have seen a non full node, and we see a node
// with non-empty left child, then the given tree is not
// a complete Binary Tree
if (flag == true)
return false;
enQueue(queue, &rear, temp_node->left); // Enqueue Left Child

}
else // If this a non-full node, set the flag as true
flag = true;
/* Ceck if right child is present*/

if(temp_node->right)
{
// If we have seen a non full node, and we see a node
// with non-empty left child, then the given tree is not
// a complete Binary Tree
if(flag == true)
return false;
enQueue(queue, &rear, temp_node->right); // Enqueue Right Child

}
else // If this a non-full node, set the flag as true
flag = true;
}
// If we reach here, then the tree is complete Bianry Tree

return true;
}
/*UTILITY FUNCTIONS*/
struct node** createQueue(int *front, int *rear)
{
struct node **queue =
(struct node **)malloc(sizeof(struct node*)*MAX_Q_SIZE);
*front = *rear = 0;
return queue;
}
void enQueue(struct node **queue, int *rear, struct node *new_node)

{
queue[*rear] = new_node;
(*rear)++;
}
struct node *deQueue(struct node **queue, int *front)

{
(*front)++;
return queue[*front - 1];
}
bool isQueueEmpty(int *front, int *rear)

{
return (*rear == *front);
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
/* Let us construct the following Binary Tree which
is not a complete Binary Tree
1
/
/\ \
4
56
*/

root->left
= newNode(2);
root->right
= newNode(3);
root->left->left
= newNode(4);
if ( isCompleteBT(root) == true )
printf ("Complete Binary Tree");
else
printf ("NOT Complete Binary Tree");
return 0;
}
Output:
NOTCompleteBinaryTree
TimeComplexity:O(n)wherenisthenumberofnodesingivenBinaryTree
AuxiliarySpace:O(n)forqueue.
=======================================================================
CheckifeachinternalnodeofaBST
hasexactlyonechild
GivenPreordertraversalofaBST,checkifeachnonleafnodehasonlyonechild.AssumethattheBST
containsuniqueentries.
Examples
Input:pre[]={20,10,11,13,12}
Output:Yes
ThegivearrayrepresentsfollowingBST.InthefollowingBST,everyinternal
nodehasexactly1child.Therefor,theoutputistrue.
20
/
10
\
11
\
13
/
12
InPreordertraversal,descendants(orPreordersuccessors)ofeverynodeappearafterthenode.Inthe
aboveexample,20isthefirstnodeinpreorderandalldescendantsof20appearafterit.Alldescendantsof
20aresmallerthanit.For10,alldescendantsaregreaterthanit.Ingeneral,wecansay,ifallinternalnodes
haveonlyonechildinaBST,thenallthedescendantsofeverynodeareeithersmallerorlargerthanthe
node.Thereasonissimple,sincethetreeisBSTandeverynodehasonlyonechild,alldescendantsofa
nodewilleitherbeonleftsideorrightside,meansalldescendantswilleitherbesmallerorgreater.
Approach1(Naive)
Thisapproachsimplyfollowstheaboveideathatallvaluesonrightsideareeithersmallerorlarger.Usetwo
loops,theouterlooppicksanelementonebyone,startingfromtheleftmostelement.Theinnerloopchecks
ifallelementsontherightsideofthepickedelementareeithersmallerorgreater.Thetimecomplexityof
thismethodwillbeO(n^2).
Approach2
Sinceallthedescendantsofanodemusteitherbelargerorsmallerthanthenode.Wecandofollowingfor
everynodeinaloop.
1.Findthenextpreordersuccessor(ordescendant)ofthenode.
2.Findthelastpreordersuccessor(lastelementinpre[])ofthenode.
3.Ifbothsuccessorsarelessthanthecurrentnode,orbothsuccessorsaregreaterthanthecurrentnode,
thencontinue.Else,returnfalse.
#include <stdio.h>
bool hasOnlyOneChild(int pre[], int size)

{
int nextDiff, lastDiff;
for (int i=0; i<size-1; i++)

{
nextDiff = pre[i] - pre[i+1];

lastDiff = pre[i] - pre[size-1];
if (nextDiff*lastDiff < 0)
return false;;
}
return true;
}
// driver program to test above function

int main()
{
int pre[] = {8, 3, 5, 7, 6};
int size = sizeof(pre)/sizeof(pre[0]);
if (hasOnlyOneChild(pre, size) == true )
printf("Yes");
else
printf("No");
return 0;
}
Output:
Yes
Approach3
1.Scanthelasttwonodesofpreorder&markthemasmin&max.
2.Scaneverynodedownthepreorderarray.Eachnodemustbeeithersmallerthantheminnodeorlarger
thanthemaxnode.Updatemin&maxaccordingly.
#include <stdio.h>
int hasOnlyOneChild(int pre[], int size)

{
// Initialize min and max using last two elements
int min, max;
if (pre[size-1] > pre[size-2])
{
max = pre[size-1];
min = pre[size-2]):
else
{
max = pre[size-2];
min = pre[size-1];
}
// Every element must be either smaller than min or

// greater than max
for (int i=size-3; i>=0; i--)
{
if (pre[i] < min)
min = pre[i];
else if (pre[i] > max)
max = pre[i];
else
return false;
}
return true;
}

int main()
{
int pre[] = {8, 3, 5, 7, 6};
int size = sizeof(pre)/sizeof(pre[0]);
if (hasOnlyOneChild(pre,size))
printf("Yes");
else
printf("No");
return 0;
}
Output:
Yes
=======================================================================
Constructaspecialtreefromgiven
preordertraversal
Givenanarraypre[]thatrepresentsPreordertraversalofaspacialbinarytreewhereeverynodehaseither
0or2children.OnemorearraypreLN[]isgivenwhichhasonlytwopossiblevaluesLandN.Thevalue
LinpreLN[]indicatesthatthecorrespondingnodeinBinaryTreeisaleafnodeandvalueNindicatesthat
thecorrespondingnodeisnonleafnode.Writeafunctiontoconstructthetreefromthegiventwoarrays.
Source:AmazonInterviewQuestion
Example:
Input:pre[]={10,30,20,5,15},preLN[]={'N','N','L','L','L'}
Output:Rootoffollowingtree
10
/\
3015
/\
205
Thefirstelementinpre[]willalwaysberoot.Sowecaneasilyfigureoutroot.Ifleftsubtreeisempty,the
rightsubtreemustalsobeemptyandpreLN[]entryforrootmustbeL.Wecansimplycreateanodeand
returnit.Ifleftandrightsubtreesarenotempty,thenrecursivelycallforleftandrightsubtreesandlinkthe
returnednodestoroot.
/* A program to construct Binary Tree from preorder traversal */
#include<stdio.h>
/* A binary tree node structure */

struct node
{
int data;
struct node *left;
struct node *right;
};
/* Utility function to create a new Binary Tree node */

{
struct node *temp = new struct node;

temp->data = data;
temp->left = NULL;
temp->right = NULL;
return temp;
}
/* A recursive function to create a Binary Tree from given pre[]

preLN[] arrays. The function returns root of tree. index_ptr is used
to update index values in recursive calls. index must be initially
passed as 0 */
struct node *constructTreeUtil(int pre[], char preLN[], int *index_ptr, int n)
{
int index = *index_ptr; // store the current value of index in pre[]
// Base Case: All nodes are constructed

if (index == n)
return NULL;
// Allocate memory for this node and increment index for

// subsequent recursive calls
struct node *temp = newNode ( pre[index] );
(*index_ptr)++;
// If this is an internal node, construct left and right subtrees and link the
subtrees
if (preLN[index] == 'N')
{
temp->left = constructTreeUtil(pre, preLN, index_ptr, n);
temp->right = constructTreeUtil(pre, preLN, index_ptr, n);
}
return temp;
}
// A wrapper over constructTreeUtil()

struct node *constructTree(int pre[], char preLN[], int n)
{
// Initialize index as 0. Value of index is used in recursion to maintain
// the current index in pre[] and preLN[] arrays.
int index = 0;
return constructTreeUtil (pre, preLN, &index, n);

}
/* This function is used only for testing */

{
if (node == NULL)
return;
/* first recur on left child */

printInorder (node->left);
/* then print the data of node */

/* now recur on right child */

printInorder (node->right);
}
/* Driver function to test above functions */

int main()
{
/* Constructing tree given in the above figure

10
/ \
30
15
/ \
20
5 */
int pre[] = {10, 30, 20, 5, 15};
char preLN[] = {'N', 'N', 'L', 'L', 'L'};

int n = sizeof(pre)/sizeof(pre[0]);
// construct the above tree

root = constructTree (pre, preLN, n);
// Test the constructed tree

printf("Following is Inorder Traversal of the Constructed Binary Tree: \n");
printInorder (root);
return 0;
}
Output:
FollowingisInorderTraversaloftheConstructedBinaryTree:
203051015
TimeComplexity:O(n)
=======================================================================
ConstructSpecialBinaryTreefrom
givenInordertraversal
GivenInorderTraversalofaSpecialBinaryTreeinwhichkeyofeverynodeisgreaterthankeysinleftand
rightchildren,constructtheBinaryTreeandreturnroot.
Examples:
Input:inorder[]={5,10,40,30,28}
Output:rootoffollowingtree
40
/\
1030
/\
528
TheideausedinConstructionofTreefromgivenInorderandPreordertraversalscanbeusedhere.Letthe
givenarrayis{1,5,10,40,30,15,28,20}.Themaximumelementingivenarraymustberoot.The
elementsonleftsideofthemaximumelementareinleftsubtreeandelementsonrightsideareinright
subtree.
40
/\
{1,5,10}{30,15,28,20}
Werecursivelyfollowabovestepforleftandrightsubtrees,andfinallygetthefollowingtree.
40
/\
1030
/\
528
//\
11520
Algorithm:buildTree()
1)Findindexofthemaximumelementinarray.ThemaximumelementmustberootofBinaryTree.
2)Createanewtreenoderootwiththedataasthemaximumvaluefoundinstep1.
3)CallbuildTreeforelementsbeforethemaximumelementandmakethebuilttreeasleftsubtreeofroot.
5)CallbuildTreeforelementsafterthemaximumelementandmakethebuilttreeasrightsubtreeofroot.
6)returnroot.
Implementation:FollowingisC/C++implementationoftheabovealgorithm.
/* program to construct tree from inorder traversal */
#include<stdio.h>
#include<stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
/* Prototypes of a utility function to get the maximum

value in inorder[start..end] */
int max(int inorder[], int strt, int end);
/* A utility function to allocate memory for a node */

/* Recursive function to construct binary of size len from

Inorder traversal inorder[]. Initial values of start and end
should be 0 and len -1. */
struct node* buildTree (int inorder[], int start, int end)
{
if (start > end)
return NULL;
/* Find index of the maximum element from Binary Tree */

int i = max (inorder, start, end);
/* Pick the maximum value and make it root */

struct node *root = newNode(inorder[i]);
/* If this is the only element in inorder[start..end],

then return it */
if (start == end)
return root;
/* Using index in Inorder traversal, construct left and

right subtress */
root->left = buildTree (inorder, start, i-1);
root->right = buildTree (inorder, i+1, end);
return root;
}
/* UTILITY FUNCTIONS */
/* Function to find index of the maximum value in arr[start...end] */
int max (int arr[], int strt, int end)
{
int i, max = arr[strt], maxind = strt;
for(i = strt+1; i <= end; i++)

{
if(arr[i] > max)
{
max = arr[i];
maxind = i;
}
}
return maxind;
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return node;
}
/* This funtcion is here just to test buildTree() */

{
if (node == NULL)
return;




int main()
{
/* Assume that inorder traversal of following tree is given
40
/
10
/
5
\
30
\
28 */
int inorder[] = {5, 10, 40, 30, 28};

int len = sizeof(inorder)/sizeof(inorder[0]);
struct node *root = buildTree(inorder, 0, len - 1);

printf("\n Inorder traversal of the constructed tree is \n");
printInorder(root);
return 0;
}
Output:
Inordertraversaloftheconstructedtreeis
510403028
TimeComplexity:O(n^2)
=======================================================================
MergetwoBSTswithlimitedextra
space
GiventwoBinarySearchTrees(BST),printtheelementsofbothBSTsinsortedform.Theexpectedtime
complexityisO(m+n)wheremisthenumberofnodesinfirsttreeandnisthenumberofnodesinsecond
tree.MaximumallowedauxiliaryspaceisO(heightofthefirsttree+heightofthesecondtree).
Examples:
FirstBST
3
/\
15
SecondBST
4
/\
26
Output:123456
FirstBST
8
/\
210
/
1
SecondBST
5
/
3
/
0
Output:01235810
Source:Googleinterviewquestion
Asimilarquestionhasbeendiscussedearlier.Letusfirstdiscussalreadydiscussedmethodsofthe
previouspostwhichwasforBalancedBSTs.Themethod1canbeappliedherealso,butthetime
complexitywillbeO(n^2)inworstcase.Themethod2canalsobeappliedhere,buttheextraspace
requiredwillbeO(n)whichviolatestheconstraintgiveninthisquestion.Method3canbeappliedherebut
thestep3ofmethod3cantbedoneinO(n)foranunbalancedBST.
ThankstoKumarforsuggestingthefollowingsolution.
Theideaistouseiterativeinordertraversal.WeusetwoauxiliarystacksfortwoBSTs.Sinceweneedto
printtheelementsinsortedform,wheneverwegetasmallerelementfromanyofthetrees,weprintit.Ifthe
elementisgreater,thenwepushitbacktostackforthenextiteration.
#include<stdio.h>
#include<stdlib.h>
// Structure of a BST Node

struct node
{
int data;
struct node *left;
struct node *right;
};
//.................... START OF STACK RELATED STUFF....................

// A stack node
struct snode
{
struct node *t;
struct snode *next;
};
// Function to add an elemt k to stack

void push(struct snode **s, struct node *k)
{
struct snode *tmp = (struct snode *) malloc(sizeof(struct snode));
//perform memory check here

tmp->t = k;
tmp->next = *s;
(*s) = tmp;
}
// Function to pop an element t from stack

struct node *pop(struct snode **s)
{
struct node *t;
struct snode *st;
st=*s;
(*s) = (*s)->next;
t = st->t;
free(st);
return t;
}
// Fucntion to check whether the stack is empty or not

int isEmpty(struct snode *s)
{
if (s == NULL )
return 1;
return 0;
}
//.................... END OF STACK RELATED STUFF....................

{
temp->data = data;
temp->left = NULL;
temp->right = NULL;
return temp;
}
/* A utility function to print Inoder traversal of a Binary Tree */

{
if (root != NULL)
{
}
}
// The function to print data of two BSTs in sorted order

void merge(struct node *root1, struct node *root2)
{
// s1 is stack to hold nodes of first BST
struct snode *s1 = NULL;
// Current node of first BST

struct node *current1 = root1;
// s2 is stack to hold nodes of second BST

struct snode *s2 = NULL;
// Current node of second BST

struct node *current2 = root2;
// If first BST is empty, then output is inorder

// traversal of second BST
if (root1 == NULL)
{
inorder(root2);
return;
}
// If second BST is empty, then output is inorder
// traversal of first BST
if (root2 == NULL)
{
inorder(root1);
return ;
}
// Run the loop while there are nodes not yet printed.
// The nodes may be in stack(explored, but not printed)
// or may be not yet explored
while (current1 != NULL || !isEmpty(s1) ||
current2 != NULL || !isEmpty(s2))
{
// Following steps follow iterative Inorder Traversal
if (current1 != NULL || current2 != NULL )
{
// Reach the leftmost node of both BSTs and push ancestors of

// leftmost nodes to stack s1 and s2 respectively
if (current1 != NULL)
{
push(&s1, current1);
current1 = current1->left;
}
if (current2 != NULL)
{
current2 = current2->left;
}
}
else
{
// If we reach a NULL node and either of the stacks is empty,
// then one tree is exhausted, ptint the other tree
if (isEmpty(s1))
{
{
current2 = pop (&s2);
current2->left = NULL;
inorder(current2);
}
return ;
}
if (isEmpty(s2))
{
{
current1 = pop (&s1);
current1->left = NULL;
inorder(current1);
}
return ;
// Pop an element from both stacks and compare the

// popped elements
current1 = pop(&s1);
current2 = pop(&s2);
// If element of first tree is smaller, then print it

// and push the right subtree. If the element is larger,
// then we push it back to the corresponding stack.
if (current1->data < current2->data)
{
printf("%d ", current1->data);
current1 = current1->right;
current2 = NULL;
}
else
{
printf("%d ", current2->data);
current2 = current2->right;
current1 = NULL;
}
}
}
}

int main()
{
struct node *root1 = NULL, *root2 = NULL;
/* Let us create the following tree as first tree

3
/ \
1
*/
root1 = newNode(3);
root1->left = newNode(1);
root1->right = newNode(5);
/* Let us create the following tree as second tree

4
/ \
2
*/
root2 = newNode(4);
// Print sorted nodes of both trees

merge(root1, root2);
return 0;
}
TimeComplexity:O(m+n)
AuxiliarySpace:O(heightofthefirsttree+heightofthesecondtree)
=======================================================================
Findthemaximumsumleaftoroot
pathinaBinaryTree
GivenaBinaryTree,findthemaximumsumpathfromaleaftoroot.Forexample,inthefollowingtree,there
arethreeleaftorootpaths8>2>10,4>2>10and7>10.Thesumsofthesethreepathsare16,4and
17respectively.Themaximumofthemis17andthepathformaximumis7>10.
10
/\
27
/\
84
Solution
1)Firstfindtheleafnodethatisonthemaximumsumpath.InthefollowingcodegetTargetLeaf()doesthis
byassigningtheresultto*target_leaf_ref.
2)Oncewehavethetargetleafnode,wecanprintthemaximumsumpathbytraversingthetree.Inthe
followingcode,printPath()doesthis.
ThemainfunctionismaxSumPath()thatusesabovetwofunctionstogetthecompletesolution.
#include<stdio.h>
#include<limits.h>
/* A tree node structure */

struct node
{
int data;
struct node *left;
struct node *right;
};
// A utility function that prints all nodes on the path from root to target_leaf
bool printPath (struct node *root, struct node *target_leaf)
{
// base case
if (root == NULL)
return false;
// return true if this node is the target_leaf or target leaf is present in

// one of its descendants
if (root == target_leaf || printPath(root->left, target_leaf) ||
printPath(root->right, target_leaf) )
{
return true;
}
return false;
}
// This function Sets the target_leaf_ref to refer the leaf node of the maximum
// path sum. Also, returns the max_sum using max_sum_ref

void getTargetLeaf (struct node *node, int *max_sum_ref, int curr_sum,
struct node **target_leaf_ref)
{
if (node == NULL)
return;
// Update current sum to hold sum of nodes on path from root to this node
curr_sum = curr_sum + node->data;
// If this is a leaf node and path to this node has maximum sum so far,
// then make this node target_leaf
if (node->left == NULL && node->right == NULL)
{
if (curr_sum > *max_sum_ref)
{
*max_sum_ref = curr_sum;
*target_leaf_ref = node;
}
}
// If this is not a leaf node, then recur down to find the target_leaf
getTargetLeaf (node->left, max_sum_ref, curr_sum, target_leaf_ref);
getTargetLeaf (node->right, max_sum_ref, curr_sum, target_leaf_ref);
}
// Returns the maximum sum and prints the nodes on max sum path
int maxSumPath (struct node *node)
{
// base case
if (node == NULL)
return 0;
struct node *target_leaf;

int max_sum = INT_MIN;
// find the target leaf and maximum sum
getTargetLeaf (node, &max_sum, 0, &target_leaf);
// print the path from root to the target leaf

printPath (node, target_leaf);
return max_sum; // return maximum sum

}

{
temp->data = data;
temp->left = NULL;
temp->right = NULL;
return temp;
}

int main()
{
/* Constructing tree given in the above figure */

root = newNode(10);
root->left = newNode(-2);
root->left->right = newNode(-4);
printf ("Following are the nodes on the maximum sum path \n");
int sum = maxSumPath(root);
printf ("\nSum of the nodes is %d ", sum);
getchar();
return 0;
}
Output:
Followingarethenodesonthemaximumsumpath
710
Sumofthenodesis17
TimeComplexity:TimecomplexityoftheabovesolutionisO(n)asitinvolvestreetraversaltwotimes.
=======================================================================
MergeTwoBalancedBinarySearch
Trees
Youaregiventwobalancedbinarysearchtreese.g.,AVLorRedBlackTree.Writeafunctionthatmerges
thetwogivenbalancedBSTsintoabalancedbinarysearchtree.Lettherebemelementsinfirsttreeandn
elementsintheothertree.YourmergefunctionshouldtakeO(m+n)time.
Inthefollowingsolutions,itisassumedthatsizesoftreesarealsogivenasinput.Ifthesizeisnotgiven,
thenwecangetthesizebytraversingthetree(Seethis).
Method1(Insertelementsoffirsttreetosecond)
TakeallelementsoffirstBSTonebyone,andinsertthemintothesecondBST.Insertinganelementtoa
selfbalancingBSTtakesLogntime(Seethis)wherenissizeoftheBST.Sotimecomplexityofthismethod
isLog(n)+Log(n+1)Log(m+n1).ThevalueofthisexpressionwillbebetweenmLognandmLog(m+n1).
Asanoptimization,wecanpickthesmallertreeasfirsttree.
Method2(MergeInorderTraversals)
1)Doinordertraversaloffirsttreeandstorethetraversalinonetemparrayarr1[].ThissteptakesO(m)
time.
2)Doinordertraversalofsecondtreeandstorethetraversalinanothertemparrayarr2[].Thissteptakes
O(n)time.
3)Thearrayscreatedinstep1and2aresortedarrays.Mergethetwosortedarraysintoonearrayofsizem
+n.ThissteptakesO(m+n)time.
4)Constructabalancedtreefromthemergedarrayusingthetechniquediscussedinthispost.Thisstep
takesO(m+n)time.
TimecomplexityofthismethodisO(m+n)whichisbetterthanmethod1.ThismethodtakesO(m+n)time
eveniftheinputBSTsarenotbalanced.
FollowingisC++implementationofthismethod.
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
// A utility unction to merge two sorted arrays into one

int *merge(int arr1[], int arr2[], int m, int n);
// A helper function that stores inorder traversal of a tree in inorder array

void storeInorder(struct node* node, int inorder[], int *index_ptr);
/* A function that constructs Balanced Binary Search Tree from a sorted array
See http://www.geeksforgeeks.org/archives/17138 */
struct node* sortedArrayToBST(int arr[], int start, int end);
/* This function merges two balanced BSTs with roots as root1 and root2.
m and n are the sizes of the trees respectively */
struct node* mergeTrees(struct node *root1, struct node *root2, int m, int n)
{
// Store inorder traversal of first tree in an array arr1[]
int *arr1 = new int[m];
int i = 0;
storeInorder(root1, arr1, &i);
// Store inorder traversal of second tree in another array arr2[]

int *arr2 = new int[n];
int j = 0;
storeInorder(root2, arr2, &j);
// Merge the two sorted array into one

int *mergedArr = merge(arr1, arr2, m, n);
// Construct a tree from the merged array and return root of the tree
return sortedArrayToBST (mergedArr, 0, m+n-1);
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
// A utility function to print inorder traversal of a given binary tree

{
if (node == NULL)
return;


}

int *merge(int arr1[], int arr2[], int m, int n)
{
// mergedArr[] is going to contain result
int *mergedArr = new int[m + n];
int i = 0, j = 0, k = 0;
// Traverse through both arrays

while (i < m && j < n)
{
// Pick the smaler element and put it in mergedArr
if (arr1[i] < arr2[j])
{
mergedArr[k] = arr1[i];
i++;
}
else
{
mergedArr[k] = arr2[j];
j++;
}
k++;
}
// If there are more elements in first array

while (i < m)
{
i++; k++;
}
// If there are more elements in second array

while (j < n)
{
j++; k++;
}
return mergedArr;
}
// A helper function that stores inorder traversal of a tree rooted with node
void storeInorder(struct node* node, int inorder[], int *index_ptr)

{
if (node == NULL)
return;

storeInorder(node->left, inorder, index_ptr);
inorder[*index_ptr] = node->data;
(*index_ptr)++; // increase index for next entry

storeInorder(node->right, inorder, index_ptr);
}
struct node* sortedArrayToBST(int arr[], int start, int end)
{
/* Base Case */
if (start > end)
return NULL;
/* Get the middle element and make it root */

int mid = (start + end)/2;
struct node *root = newNode(arr[mid]);
/* Recursively construct the left subtree and make it

left child of root */
root->left = sortedArrayToBST(arr, start, mid-1);
/* Recursively construct the right subtree and make it

right child of root */
root->right = sortedArrayToBST(arr, mid+1, end);
return root;
}

int main()
{
/* Create following tree as first balanced BST
100
/ \
50
300
/\
20 70
*/
struct node *root1 = newNode(100);
root1->left
= newNode(50);
root1->right
= newNode(300);
root1->left->left
= newNode(20);
root1->left->right = newNode(70);
/* Create following tree as second balanced BST

80
/ \
40
120
*/
root2->left
= newNode(40);
root2->right
= newNode(120);
struct node *mergedTree = mergeTrees(root1, root2, 5, 3);
printf ("Following is Inorder traversal of the merged tree \n");

printInorder(mergedTree);
getchar();
return 0;
}
Output:
FollowingisInordertraversalofthemergedtree
2040507080100120300
Method3(InPlaceMergeusingDLL)
WecanuseaDoublyLinkedListtomergetreesinplace.Followingarethesteps.
1)ConvertthegiventwoBinarySearchTreesintodoublylinkedlistinplace(Referthispostforthisstep).
2)MergethetwosortedLinkedLists(Referthispostforthisstep).
3)BuildaBalancedBinarySearchTreefromthemergedlistcreatedinstep2.(Referthispostforthisstep)
TimecomplexityofthismethodisalsoO(m+n)andthismethoddoesconversioninplace.
=======================================================================
AVLTree|Set2(Deletion)
WehavediscussedAVLinsertioninthepreviouspost.Inthispost,wewillfollowasimilarapproachfor
deletion.
Stepstofollowfordeletion.
TomakesurethatthegiventreeremainsAVLaftereverydeletion,wemustaugmentthestandardBST
deleteoperationtoperformsomerebalancing.Followingaretwobasicoperationsthatcanbeperformedto
rebalanceaBSTwithoutviolatingtheBSTproperty(keys(left)<key(root)<keys(right)).
1)LeftRotation
2)RightRotation
T1,T2andT3aresubtreesofthetreerootedwithy(onleftside)
orx(onrightside)
yx
/\RightRotation/\
xT3>T1y
/\</\
T1T2LeftRotationT2T3
Keysinbothoftheabovetreesfollowthefollowingorder
keys(T1)<key(x)<keys(T2)<key(y)<keys(T3)
SoBSTpropertyisnotviolatedanywhere.
Letwbethenodetobedeleted
1)PerformstandardBSTdeleteforw.
2)Startingfromw,travelupandfindthefirstunbalancednode.Letzbethefirstunbalancednode,ybethe
largerheightchildofz,andxbethelargerheightchildofy.Notethatthedefinitionsofxandyaredifferent
frominsertionhere.
3)Rebalancethetreebyperformingappropriaterotationsonthesubtreerootedwithz.Therecanbe4
possiblecasesthatneedstobehandledasx,yandzcanbearrangedin4ways.Followingarethepossible
4arrangements:
a)yisleftchildofzandxisleftchildofy(LeftLeftCase)
b)yisleftchildofzandxisrightchildofy(LeftRightCase)
c)yisrightchildofzandxisrightchildofy(RightRightCase)
d)yisrightchildofzandxisleftchildofy(RightLeftCase)
Likeinsertion,followingaretheoperationstobeperformedinabovementioned4cases.Notethat,unlike
insertion,fixingthenodezwontfixthecompleteAVLtree.Afterfixingz,wemayhavetofixancestorsofz
aswell(Seethisvideolectureforproof)
a)LeftLeftCase
T1,T2,T3andT4aresubtrees.
zy
/\/\
yT4RightRotate(z)xz
/\>/\/\
xT3T1T2T3T4
/\
T1T2
b)LeftRightCase
zzx
/\/\/\
yT4LeftRotate(y)xT4RightRotate(z)yz
/\>/\>/\/\
T1xyT3T1T2T3T4
/\/\
T2T3T1T2
c)RightRightCase
zy
/\/\
T1yLeftRotate(z)zx
/\>/\/\
T2xT1T2T3T4
/\
T3T4
d)RightLeftCase
zzx
/\/\/\
T1yRightRotate(y)T1xLeftRotate(z)zx
/\>/\>/\/\
xT4T2yT1T2T3T4
/\/\
T2T3T3T4
Unlikeinsertion,indeletion,afterweperformarotationatz,wemayhavetoperformarotationatancestors
ofz.Thus,wemustcontinuetotracethepathuntilwereachtheroot.
Cimplementation
FollowingistheCimplementationforAVLTreeDeletion.ThefollowingCimplementationusestherecursive
BSTdeleteasbasis.IntherecursiveBSTdelete,afterdeletion,wegetpointerstoallancestorsonebyone
inbottomupmanner.Sowedontneedparentpointertotravelup.Therecursivecodeitselftravelsupand
visitsalltheancestorsofthedeletednode.
1)PerformthenormalBSTdeletion.
2)Thecurrentnodemustbeoneoftheancestorsofthedeletednode.Updatetheheightofthecurrent
node.
3)Getthebalancefactor(leftsubtreeheightrightsubtreeheight)ofthecurrentnode.
4)Ifbalancefactorisgreaterthan1,thenthecurrentnodeisunbalancedandweareeitherinLeftLeftcase
orLeftRightcase.TocheckwhetheritisLeftLeftcaseorLeftRightcase,getthebalancefactorofleft
subtree.Ifbalancefactoroftheleftsubtreeisgreaterthanorequalto0,thenitisLeftLeftcase,elseLeft
Rightcase.
5)Ifbalancefactorislessthan1,thenthecurrentnodeisunbalancedandweareeitherinRightRightcase
orRightLeftcase.TocheckwhetheritisRightRightcaseorRightLeftcase,getthebalancefactorofright
subtree.Ifthebalancefactoroftherightsubtreeissmallerthanorequalto0,thenitisRightRightcase,
elseRightLeftcase.
#include<stdio.h>
#include<stdlib.h>
// An AVL tree node

struct node
{
int key;
struct node *left;
struct node *right;
int height;
};

int max(int a, int b);
// A utility function to get height of the tree

int height(struct node *N)
{
if (N == NULL)
return 0;
return N->height;
}

{
return (a > b)? a : b;
}
{
node->key
node->left
= key;
= NULL;
node->right = NULL;
node->height = 1; // new node is initially added at leaf
return(node);
}

struct node *rightRotate(struct node *y)
{
struct node *x = y->left;
struct node *T2 = x->right;
// Perform rotation
x->right = y;
y->left = T2;
// Update heights
y->height = max(height(y->left), height(y->right))+1;
x->height = max(height(x->left), height(x->right))+1;
// Return new root

return x;
}

{
struct node *T2 = y->left;
// Perform rotation
y->left = x;
x->right = T2;
// Update heights
// Return new root

return y;
}
// Get Balance factor of node N

int getBalance(struct node *N)
{
if (N == NULL)
return 0;
return height(N->left) - height(N->right);
}
struct node* insert(struct node* node, int key)

{
/* 1. Perform the normal BST rotation */
if (node == NULL)
return(newNode(key));

else
/* 2. Update height of this ancestor node */

node->height = max(height(node->left), height(node->right)) + 1;
/* 3. Get the balance factor of this ancestor node to check whether

this node became unbalanced */
int balance = getBalance(node);
// If this node becomes unbalanced, then there are 4 cases
// Left Left Case

if (balance > 1 && key < node->left->key)
return rightRotate(node);
// Right Right Case

if (balance < -1 && key > node->right->key)
return leftRotate(node);
// Left Right Case

if (balance > 1 && key > node->left->key)
{
node->left = leftRotate(node->left);
}
// Right Left Case

if (balance < -1 && key < node->right->key)
{
node->right = rightRotate(node->right);
}

return node;
}
/* Given a non-empty binary search tree, return the node with minimum
key value found in that tree. Note that the entire tree does not
need to be searched. */
struct node * minValueNode(struct node* node)
{
struct node* current = node;
/* loop down to find the leftmost leaf */

while (current->left != NULL)
current = current->left;
return current;
}
struct node* deleteNode(struct node* root, int key)

{
// STEP 1: PERFORM STANDARD BST DELETE
if (root == NULL)
return root;
// If the key to be deleted is smaller than the root's key,

// then it lies in left subtree
if ( key < root->key )
root->left = deleteNode(root->left, key);
// If the key to be deleted is greater than the root's key,

// then it lies in right subtree
else if( key > root->key )
root->right = deleteNode(root->right, key);
// if key is same as root's key, then This is the node

// to be deleted
else
{
// node with only one child or no child
if( (root->left == NULL) || (root->right == NULL) )
{
struct node *temp = root->left ? root->left : root->right;
// No child case
if(temp == NULL)
{
temp = root;
root = NULL;
}
else // One child case
*root = *temp; // Copy the contents of the non-empty child
free(temp);
}
else
{
// node with two children: Get the inorder successor (smallest
// in the right subtree)
struct node* temp = minValueNode(root->right);
// Copy the inorder successor's data to this node

root->key = temp->key;
// Delete the inorder successor

root->right = deleteNode(root->right, temp->key);
}
}
// If the tree had only one node then return

if (root == NULL)
return root;
// STEP 2: UPDATE HEIGHT OF THE CURRENT NODE

root->height = max(height(root->left), height(root->right)) + 1;
// STEP 3: GET THE BALANCE FACTOR OF THIS NODE (to check whether
// this node became unbalanced)
int balance = getBalance(root);
// Left Left Case

if (balance > 1 && getBalance(root->left) >= 0)
return rightRotate(root);
// Left Right Case

if (balance > 1 && getBalance(root->left) < 0)
{
return rightRotate(root);
}
// Right Right Case

if (balance < -1 && getBalance(root->right) <= 0)
return leftRotate(root);
// Right Left Case

if (balance < -1 && getBalance(root->right) > 0)
{
return leftRotate(root);
}
return root;
}

{
if(root != NULL)
{
}
}

int main()
{

/* The constructed AVL Tree would be

9
/ \
1 10
/ \\
0
/
11
/ \
-1
2 6
*/
printf("Pre order traversal of the constructed AVL tree is \n");

preOrder(root);
root = deleteNode(root, 10);
/* The AVL Tree after deletion of 10

1
/ \
0 9
/
-1
/ \
5
11
/ \
2 6
*/
printf("\nPre order traversal after deletion of 10 \n");

preOrder(root);
return 0;
}
Output:
PreordertraversaloftheconstructedAVLtreeis
91015261011
Preordertraversalafterdeletionof10
101952611
TimeComplexity:Therotationoperations(leftandrightrotate)takeconstanttimeasonlyfewpointersare
beingchangedthere.Updatingtheheightandgettingthebalancefactoralsotakeconstanttime.Sothetime
complexityofAVLdeleteremainssameasBSTdeletewhichisO(h)wherehisheightofthetree.Since
AVLtreeisbalanced,theheightisO(Logn).SotimecomplexityofAVLdeleteisO(Logn).
=======================================================================
VerticalSuminagivenBinaryTree
GivenaBinaryTree,findverticalsumofthenodesthatareinsameverticalline.Printallsumsthrough
differentverticallines.
Examples:
1
/\
23
/\/\
4567
Thetreehas5verticallines
VerticalLine1hasonlyonenode4=>verticalsumis4
VerticalLine2:hasonlyonenode2=>verticalsumis2
VerticalLine3:hasthreenodes:1,5,6=>verticalsumis1+5+6=12
Soexpectedoutputis4,2,12,3and7
Solution:
WeneedtochecktheHorizontalDistancesfromrootforallnodes.IftwonodeshavethesameHorizontal
Distance(HD),thentheyareonsameverticalline.TheideaofHDissimple.HDforrootis0,arightedge
(edgeconnectingtorightsubtree)isconsideredas+1horizontaldistanceandaleftedgeisconsideredas
1horizontaldistance.Forexample,intheabovetree,HDforNode4isat2,HDforNode2is1,HDfor5
and6is0andHDfornode7is+2.
WecandoinordertraversalofthegivenBinaryTree.Whiletraversingthetree,wecanrecursivelycalculate
HDs.Weinitiallypassthehorizontaldistanceas0forroot.Forleftsubtree,wepasstheHorizontalDistance
asHorizontaldistanceofrootminus1.Forrightsubtree,wepasstheHorizontalDistanceasHorizontal
Distanceofrootplus1.
FollowingisJavaimplementationforthesame.HashMapisusedtostoretheverticalsumsfordifferent
horizontaldistances.ThankstoNagesforsuggestingthismethod.
import java.util.HashMap;
// Class for a tree node

class TreeNode {
// data members
private int key;
private TreeNode left;
private TreeNode right;
// Accessor methods
public int key()
{ return key; }
public TreeNode left() { return left; }

public TreeNode right() { return right; }
// Constructor
public TreeNode(int key) { this.key = key; left = null; right = null; }
// Methods to set left and right subtrees

public void setLeft(TreeNode left)
{ this.left = left; }
public void setRight(TreeNode right) { this.right = right; }

}
// Class for a Binary Tree

class Tree {
private TreeNode root;
// Constructors
public Tree() { root = null; }
public Tree(TreeNode n) { root = n; }
// Method to be called by the consumer classes like Main class

public void VerticalSumMain() { VerticalSum(root); }
// A wrapper over VerticalSumUtil()

private void VerticalSum(TreeNode root) {
// base case
// Creates an empty hashMap hM

HashMap<Integer, Integer> hM = new HashMap<Integer, Integer>();
// Calls the VerticalSumUtil() to store the vertical sum values in hM

VerticalSumUtil(root, 0, hM);
// Prints the values stored by VerticalSumUtil()

if (hM != null) {
System.out.println(hM.entrySet());
}
}
// Traverses the tree in Inoorder form and builds a hashMap hM that

// contains the vertical sum
private void VerticalSumUtil(TreeNode root, int hD,
HashMap<Integer, Integer> hM) {
// base case
// Store the values in hM for left subtree

VerticalSumUtil(root.left(), hD - 1, hM);
// Update vertical sum for hD of this node

int prevSum = (hM.get(hD) == null) ? 0 : hM.get(hD);
hM.put(hD, prevSum + root.key());
// Store the values in hM for right subtree

VerticalSumUtil(root.right(), hD + 1, hM);
}
}
// Driver class to test the verticalSum methods

public class Main {

/* Create following Binary Tree
1
/\
2
/\
4
56
/\
7
*/
root.setLeft(new TreeNode(2));
root.setRight(new TreeNode(3));
root.left().setLeft(new TreeNode(4));
root.left().setRight(new TreeNode(5));
root.right().setLeft(new TreeNode(6));
root.right().setRight(new TreeNode(7));
Tree t = new Tree(root);
System.out.println("Following are the values of vertical sums with "

+ "the positions of the columns with respect to root ");
t.VerticalSumMain();
}
}
Seethisforasamplerun.
TimeComplexity:O(n)
==================================================================================
AVLTree|Set1(Insertion)
AVLtreeisaselfbalancingBinarySearchTree(BST)wherethedifferencebetweenheightsofleftandright
subtreescannotbemorethanoneforallnodes.
WhyAVLTrees?
MostoftheBSToperations(e.g.,search,max,min,insert,delete..etc)takeO(h)timewherehistheheight
oftheBST.ThecostoftheseoperationsmaybecomeO(n)foraskewedBinarytree.Ifwemakesurethat
heightofthetreeremainsO(Logn)aftereveryinsertionanddeletion,thenwecanguaranteeanupper
boundofO(Logn)foralltheseoperations.TheheightofanAVLtreeisalwaysO(Logn)wherenisthe
numberofnodesinthetree(Seethisvideolectureforproof).
Insertion
TomakesurethatthegiventreeremainsAVLaftereveryinsertion,wemustaugmentthestandardBST
insertoperationtoperformsomerebalancing.Followingaretwobasicoperationsthatcanbeperformedto
rebalanceaBSTwithoutviolatingtheBSTproperty(keys(left)<key(root)<keys(right)).
1)LeftRotation
2)RightRotation
T1,T2andT3aresubtreesofthetreerootedwithy(onleftside)
orx(onrightside)
yx
/\RightRotation/\
xT3>T1y
/\</\
T1T2LeftRotationT2T3
Keysinbothoftheabovetreesfollowthefollowingorder
keys(T1)<key(x)<keys(T2)<key(y)<keys(T3)
SoBSTpropertyisnotviolatedanywhere.
Stepstofollowforinsertion
Letthenewlynsertednodebew
1)PerformstandardBSTinsertforw.
2)Startingfromw,travelupandfindthefirstunbalancednode.Letzbethefirstunbalancednode,ybethe
childofzthatcomesonthepathfromwtozandxbethegrandchildofzthatcomesonthepathfromwto
z.
3)Rebalancethetreebyperformingappropriaterotationsonthesubtreerootedwithz.Therecanbe4
possiblecasesthatneedstobehandledasx,yandzcanbearrangedin4ways.Followingarethepossible
4arrangements:
a)yisleftchildofzandxisleftchildofy(LeftLeftCase)
b)yisleftchildofzandxisrightchildofy(LeftRightCase)
c)yisrightchildofzandxisrightchildofy(RightRightCase)
d)yisrightchildofzandxisleftchildofy(RightLeftCase)
Followingaretheoperationstobeperformedinabovementioned4cases.Inallofthecases,weonlyneed
torebalancethesubtreerootedwithzandthecompletetreebecomesbalancedastheheightofsubtree
(Afterappropriaterotations)rootedwithzbecomessameasitwasbeforeinsertion.(Seethisvideolecture
forproof)
a)LeftLeftCase
T1,T2,T3andT4aresubtrees.
zy
/\/\
yT4RightRotate(z)xz
/\>/\/\
xT3T1T2T3T4
/\
T1T2
b)LeftRightCase
zzx
/\/\/\
yT4LeftRotate(y)xT4RightRotate(z)yz
/\>/\>/\/\
T1xyT3T1T2T3T4
/\/\
T2T3T1T2
c)RightRightCase
zy
/\/\
T1yLeftRotate(z)zx
/\>/\/\
T2xT1T2T3T4
/\
T3T4
d)RightLeftCase
zzx
/\/\/\
T1yRightRotate(y)T1xLeftRotate(z)zy
/\>/\>/\/\
xT4T2yT1T2T3T4
/\/\
T2T3T3T4
Cimplementation
FollowingistheCimplementationforAVLTreeInsertion.ThefollowingCimplementationusestherecursive
BSTinserttoinsertanewnode.IntherecursiveBSTinsert,afterinsertion,wegetpointerstoallancestors
onebyoneinbottomupmanner.Sowedontneedparentpointertotravelup.Therecursivecodeitself
travelsupandvisitsalltheancestorsofthenewlyinsertednode.
1)PerformthenormalBSTinsertion.
2)Thecurrentnodemustbeoneoftheancestorsofthenewlyinsertednode.Updatetheheightofthe
currentnode.
3)Getthebalancefactor(leftsubtreeheightrightsubtreeheight)ofthecurrentnode.
4)Ifbalancefactorisgreaterthan1,thenthecurrentnodeisunbalancedandweareeitherinLeftLeftcase
orleftRightcase.Tocheckwhetheritisleftleftcaseornot,comparethenewlyinsertedkeywiththekeyin
leftsubtreeroot.
5)Ifbalancefactorislessthan1,thenthecurrentnodeisunbalancedandweareeitherinRightRightcase
orRightLeftcase.TocheckwhetheritisRightRightcaseornot,comparethenewlyinsertedkeywiththe
keyinrightsubtreeroot.
#include<stdio.h>
#include<stdlib.h>
// An AVL tree node

struct node
{
int key;
struct node *left;
struct node *right;
int height;
};

// A utility function to get height of the tree

int height(struct node *N)
{
if (N == NULL)
return 0;
return N->height;
}

{
return (a > b)? a : b;
}
{
node->key
node->left
= key;
= NULL;
node->right = NULL;
node->height = 1; // new node is initially added at leaf
return(node);
}

struct node *rightRotate(struct node *y)
{
struct node *x = y->left;
struct node *T2 = x->right;
// Perform rotation
x->right = y;
y->left = T2;
// Update heights
// Return new root

return x;
}

{
struct node *T2 = y->left;
// Perform rotation
y->left = x;
x->right = T2;
// Update heights
// Return new root

return y;
}
// Get Balance factor of node N

int getBalance(struct node *N)
{
if (N == NULL)
return 0;
return height(N->left) - height(N->right);
}
struct node* insert(struct node* node, int key)

{
/* 1. Perform the normal BST rotation */
if (node == NULL)
return(newNode(key));

else
/* 2. Update height of this ancestor node */

node->height = max(height(node->left), height(node->right)) + 1;
/* 3. Get the balance factor of this ancestor node to check whether

this node became unbalanced */
int balance = getBalance(node);
// Left Left Case

if (balance > 1 && key < node->left->key)
// Right Right Case

if (balance < -1 && key > node->right->key)
// Left Right Case

if (balance > 1 && key > node->left->key)
{
node->left = leftRotate(node->left);
}
// Right Left Case

if (balance < -1 && key < node->right->key)
{
node->right = rightRotate(node->right);
}
return node;
}

{
if(root != NULL)
{
}
}

int main()
{

/* The constructed AVL Tree would be

30
/ \
20
40
/ \\
10 2550
*/
printf("Pre order traversal of the constructed AVL tree is \n");
preOrder(root);
return 0;
}
Output:
PreordertraversaloftheconstructedAVLtreeis
302010254050
TimeComplexity:Therotationoperations(leftandrightrotate)takeconstanttimeasonlyfewpointersare
beingchangedthere.Updatingtheheightandgettingthebalancefactoralsotakeconstanttime.Sothetime
complexityofAVLinsertremainssameasBSTinsertwhichisO(h)wherehisheightofthetree.SinceAVL
treeisbalanced,theheightisO(Logn).SotimecomplexityofAVLinsertisO(Logn).
TheAVLtreeandotherselfbalancingsearchtreeslikeRedBlackareusefultogetallbasicoperationsdone
inO(Logn)time.TheAVLtreesaremorebalancedcomparedtoRedBlackTrees,buttheymaycausemore
rotationsduringinsertionanddeletion.Soifyourapplicationinvolvesmanyfrequentinsertionsand
deletions,thenRedBlacktreesshouldbepreferred.Andiftheinsertionsanddeletionsarelessfrequent
andsearchismorefrequentoperation,thenAVLtreeshouldbepreferredoverRedBlackTree.
Followingaresomepreviouspoststhathaveusedselfbalancingsearchtrees.
Medianinastreamofintegers(runningintegers)
Maximumofallsubarraysofsizek
Countsmallerelementsonrightside
=======================================================================
FindthelargestBSTsubtreeina
givenBinaryTree
GivenaBinaryTree,writeafunctionthatreturnsthesizeofthelargestsubtreewhichisalsoaBinary
SearchTree(BST).IfthecompleteBinaryTreeisBST,thenreturnthesizeofwholetree.
Examples:
Input:
5
/\
24
/\
13
Output:3
ThefollowingsubtreeisthemaximumsizeBSTsubtree
2
/\
13
Input:
50
/\
3060
/\/\
5204570
/\
6580
Output:5
ThefollowingsubtreeisthemaximumsizeBSTsubtree
60
/\
4570
/\
6580
Method1(Simplebutinefficient)
Startfromrootanddoaninordertraversalofthetree.ForeachnodeN,checkwhetherthesubtreerooted
withNisBSTornot.IfBST,thenreturnsizeofthesubtreerootedwithN.Else,recurdowntheleftandright
subtreesandreturnthemaximumofvaluesreturnedbyleftandrightsubtrees.
/*
See http://www.geeksforgeeks.org/archives/632 for implementation of size()
See Method 3 of http://www.geeksforgeeks.org/archives/3042 for

implementation of isBST()
max() returns maximum of two integers

*/
int largestBST(struct node *root)

{
if (isBST(root))
return size(root);
else
return max(largestBST(root->left), largestBST(root->right));
}
TimeComplexity:TheworstcasetimecomplexityofthismethodwillbeO(n^2).Consideraskewedtreefor
worstcaseanalysis.
Method2(TrickyandEfficient)
Inmethod1,wetraversethetreeintopdownmanneranddoBSTtestforeverynode.Ifwetraversethe
treeinbottomupmanner,thenwecanpassinformationaboutsubtreestotheparent.Thepassed
informationcanbeusedbytheparenttodoBSTtest(forparentnode)onlyinconstanttime(orO(1)time).
AleftsubtreeneedtotelltheparentwhetheritisBSTornotandalsoneedtopassmaximumvalueinit.So
thatwecancomparethemaximumvaluewiththeparentsdatatochecktheBSTproperty.Similarly,the
rightsubtreeneedtopasstheminimumvalueupthetree.Thesubtreesneedtopassthefollowing
informationupthetreeforthefindingthelargestBST.
1)WhetherthesubtreeitselfisBSTornot(Inthefollowingcode,is_bst_refisusedforthispurpose)
2)Ifthesubtreeisleftsubtreeofitsparent,thenmaximumvalueinit.Andifitisrightsubtreethenminimum
valueinit.
3)SizeofthissubtreeifthissubtreeisBST(Inthefollowingcode,returnvalueoflargestBSTtil()isusedfor
thispurpose)
max_refisusedforpassingthemaximumvalueupthetreeandmin_ptrisusedforpassingminimumvalue
upthetree.
#include <stdio.h>
#include <stdlib.h>
#include <limits.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
int largestBSTUtil(struct node* node, int *min_ref, int *max_ref,

int *max_size_ref, bool *is_bst_ref);
/* Returns size of the largest BST subtree in a Binary Tree

(efficient version). */
int largestBST(struct node* node)
{
// Set the initial values for calling largestBSTUtil()
int min = INT_MAX; // For minimum value in right subtree
int max = INT_MIN; // For maximum value in left subtree
int max_size = 0; // For size of the largest BST

bool is_bst = 0;
largestBSTUtil(node, &min, &max, &max_size, &is_bst);
return max_size;
}
/* largestBSTUtil() updates *max_size_ref for the size of the largest BST

subtree.
Also, if the tree rooted with node is non-empty and a BST,
then returns size of the tree. Otherwise returns 0.*/
int largestBSTUtil(struct node* node, int *min_ref, int *max_ref,

int *max_size_ref, bool *is_bst_ref)
{
/* Base Case */
if (node == NULL)
{
*is_bst_ref = 1; // An empty tree is BST
return 0;
// Size of the BST is 0
int min = INT_MAX;
/* A flag variable for left subtree property

i.e., max(root->left) < root->data */
bool left_flag = false;
/* A flag variable for right subtree property

i.e., min(root->right) > root->data */
bool right_flag = false;
int ls, rs; // To store sizes of left and right subtrees
/* Following tasks are done by recursive call for left subtree

a) Get the maximum value in left subtree (Stored in *max_ref)
b) Check whether Left Subtree is BST or not (Stored in *is_bst_ref)
c) Get the size of maximum size BST in left subtree (updates *max_size) */
*max_ref = INT_MIN;
ls = largestBSTUtil(node->left, min_ref, max_ref, max_size_ref, is_bst_ref);
if (*is_bst_ref == 1 && node->data > *max_ref)
left_flag = true;
/* Before updating *min_ref, store the min value in left subtree. So that we
have the correct minimum value for this subtree */
min = *min_ref;
/* The following recursive call does similar (similar to left subtree)
task for right subtree */

*min_ref = INT_MAX;
rs = largestBSTUtil(node->right, min_ref, max_ref, max_size_ref, is_bst_ref);
if (*is_bst_ref == 1 && node->data < *min_ref)
right_flag = true;
// Update min and max values for the parent recursive calls
if (min < *min_ref)
*min_ref = min;
if (node->data < *min_ref) // For leaf nodes
*min_ref = node->data;
if (node->data > *max_ref)
*max_ref = node->data;
/* If both left and right subtrees are BST. And left and right
subtree properties hold for this node, then this tree is BST.
So return the size of this tree */
if(left_flag && right_flag)
{
if (ls + rs + 1 > *max_size_ref)
*max_size_ref = ls + rs + 1;
return ls + rs + 1;
}
else
{
//Since this subtree is not BST, set is_bst flag for parent calls
*is_bst_ref = 0;
return 0;
}
}

int main()
{
/* Let us construct the following Tree
50
/
10
60
/ \
5
/ \
20
55
/
45 65
70
/ \
80
*/

root->left
= newNode(10);
root->right
= newNode(60);
/* The complete tree is not BST as 45 is in right subtree of 50.

The following subtree is the largest BST
60
/ \
5570
/
45
/ \
65
80
*/
printf(" Size of the largest BST is %d", largestBST(root));
getchar();
return 0;
}
TimeComplexity:O(n)wherenisthenumberofnodesinthegivenBinaryTree.
================================================================
=======
ConvertagiventreetoitsSumTree
GivenaBinaryTreewhereeachnodehaspositiveandnegativevalues.Convertthistoatreewhereeach
nodecontainsthesumoftheleftandrightsubtreesintheoriginaltree.Thevaluesofleafnodesare
changedto0.
Forexample,thefollowingtree
10
/\
26
/\/\
8475
shouldbechangedto
20(42+12+6)
/\
4(84)12(7+5)
/\/\
0000
Solution:
Doatraversalofthegiventree.Inthetraversal,storetheoldvalueofthecurrentnode,recursivelycallfor
leftandrightsubtreesandchangethevalueofcurrentnodeassumofthevaluesreturnedbytherecursive
calls.Finallyreturnthesumofnewvalueandvalue(whichissumofvaluesinthesubtreerootedwiththis
node).
#include<stdio.h>

struct node
{
int data;
struct node *left;
struct node *right;
};
// Convert a given tree to a tree where every node contains sum of values of
// nodes in left and right subtrees in the original tree
int toSumTree(struct node *node)
{
// Base case
if(node == NULL)
return 0;
// Store the old value

int old_val = node->data;
// Recursively call for left and right subtrees and store the sum as
// new value of this node
node->data = toSumTree(node->left) + toSumTree(node->right);
// Return the sum of values of nodes in left and right subtrees and
// old_value of this node
return node->data + old_val;
}

{
if (node == NULL)
return;
}

{
temp->data = data;
temp->left = NULL;
temp->right = NULL;
return temp;
}

int main()
{
int x;

root = newNode(10);
root->left = newNode(-2);
root->left->right = newNode(-4);
toSumTree(root);
// Print inorder traversal of the converted tree to test result of toSumTree()

printf("Inorder Traversal of the resultant tree is: \n");
printInorder(root);
getchar();
return 0;
}
Output:
InorderTraversaloftheresultanttreeis:
040200120
TimeComplexity:Thesolutioninvolvesasimpletraversalofthegiventree.SothetimecomplexityisO(n)
wherenisthenumberofnodesinthegivenBinaryTree.
=======================================================================
PopulateInorderSuccessorforall
nodes
GivenaBinaryTreewhereeachnodehasfollowingstructure,writeafunctiontopopulatenextpointerforall
nodes.Thenextpointerforeverynodeshouldbesettopointtoinordersuccessor.
struct node
{
int data;
struct node* left;
struct node* right;
struct node* next;
}
Initially,allnextpointershaveNULLvalues.Yourfunctionshouldfillthesenextpointerssothattheypointto
inordersuccessor.
Solution(UseReverseInorderTraversal)
Traversethegiventreeinreverseinordertraversalandkeeptrackofpreviouslyvisitednode.Whenanode
isbeingvisited,assignpreviouslyvisitednodeasnext.
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node *left;
struct node *right;
struct node *next;
};
/* Set next of p and all descendents of p by traversing them in reverse Inorder */

void populateNext(struct node* p)
{
// The first visited node will be the rightmost node
// next of the rightmost node will be NULL
static struct node *next = NULL;
if (p)
{
// First set the next pointer in right subtree
populateNext(p->right);
// Set the next as previously visited node in reverse Inorder

p->next = next;
// Change the prev for subsequent node

next = p;
// Finally, set the next pointer in left subtree

populateNext(p->left);
}
}
struct node* newnode(int data)
{
node->data = data;
node->left = NULL;
node->right = NULL;
node->next = NULL;
return(node);
}

int main()
{

10
8 12
/
3
*/
struct node *root = newnode(10);
root->left
= newnode(8);
root->right
= newnode(12);
// Populates nextRight pointer in all nodes

populateNext(root);
// Let us see the populated values

struct node *ptr = root->left->left;
while(ptr)
{
// -1 is printed if there is no successor
printf("Next of %d is %d \n", ptr->data, ptr->next? ptr->next->data: -1);
ptr = ptr->next;
}
return 0;
}
Wecanavoidtheuseofstaticvariablebypassingreferencetonextasparamater.
// An implementation that doesn't use static variable
// A wrapper over populateNextRecur

void populateNext(struct node *root)
{
// The first visited node will be the rightmost node
// next of the rightmost node will be NULL
struct node *next = NULL;
populateNextRecur(root, &next);
}
/* Set next of all descendents of p by traversing them in reverse Inorder */

void populateNextRecur(struct node* p, struct node **next_ref)
{
if (p)
{
// First set the next pointer in right subtree
populateNextRecur(p->right, next_ref);
// Set the next as previously visited node in reverse Inorder

p->next = *next_ref;
// Change the prev for subsequent node

*next_ref = p;
// Finally, set the next pointer in right subtree

populateNextRecur(p->left, next_ref);
}
}
TimeComplexity:O(n)
=======================================================================
SortedArraytoBalancedBST
Givenasortedarray.WriteafunctionthatcreatesaBalancedBinarySearchTreeusingarrayelements.
Examples:
Input:Array{1,2,3}
Output:ABalancedBST
2
/\
13
Input:Array{1,2,3,4}
Output:ABalancedBST
3
/\
24
/
1
Algorithm
Inthepreviouspost,wediscussedconstructionofBSTfromsortedLinkedList.Constructingfromsorted
arrayinO(n)timeissimpleraswecangetthemiddleelementinO(1)time.Followingisasimplealgorithm
wherewefirstfindthemiddlenodeoflistandmakeitrootofthetreetobeconstructed.
1)GettheMiddleofthearrayandmakeitroot.
2)Recursivelydosameforlefthalfandrighthalf.
a)Getthemiddleoflefthalfandmakeitleftchildoftheroot
createdinstep1.
b)Getthemiddleofrighthalfandmakeitrightchildofthe
rootcreatedinstep1.
FollowingisCimplementationoftheabovealgorithm.ThemaincodewhichcreatesBalancedBSTis
highlighted.
#include<stdio.h>
#include<stdlib.h>
/* A Binary Tree node */

struct TNode
{
int data;
struct TNode* left;
struct TNode* right;
};
struct TNode* newNode(int data);
/* A function that constructs Balanced Binary Search Tree from a sorted array */
struct TNode* sortedArrayToBST(int arr[], int start, int end)
{
/* Base Case */
if (start > end)
return NULL;

struct TNode *root = newNode(arr[mid]);


return root;
}

struct TNode* newNode(int data)
{
struct TNode* node = (struct TNode*)
malloc(sizeof(struct TNode));
node->data = data;
node->left = NULL;
node->right = NULL;
return node;
}
/* A utility function to print preorder traversal of BST */

void preOrder(struct TNode* node)
{
if (node == NULL)
return;
preOrder(node->left);
preOrder(node->right);
}

int main()
{
int arr[] = {1, 2, 3, 4, 5, 6, 7};
/* Convert List to BST */

struct TNode *root = sortedArrayToBST(arr, 0, n-1);
printf("\n PreOrder Traversal of constructed BST ");
preOrder(root);
return 0;
}
TimeComplexity:O(n)
FollowingistherecurrancerelationforsortedArrayToBST().
T(n)=2T(n/2)+C
T(n)>Timetakenforanarrayofsizen
C>Constant(Findingmiddleofarrayandlinkingroottoleft
andrightsubtreestakeconstanttime)
TheaboverecurrencecanbesolvedusingMasterTheoremasitfallsincase2.
================================================================
=======
Connectnodesatsamelevelusing
constantextraspace
Writeafunctiontoconnectalltheadjacentnodesatthesamelevelinabinarytree.Structureofthegiven
BinaryTreenodeislikefollowing.
struct node {
int data;
struct node* left;
struct node* right;
struct node* nextRight;

}
Initially,allthenextRightpointerspointtogarbagevalues.Yourfunctionshouldsetthesepointerstopoint
nextrightforeachnode.Youcanuseonlyconstantextraspace.
Example
InputTree
A
/\
BC
/\\
DEF
OutputTree
A>NULL
/\
B>C>NULL
/\\
D>E>F>NULL
Wediscussedtwodifferentapproachestodoitinthepreviouspost.Theauxiliaryspacerequiredinbothof
thoseapproachesisnotconstant.Also,themethod2discussedthereonlyworksforcompleteBinaryTree.
Inthispost,wewillfirstmodifythemethod2tomakeitworkforallkindoftrees.Afterthat,wewillremove
recursionfromthismethodsothattheextraspacebecomesconstant.
ARecursiveSolution
Inthemethod2ofpreviouspost,wetraversedthenodesinpreorderfashion.InsteadoftraversinginPre
Orderfashion(root,left,right),ifwetraversethenextRightnodebeforetheleftandrightchildren(root,
nextRight,left),thenwecanmakesurethatallnodesatlevelihavethenextRightset,beforetheleveli+1
nodes.Letusconsiderthefollowingexample(sameexampleaspreviouspost).Themethod2failsforright
childofnode4.Inthismethod,wemakesurethatallnodesatthe4slevel(level2)havenextRightset,
beforewetrytosetthenextRightof9.SowhenwesetthenextRightof9,wesearchforanonleafnodeon
rightsideofnode4(getNextRight()doesthisforus).
1Level0
/\
23Level1
/\/\
4567Level2
/\/\
891011Level3
void connectRecur(struct node* p);

struct node *getNextRight(struct node *p);
// Sets the nextRight of root and calls connectRecur() for other nodes
void connect (struct node *p)
{
// Set the nextRight for root
p->nextRight = NULL;
// Set the next right for rest of the nodes (other than root)
connectRecur(p);
}
/* Set next right of all descendents of p. This function makes sure that
nextRight of nodes ar level i is set before level i+1 nodes. */
void connectRecur(struct node* p)
{
// Base case
if (!p)
return;
/* Before setting nextRight of left and right children, set nextRight

of children of other nodes at same level (because we can access
children of other nodes using p's nextRight only) */
if (p->nextRight != NULL)
connectRecur(p->nextRight);
/* Set the nextRight pointer for p's left child */

if (p->left)
{
if (p->right)
{
p->left->nextRight = p->right;
p->right->nextRight = getNextRight(p);
}
else
p->left->nextRight = getNextRight(p);
/* Recursively call for next level nodes. Note that we call only
for left child. The call for left child will call for right child */
connectRecur(p->left);
}
/* If left child is NULL then first node of next level will either be
p->right or getNextRight(p) */
else if (p->right)
{
p->right->nextRight = getNextRight(p);
connectRecur(p->right);
}
else
connectRecur(getNextRight(p));
}
/* This function returns the leftmost child of nodes at the same level as p.
This function is used to getNExt right of p's right child
If right child of p is NULL then this can also be used for the left child */
struct node *getNextRight(struct node *p)
{
struct node *temp = p->nextRight;
/* Traverse nodes at p's level and find and return

the first node's first child */
while(temp != NULL)
{
if(temp->left != NULL)
return temp->left;
if(temp->right != NULL)
return temp->right;
temp = temp->nextRight;
}
// If all the nodes at p's level are leaf nodes then return NULL
return NULL;
}
AnIterativeSolution
Therecursiveapproachdiscussedabovecanbeeasilyconvertedtoiterative.Intheiterativeversion,weuse
nestedloop.Theouterloop,goesthroughallthelevelsandtheinnerloopgoesthroughallthenodesat
everylevel.Thissolutionusesconstantspace.
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node *left;
struct node *right;
struct node *nextRight;
};
/* This function returns the leftmost child of nodes at the same level as p.
This function is used to getNExt right of p's right child
If right child of is NULL then this can also be sued for the left child */
struct node *getNextRight(struct node *p)
{
struct node *temp = p->nextRight;
/* Traverse nodes at p's level and find and return

the first node's first child */
while (temp != NULL)
{
if (temp->left != NULL)
return temp->left;
if (temp->right != NULL)
return temp->right;
temp = temp->nextRight;
}
// If all the nodes at p's level are leaf nodes then return NULL
return NULL;
}
/* Sets nextRight of all nodes of a tree with root as p */

void connect(struct node* p)
{
struct node *temp;
if (!p)
return;
// Set nextRight for root

// set nextRight of all levels one by one

while (p != NULL)
{
struct node *q = p;
/* Connect all childrem nodes of p and children nodes of all other nodes
at same level as p */
while (q != NULL)
{
// Set the nextRight pointer for p's left child
if (q->left)
{
// If q has right child, then right child is nextRight of
// p and we also need to set nextRight of right child
if (q->right)
q->left->nextRight = q->right;
else
q->left->nextRight = getNextRight(q);
}
if (q->right)
q->right->nextRight = getNextRight(q);
// Set nextRight for other nodes in pre order fashion

q = q->nextRight;
}
// start from the first node of next level

if (p->left)
p = p->left;
else if (p->right)
p = p->right;
else
p = getNextRight(p);
}
}
{
node->data = data;
node->left = NULL;
node->right = NULL;
node->nextRight = NULL;
return(node);
}

int main()
{

10
8 2
/
90
*/
root->left
= newnode(8);
root->right
= newnode(2);
root->right->right
= newnode(90);

connect(root);
// Let us check the values of nextRight pointers

printf("Following are populated nextRight pointers in the tree "
"(-1 is printed if there is no nextRight) \n");
printf("nextRight of %d is %d \n", root->data,
root->nextRight? root->nextRight->data: -1);
printf("nextRight of %d is %d \n", root->left->data,
root->left->nextRight? root->left->nextRight->data: -1);
printf("nextRight of %d is %d \n", root->right->data,
root->right->nextRight? root->right->nextRight->data: -1);
printf("nextRight of %d is %d \n", root->left->left->data,
root->left->left->nextRight? root->left->left->nextRight->data: -1);
printf("nextRight of %d is %d \n", root->right->right->data,
root->right->right->nextRight? root->right->right->nextRight->data: -1);
getchar();
return 0;
}
Output:
FollowingarepopulatednextRightpointersinthetree(1isprintedif
thereisnonextRight)
nextRightof10is1
nextRightof8is2
nextRightof2is1
nextRightof3is90
nextRightof90is1
=======================================================================
Connectnodesatsamelevel
Writeafunctiontoconnectalltheadjacentnodesatthesamelevelinabinarytree.Structureofthegiven
BinaryTreenodeislikefollowing.
struct node{
int data;
struct node* left;
struct node* right;
struct node* nextRight;
}
Initially,allthenextRightpointerspointtogarbagevalues.Yourfunctionshouldsetthesepointerstopoint
nextrightforeachnode.
Example
InputTree
A
/\
BC
/\\
DEF
OutputTree
A>NULL
/\
B>C>NULL
/\\
D>E>F>NULL
Method1(ExtendLevelOrderTraversalorBFS)
Considerthemethod2ofLevelOrderTraversal.Themethod2caneasilybeextendedtoconnectnodesof
samelevel.Wecanaugmentqueueentriestocontainlevelofnodesalsowhichis0forroot,1forroots
childrenandsoon.Soaqueuenodewillnowcontainapointertoatreenodeandanintegerlevel.Whenwe
enqueueanode,wemakesurethatcorrectlevelvaluefornodeisbeingsetinqueue.TosetnextRight,for
everynodeN,wedequeuethenextnodefromqueue,ifthelevelnumberofnextnodeissame,wesetthe
nextRightofNasaddressofthedequeuednode,otherwisewesetnextRightofNasNULL.
TimeComplexity:O(n)
Method2(ExtendPreOrderTraversal)
ThisapproachworksonlyforCompleteBinaryTrees.InthismethodwesetnextRightinPreOrderfashion
tomakesurethatthenextRightofparentissetbeforeitschildren.Whenweareatnodep,wesetthe
nextRightofitsleftandrightchildren.Sincethetreeiscompletetree,nextRightofpsleftchild
(p>left>nextRight)willalwaysbepsrightchild,andnextRightofpsrightchild(p>right>nextRight)will
alwaysbeleftchildofpsnextRight(ifpisnottherightmostnodeatitslevel).Ifpistherightmostnode,then
nextRightofpsrightchildwillbeNULL.
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node *left;
struct node *right;
struct node *nextRight;
};
void connectRecur(struct node* p);
// Sets the nextRight of root and calls connectRecur() for other nodes
void connect (struct node *p)
{
// Set the nextRight for root
// Set the next right for rest of the nodes (other than root)
connectRecur(p);
}
/* Set next right of all descendents of p.

Assumption: p is a compete binary tree */
void connectRecur(struct node* p)
{
// Base case
if (!p)
return;
// Set the nextRight pointer for p's left child

if (p->left)
p->left->nextRight = p->right;
// Set the nextRight pointer for p's right child

// p->nextRight will be NULL if p is the right most child at its level
if (p->right)
p->right->nextRight = (p->nextRight)? p->nextRight->left: NULL;
// Set nextRight for other nodes in pre order fashion

connectRecur(p->left);
connectRecur(p->right);
}
{
node->data = data;
node->left = NULL;
node->right = NULL;
node->nextRight = NULL;
return(node);
}

int main()
{

10
/
8
\
2
/
3
*/
root->left
= newnode(8);
root->right
= newnode(2);

connect(root);
// Let us check the values of nextRight pointers

printf("Following are populated nextRight pointers in the tree "
"(-1 is printed if there is no nextRight) \n");
printf("nextRight of %d is %d \n", root->data,
root->nextRight? root->nextRight->data: -1);
printf("nextRight of %d is %d \n", root->left->data,
root->left->nextRight? root->left->nextRight->data: -1);
printf("nextRight of %d is %d \n", root->right->data,
root->right->nextRight? root->right->nextRight->data: -1);
printf("nextRight of %d is %d \n", root->left->left->data,
root->left->left->nextRight? root->left->left->nextRight->data: -1);
getchar();
return 0;
}
ThankstoDhanyaforsuggestingthisapproach.
TimeComplexity:O(n)
Whydoesntmethod2workfortreeswhicharenotCompleteBinaryTrees?
Letusconsiderfollowingtreeasanexample.InMethod2,wesetthenextRightpointerinpreorderfashion.
Whenweareatnode4,wesetthenextRightofitschildrenwhichare8and9(thenextRightof4isalready
setasnode5).nextRightof8willsimplybesetas9,butnextRightof9willbesetasNULLwhichis
incorrect.WecantsetthecorrectnextRight,becausewhenwesetnextRightof9,weonlyhavenextRight
ofnode4andancestorsofnode4,wedonthavenextRightofnodesinrightsubtreeofroot.
1
/\
23
/\/\
4567
/\/\
891011
=======================================================================
Trie|(Delete)
Inthepreviouspostontriewehavedescribedhowtoinsertandsearchanodeintrie.Hereisanalgorithm
howtodeleteanodefromtrie.
Duringdeleteoperationwedeletethekeyinbottomupmannerusingrecursion.Thefollowingarepossible
conditionswhendeletingkeyfromtrie,
1.
Keymaynotbethereintrie.Deleteoperationshouldnotmodifytrie.
2.
Keypresentasuniquekey(nopartofkeycontainsanotherkey(prefix),northekeyitselfisprefixof
anotherkeyintrie).Deleteallthenodes.
3.
Keyisprefixkeyofanotherlongkeyintrie.Unmarktheleafnode.
4.
Keypresentintrie,havingatleastoneotherkeyasprefixkey.Deletenodesfromendofkeyuntil
firstleafnodeoflongestprefixkey.
Thehighlightedcodepresentsalgorithmtoimplementaboveconditions.(Onemaybeindilemmahowa
pointerpassedtodeletehelperisreflectingchangesfromdeleteHelpertodeleteKey.Notethatweare
holdingtrieasanADTintrie_tnode,whichispassedbyreferenceorpointer).
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#define ARRAY_SIZE(a) sizeof(a)/sizeof(a[0])
// Alphabet size (# of symbols)
#define ALPHABET_SIZE (26)

#define INDEX(c) ((int)c - (int)'a')
#define FREE(p) \
free(p); \
p = NULL;
// forward declration
typedef struct trie_node trie_node_t;
// trie node
struct trie_node
{
int value; // non zero if leaf
trie_node_t *children[ALPHABET_SIZE];
};
// trie ADT
typedef struct trie trie_t;
struct trie
{
trie_node_t *root;
int count;
};
trie_node_t *getNode(void)
{
trie_node_t *pNode = NULL;
pNode = (trie_node_t *)malloc(sizeof(trie_node_t));
if( pNode )
{
int i;
pNode->value
= 0;
for(i = 0; i < ALPHABET_SIZE; i++)

{
pNode->children[i] = NULL;
}
}
return pNode;
}
void initialize(trie_t *pTrie)

{
pTrie->root = getNode();
pTrie->count = 0;
}
void insert(trie_t *pTrie, char key[])

{
int level;
int length = strlen(key);
int index;
trie_node_t *pCrawl;
pTrie->count++;
pCrawl = pTrie->root;
for( level = 0; level < length; level++ )

{
index = INDEX(key[level]);
if( pCrawl->children[index] )
{
// Skip current node
pCrawl = pCrawl->children[index];
}
else
{
// Add new node
pCrawl->children[index] = getNode();
}
}
// mark last node as leaf (non zero)

pCrawl->value = pTrie->count;
}
int search(trie_t *pTrie, char key[])

{
int level;
int index;

{
index = INDEX(key[level]);
if( !pCrawl->children[index] )
{
return 0;
}
}
return (0 != pCrawl && pCrawl->value);

}
int leafNode(trie_node_t *pNode)

{
return (pNode->value != 0);

}
int isItFreeNode(trie_node_t *pNode)

{
int i;
{
if( pNode->children[i] )
return 0;
}
return 1;
}
bool deleteHelper(trie_node_t *pNode, char key[], int level, int len)

{
if( pNode )
{
// Base case
if( level == len )
{
if( pNode->value )
{
// Unmark leaf node
pNode->value = 0;
// If empty, node to be deleted

if( isItFreeNode(pNode) )
{
return true;
}
return false;
}
}
else // Recursive case
{
int index = INDEX(key[level]);
if( deleteHelper(pNode->children[index], key, level+1, len) )

{
// last node marked, delete it
FREE(pNode->children[index]);
// recursively climb up, and delete eligible nodes

return ( !leafNode(pNode) && isItFreeNode(pNode) );
}
}
}
return false;
}
void deleteKey(trie_t *pTrie, char key[])

{
int len = strlen(key);
if( len > 0 )

{
deleteHelper(pTrie->root, key, 0, len);
}
}
int main()
{
char keys[][8] = {"she", "sells", "sea", "shore", "the", "by", "sheer"};
trie_t trie;
initialize(&trie);
for(int i = 0; i < ARRAY_SIZE(keys); i++)

{
insert(&trie, keys[i]);
deleteKey(&trie, keys[0]);
printf("%s %s\n", "she", search(&trie, "she") ? "Present in trie" : "Not present

in trie");
return 0;
}
=======================================================================
Trie|(InsertandSearch)
Trieisanefficientinformationretrievaldatastructure.Usingtrie,searchcomplexitiescanbebroughtto
optimallimit(keylength).Ifwestorekeysinbinarysearchtree,awellbalancedBSTwillneedtime
proportionaltoM*logN,whereMismaximumstringlengthandNisnumberofkeysintree.Usingtrie,we
cansearchthekeyinO(M)time.Howeverthepenaltyisontriestoragerequirements.
Everynodeoftrieconsistsofmultiplebranches.Eachbranchrepresentsapossiblecharacterofkeys.We
needtomarkthelastnodeofeverykeyasleafnode.Atrienodefieldvaluewillbeusedtodistinguishthe
nodeasleafnode(thereareotherusesofthevaluefield).AsimplestructuretorepresentnodesofEnglish
alphabetcanbeasfollowing,
structtrie_node
{
intvalue/*Usedtomarkleafnodes*/
trie_node_t*children[ALPHABET_SIZE]
}
Insertingakeyintotrieissimpleapproach.Everycharacterofinputkeyisinsertedasanindividualtrie
node.Notethatthechildrenisanarrayofpointerstonextleveltrienodes.Thekeycharacteractsasan
indexintothearraychildren.Iftheinputkeyisneworanextensionofexistingkey,weneedtoconstruct
nonexistingnodesofthekey,andmarkleafnode.Iftheinputkeyisprefixofexistingkeyintrie,wesimply
markthelastnodeofkeyasleaf.Thekeylengthdeterminestriedepth.
Searchingforakeyissimilartoinsertoperation,howeverweonlycomparethecharactersandmovedown.
Thesearchcanterminateduetoendofstringorlackofkeyintrie.Intheformercase,ifthevaluefieldof
lastnodeisnonzerothenthekeyexistsintrie.Inthesecondcase,thesearchterminateswithoutexamining
allthecharactersofkey,sincethekeyisnotpresentintrie.
Thefollowingpictureexplainsconstructionoftrieusingkeysgivenintheexamplebelow,
root
/\\
tab
|||
hny
||\|
esye
/||
irw
|||
ree
|
r
Inthepicture,everycharacterisoftypetrie_node_t.Forexample,therootisoftypetrie_node_t,andits
childrena,bandtarefilled,allothernodesofrootwillbeNULL.Similarly,aatthenextlevelishavingonly
onechild(n),allotherchildrenareNULL.Theleafnodesareinblue.
InsertandsearchcostsO(key_length),howeverthememoryrequirementsoftrieisO(ALPHABET_SIZE*
key_length*N)whereNisnumberofkeysintrie.Thereareefficientrepresentationoftrienodes(e.g.
compressedtrie,ternarysearchtree,etc.)tominimizememoryrequirementsoftrie.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#define ARRAY_SIZE(a) sizeof(a)/sizeof(a[0])
// Alphabet size (# of symbols)

#define ALPHABET_SIZE (26)
// Converts key current character into index
// use only 'a' through 'z' and lower case
#define CHAR_TO_INDEX(c) ((int)c - (int)'a')
// trie node
typedef struct trie_node trie_node_t;
struct trie_node
{
int value;
trie_node_t *children[ALPHABET_SIZE];
};
// trie ADT
typedef struct trie trie_t;
struct trie
{
trie_node_t *root;
int count;
};
// Returns new trie node (initialized to NULLs)

trie_node_t *getNode(void)
{
trie_node_t *pNode = NULL;
pNode = (trie_node_t *)malloc(sizeof(trie_node_t));
if( pNode )
{
int i;
pNode->value = 0;

{
pNode->children[i] = NULL;
}
}
return pNode;
}
// Initializes trie (root is dummy node)

void initialize(trie_t *pTrie)
{
pTrie->root = getNode();
pTrie->count = 0;
}
// If not present, inserts key into trie

// If the key is prefix of trie node, just marks leaf node
void insert(trie_t *pTrie, char key[])
{
int level;
int index;
pTrie->count++;

{
index = CHAR_TO_INDEX(key[level]);
{
pCrawl->children[index] = getNode();
}
}
// mark last node as leaf

pCrawl->value = pTrie->count;
}
// Returns non zero, if key presents in trie

int search(trie_t *pTrie, char key[])
{
int level;
int index;

{
index = CHAR_TO_INDEX(key[level]);
{
return 0;
}
}
return (0 != pCrawl && pCrawl->value);

}
// Driver
int main()
{
// Input keys (use only 'a' through 'z' and lower case)
char keys[][8] = {"the", "a", "there", "answer", "any", "by", "bye", "their"};
trie_t trie;
char output[][32] = {"Not present in trie", "Present in trie"};
initialize(&trie);
// Construct trie
for(int i = 0; i < ARRAY_SIZE(keys); i++)
{
insert(&trie, keys[i]);
}
// Search for different keys

printf("%s --- %s\n", "the", output[search(&trie, "the")] );
printf("%s --- %s\n", "these", output[search(&trie, "these")] );
printf("%s --- %s\n", "their", output[search(&trie, "their")] );
printf("%s --- %s\n", "thaw", output[search(&trie, "thaw")] );
return 0;
}
=======================================================================
Checkifabinarytreeissubtreeof
anotherbinarytree|Set1
Giventwobinarytrees,checkifthefirsttreeissubtreeofthesecondone.AsubtreeofatreeTisatreeS
consistingofanodeinTandallofitsdescendantsinT.Thesubtreecorrespondingtotherootnodeisthe
entiretreethesubtreecorrespondingtoanyothernodeiscalledapropersubtree.
Forexample,inthefollowingcase,treeSisasubtreeoftreeT.
TreeS
10
/\
46
\
30
TreeT
26
/\
103
/\\
463
\
30
Solution:TraversethetreeTinpreorderfashion.Foreveryvisitednodeinthetraversal,seeifthesubtree
rootedwiththisnodeisidenticaltoS.
FollowingisCimplementationforthis.
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node* left;
struct node* right;
};
/* A utility function to check whether trees with roots as root1 and

root2 are identical or not */
bool areIdentical(struct node * root1, struct node *root2)
{
/* base cases */
if (root1 == NULL && root2 == NULL)
return true;
if (root1 == NULL || root2 == NULL)

return false;
/* Check if the data of both roots is same and data of left and right
subtrees are also same */
return (root1->data == root2->data
&&
areIdentical(root1->left, root2->left) &&

areIdentical(root1->right, root2->right) );
}
/* This function returns true if S is a subtree of T, otherwise false */

bool isSubtree(struct node *T, struct node *S)
{
/* base cases */
if (S == NULL)
return true;
if (T == NULL)
return false;
/* Check the tree with root as current node */

if (areIdentical(T, S))
return true;
/* If the tree with root as current node doesn't match then

try left and right subtrees one by one */
return isSubtree(T->left, S) ||
isSubtree(T->right, S);
}
/* Helper function that allocates a new node with the given data
and NULL left and right pointers. */
{
struct node* node =
(struct node*)malloc(sizeof(struct node));
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Driver program to test above function */

int main()
{
/* Construct the following tree
26
/
10 3
/
\
30
*/
struct node *T
= newNode(26);
T->right
= newNode(3);
T->right->right
= newNode(3);
T->left
= newNode(10);
T->left->left
= newNode(4);
T->left->left->right = newNode(30);
T->left->right
= newNode(6);
/* Construct the following tree

10
/
4
\
6
\
30
*/
struct node *S = newNode(10);
S->right
= newNode(6);
S->left
= newNode(4);
S->left->right = newNode(30);
if (isSubtree(T, S))
printf("Tree S is subtree of tree T");
else
printf("Tree S is not a subtree of tree T");
getchar();
return 0;
}
Output:
TreeSissubtreeoftreeT
TimeComplexity:TimeworstcasecomplexityofabovesolutionisO(mn)wheremandnarenumberof
nodesingiventwotrees.
WecansolvetheaboveprobleminO(n)time.PleasereferCheckifabinarytreeissubtreeofanother
binarytree|Set2forO(n)solution.
=======================================================================
DecisionTreesFake(Counterfeit)
CoinPuzzle(12CoinPuzzle)
Letussolvetheclassicfakecoinpuzzleusingdecisiontrees.Therearethetwodifferentvariantsofthe
puzzlegivenbelow.Iamprovidingdescriptionofboththepuzzlesbelow,trytosolveonyourown,assume
N=8.
Easy:GivenatwopanfairbalanceandNidenticallylookingcoins,outofwhichonlyonecoinislighter(or
heavier).Tofigureouttheoddcoin,howmanyminimumnumberofweighingarerequiredintheworst
case?
Difficult:GivenatwopanfairbalanceandNidenticallylookingcoinsoutofwhichonlyonecoinmaybe
defective.Howcanwetracewhichcoin,ifany,isoddoneandalsodeterminewhetheritislighterorheavier
inminimumnumberoftrialsintheworstcase?
Letusstartwithrelativelysimpleexamples.Afterreadingeveryproblemtrytosolveonyourown.
Problem1:(Easy)
Given5coinsoutofwhichonecoinislighter.Intheworstcase,howmanyminimumnumberofweighing
arerequiredtofigureouttheoddcoin?
Namethecoinsas1,2,3,4and5.Weknowthatonecoinislighter.Consideringbestoutcomeofbalance,
wecangroupthecoinsintwodifferentways,[(1,2),(3,4)and(5)],or[(12),(34)and(5)].Wecaneasilyrule
outgroupslike[(123)and(45)],aswewillgetobviousanswer.Anyothercombinationwillfallintooneof
thesetwogroups,like[(2)(45)and(13)],etc.
Considerthefirstgroup,pairs(1,2)and(3,4).Wecancheck(1,2),iftheyareequalwegoaheadwith(3,
4).Weneedtwoweighinginworstcase.Thesameanalogycanbeappliedwhenthecoininheavier.
Withthesecondgroup,weigh(12)and(34).Iftheybalance(5)isdefectiveone,otherwisepickthelighter
pair,andweneedonemoreweighingtofindoddone.
Boththecombinationsneedtwoweighingincaseof5coinswithpriorinformationofonecoinislighter.
Analysis:Ingeneral,ifweknowthatthecoinisheavyorlight,wecantracethecoininlog3(N)trials
(roundedtonextinteger).Ifwerepresenttheoutcomeofbalanceasternarytree,everyleafrepresentan
outcome.SinceanycoinamongNcoinscanbedefective,weneedtogeta3arytreehavingminimumofN
leaves.A3arytreeatkthlevelwillhave3kleavesandhenceweneed3k>=N.
Inotherwords,inktrialswecanexamineupto3kcoins,ifweknowwhetherthedefectivecoinisheavieror
lighter.Giventhatacoinisheavier,verifythat3trialsaresufficienttofindtheoddcoinamong12coins,
because32<12<33.
Problem2:(Difficult)
Wearegiven4coins,outofwhichonlyonecoinmaybedefective.Wedontknow,whetherallcoinsare
genuineoranydefectiveoneispresent.Howmanynumberofweighingarerequiredinworstcasetofigure
outtheoddcoin,ifpresent?Wealsoneedtotellwhetheritisheavierorlighter.
Fromtheaboveanalysiswemaythinkthatk=2trialsaresufficient,sinceatwolevel3arytreeyields9
leaveswhichisgreaterthanN=4(readtheproblemonceagain).Notethatitisimpossibletosolveabove4
coinsproblemintwoweighing.Thedecisiontreeconfirmsthefact(trytodraw).
Wecangroupthecoinsintwodifferentways,[(12,34)]or[(1,2)and(3,4)].Letusconsiderthecombination
(12,34),thecorrespondingdecisiontreeisgivenbelow.Blueleavesarevalidoutcomes,andredleavesare
impossiblecases.Wearrivedatimpossiblecasesduetotheassumptionsmadeearlieronthepath.
Theoutcomecanbe(12)<(34)i.e.wegoontoleftsubtreeor(12)>(34)i.e.wegoontorightsubtree.
Theleftsubtreeispossibleintwoways,
A)Either1or2canbelighterOR
B)Either3or4canbeheavier.
Furtherontheleftsubtree,assecondtrial,weweigh(1,2)or(3,4).Letusconsider(3,4)astheanalogyfor
(1,2)issimilar.Theoutcomeofsecondtrailcanbethreeways
A)(3)<(4)yielding4asdefectiveheaviercoin,OR
B)(3)>(4)yielding3asdefectiveheaviercoinOR
C)(3)=(4),yieldingambiguity.Hereweneedonemoreweighingtocheckagenuinecoinagainst1
or2.InthefigureItook(3,2)where3isconfirmedasgenuine.Wecanget(3)>(2)inwhich2is
lighter,or(3)=(2)inwhich1islighter.Notethatitimpossibletoget(3)<(2),itcontradictsour
assumptionleanedtoleftside.
Similarlywecananalyzetherightsubtree.Weneedtwomoreweighingsonrightsubtreeaswell.
Overallweneed3weighingstotracetheoddcoin.Notethatweareunabletoutilizetwooutcomesof3ary
trees.Also,thetreeisnotfulltree,middlebranchterminatedafterfirstweighing.Infact,wecanget27
leavesof3levelfull3arytree,butonlywegot11leavesincludingimpossiblecases.
Analysis:GivenNcoins,allmaybegenuineoronlyonecoinisdefective.Weneedadecisiontreewith
atleast(2N+1)leavescorrespondtotheoutputs.BecausetherecanbeNleavestobelighter,orNleaves
tobeheavieroronegenuinecase,ontotal(2N+1)leaves.
Asexplainedearlierternarytreeatlevelk,canhaveutmost3kleavesandweneedatreewithleavesof3k>
(2N+1).
Inotherwords,weneedatleastk>log3(2N+1)weighingtofindthedefectiveone.
Observetheabovefigurethatnotallthebranchesaregeneratingleaves,i.e.wearemissingvalidoutputs
undersomebranchesthatleadingtomorenumberoftrials.Whenpossible,weshouldgroupthecoinsin
suchawaythateverybranchisgoingtoyieldvalidoutput(insimpletermsgeneratefull3arytree).Problem
4describesthisapproachof12coins.
Problem3:(Specialcaseoftwopanbalance)
Wearegiven5coins,agroupof4coinsoutofwhichonecoinisdefective(wedontknowwhetheritis
heavierorlighter),andonecoinisgenuine.Howmanyweighingarerequiredinworstcasetofigureoutthe
oddcoinwhetheritisheavierorlighter?
Labelthecoinsas1,2,3,4andG(genuine).Wenowhavesomeinformationoncoinpurity.Weneedto
makeusethatinthegroupings.
Wecanbestgroupthemas[(G1,23)and(4)].Anyothergroupcantgeneratefull3arytree,tryyourself.
Thefollowingdiagramexplainstheprocedure.

Themiddlecase(G1)=(23)isselfexplanatory,i.e.1,2,3aregenuineand4thcoincanbefiguredout
lighterorheavierinonemoretrial.
Theleftsideoftreecorrespondstothecase(G1)<(23).Thisispossibleintwoways,either1shouldbe
lighteroreitherof(2,3)shouldbeheavier.Theformerinstanceisobviouswhennextweighing(2,3)is
balanced,yielding1aslighter.Thelaterinstancecouldbe(2)<(3)yielding3asheavieror(2)>(3)yielding
2asheavier.Theleafnodesonleftbrancharenamedtoreflecttheseoutcomes.
Therightsideoftreecorrespondstothecase(G1)>(23).Thisispossibleintwoways,either1isheavieror
eitherof(2,3)shouldbelighter.Theformerinstanceisobviouswhenthenextweighing(2,3)isbalanced,
yielding1asheavier.Thelatercasecouldbe(2)<(3)yielding2aslightercoin,or(2)>(3)yielding3as
lighter.
Intheaboveproblem,underanypossibilityweneedonlytwoweighing.Weareabletousealloutcomesof
twolevelfull3arytree.Westartedwith(N+1)=5coinswhereN=4,weendupwith(2N+1)=9leaves.
Infactweshouldhave11outcomessincewestaredwith5coins,whereareother2outcomes?
Thesetwooutcomescanbedeclaredattherootoftreeitself(priortofirstweighing),canyoufigure
thesetwooutcomes?
Ifweobservethefigure,afterthefirstweighingtheproblemreducedtoweknowthreecoins,eitheronecan
belighter(heavier)oroneamongothertwocanbeheavier(lighter).Thiscanbesolvedinoneweighing
(readProblem1).
Analysis:Given(N+1)coins,oneisgenuineandtherestNcanbegenuineoronlyonecoinisdefective.
Therequireddecisiontreeshouldresultinminimumof(2N+1)leaves.Sincethetotalpossibleoutcomes
are(2(N+1)+1),numberofweighing(trials)aregivenbytheheightofternarytree,k>=log3[2(N+1)+1].
Notetheequalitysign.
RearrangingkandN,wecanweighmaximumofN<=(3k3)/2coinsinktrials.
Problem4:(Theclassic12coinpuzzle)
Youaregiventwopanfairbalance.Youhave12identicallylookingcoinsoutofwhichonecoinmaybe
lighterorheavier.Howcanyoufindoddcoin,ifany,inminimumtrials,alsodeterminewhetherdefective
coinislighterorheavier,intheworstcase?
Howdoyouwanttogroupthem?Bisetortriset?Clearlywecandiscardtheoptionofdividingintotwo
equalgroups.Itcantleadtobesttree.Fromtheabovetwoexamples,wecanensurethatthedecisiontree
canbeusedinoptimalwayifwecanrevealatleasetonegenuinecoin.Remembertogroupcoinssuchthat
thefirstweighingrevealsatleastonegenuinecoin.
Letusnamethecoinsas1,2,8,A,B,CandD.Wecancombinethecoinsinto3groups,namely(1234),
(5678)and(ABCD).Weigh(1234)and(5678).Youareencouragedtodrawdecisiontreewhilereadingthe
procedure.Theoutcomecanbethreeways,
1.
(1234)=(5678),bothgroupsareequal.Defectivecoinmaybein(ABCD)group.
2.
(1234)<(5678),i.e.firstgroupislessinweightthansecondgroup.
3.
(1234)>(5678),i.e.firstgroupismoreinweightthansecondgroup.
Theoutput(1)canbesolvedintwomoreweighingasspecialcaseoftwopanbalancegiveninProblem3.
Weknowthatgroups(1234)and(5678)aregenuineanddefectivecoinmaybein(ABCD).Pickone
genuinecoinfromanyofweighedgroups,andproceedwith(ABCD)asexplainedinProblem3.
Outcomes(2)and(3)arespecial.Inboththecases,weknowthat(ABCD)isgenuine.Andalso,weknowa
setofcoinsbeinglighterandasetofcoinsbeingheavier.Weneedtoshuffletheweighedtwogroupsin
suchawaythatweendupwithsmallerheightdecisiontree.
Considerthesecondoutcomewhere(1234)<(5678).Itispossiblewhenanycoinamong(1,2,3,4)is
lighteroranycoinamong(5,6,7,8)isheavier.Werevealedlighterorheavierpossibilityafterfirst
weighing.IfweproceedasinProblem1,wewillnotgeneratebestdecisiontree.Letusshufflecoinsas
(1235)and(4BCD)asnewgroups(therearedifferentshufflespossible,theyalsoleadtominimum
weighing,canyoutry?).Ifweweighthesetwogroupsagaintheoutcomecanbethreeways,i)(1235)<
(4BCD)yieldingoneamong1,2,3islighterwhichissimilartoProblem1explainedabove,weneedone
moreweighing,ii)(1235)=(4BCD)yieldingoneamong6,7,8isheavierwhichissimilartoProblem1
explainedabove,weneedonemoreweighingiii)(1235)>(4BCD)yieldingeither5asheaviercoinor4as
lightercoin,attheexpenseofonemoreweighing.
Similarwaywecanalsosolvetherightsubtree(thirdoutcomewhere(1234)>(5678))intwomoreweighing.
Weareabletosolvethe12coinpuzzlein3weighingintheworstcase.
FewInterestingPuzzles:
1.
SolveProblem4withN=8andN=13,Howmanyminimumtrialsarerequiredineachcase?
2.
Givenafunctionintweigh(A[],B[])whereAandBarearrays(neednotbeequalsize).Thefunction
returns1,0or1.Itreturns0ifsumofallelementsinAandBareequal,1ifA<Band1ifA>B.
Givenanarrayof12elements,allelementsareequalexceptone.Theoddelementcanbeasthat
ofothers,smallerorgreaterthanothers.Writeaprogramtofindtheoddelement(ifany)using
weigh()minimumnumberoftimes.
3.
Youmighthaveseen3panbalanceinsciencelabsduringschooldays.Givena3panbalance(4
outcomes)andNcoins,howmanyminimumtrialsareneededtofigureoutoddcoin?
=======================================================================
CheckifagivenBinaryTreeis
SumTree
WriteafunctionthatreturnstrueifthegivenBinaryTreeisSumTreeelsefalse.ASumTreeisaBinaryTree
wherethevalueofanodeisequaltosumofthenodespresentinitsleftsubtreeandrightsubtree.An
emptytreeisSumTreeandsumofanemptytreecanbeconsideredas0.Aleafnodeisalsoconsideredas
SumTree.
FollowingisanexampleofSumTree.
26
/\
103
/\\
463
Method1(Simple)
Getthesumofnodesinleftsubtreeandrightsubtree.Checkifthesumcalculatedisequaltorootsdata.
Also,recursivelycheckiftheleftandrightsubtreesareSumTrees.
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node* left;
struct node* right;
};
/* A utility function to get the sum of values in tree with root

as root */
int sum(struct node *root)
{
if(root == NULL)
return 0;
return sum(root->left) + root->data + sum(root->right);
}
/* returns 1 if sum property holds for the given

node and both of its children */
int isSumTree(struct node* node)
{
int ls, rs;
/* If node is NULL or it's a leaf node then

return true */
if(node == NULL ||
(node->left == NULL && node->right == NULL))
return 1;
/* Get sum of nodes in left and right subtrees */

ls = sum(node->left);
rs = sum(node->right);
/* if the node and both of its children satisfy the

property return 1 else 0*/
if((node->data == ls + rs)&&
isSumTree(node->left) &&
isSumTree(node->right))
return 1;
return 0;
}
/*
Helper function that allocates a new node
with the given data and NULL left and right
pointers.
*/
{
struct node* node =
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(10);
root->right
= newNode(3);
root->left->left
= newNode(4);
if(isSumTree(root))
printf("The given tree is a SumTree ");
else
printf("The given tree is not a SumTree ");
getchar();
return 0;
}
TimeComplexity:O(n^2)inworstcase.Worstcaseoccursforaskewedtree.
Method2(Tricky)
TheMethod1usessum()togetthesumofnodesinleftandrightsubtrees.Themethod2usesfollowing
rulestogetthesumdirectly.
1)Ifthenodeisaleafnodethensumofsubtreerootedwiththisnodeisequaltovalueofthisnode.
2)Ifthenodeisnotaleafnodethensumofsubtreerootedwiththisnodeistwicethevalueofthisnode
(AssumingthatthetreerootedwiththisnodeisSumTree).
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node* left;
struct node* right;
};
/* Utillity function to check if the given node is leaf or not */

int isLeaf(struct node *node)
{
if(node == NULL)
return 0;
if(node->left == NULL && node->right == NULL)
return 1;
return 0;
}
/* returns 1 if SumTree property holds for the given

tree */
{
int ls; // for sum of nodes in left subtree
int rs; // for sum of nodes in right subtree

return true */
if(node == NULL || isLeaf(node))
return 1;
if( isSumTree(node->left) && isSumTree(node->right))

{
// Get the sum of nodes in left subtree
if(node->left == NULL)
ls = 0;
else if(isLeaf(node->left))
ls = node->left->data;
else
ls = 2*(node->left->data);
// Get the sum of nodes in right subtree

if(node->right == NULL)
rs = 0;
else if(isLeaf(node->right))
rs = node->right->data;
else
rs = 2*(node->right->data);
/* If root's data is equal to sum of nodes in left

and right subtrees then return 1 else return 0*/
return(node->data == ls + rs);
}
return 0;
}
/* Helper function that allocates a new node

pointers.
*/
{
struct node* node =
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(10);
root->right
= newNode(3);
root->left->left
= newNode(4);
if(isSumTree(root))
else
getchar();
return 0;
}
TimeComplexity:O(n)
WriteafunctionthatreturnstrueifthegivenBinaryTreeisSumTreeelsefalse.ASumTreeisaBinaryTree
wherethevalueofanodeisequaltosumofthenodespresentinitsleftsubtreeandrightsubtree.An
emptytreeisSumTreeandsumofanemptytreecanbeconsideredas0.Aleafnodeisalsoconsideredas
SumTree.
FollowingisanexampleofSumTree.
26
/\
103
/\\
463
Method1(Simple)
Getthesumofnodesinleftsubtreeandrightsubtree.Checkifthesumcalculatedisequaltorootsdata.
Also,recursivelycheckiftheleftandrightsubtreesareSumTrees.
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node* left;
struct node* right;
};
/* A utility function to get the sum of values in tree with root

as root */
int sum(struct node *root)
{
if(root == NULL)
return 0;
return sum(root->left) + root->data + sum(root->right);
}
/* returns 1 if sum property holds for the given

node and both of its children */
{
int ls, rs;

return true */
if(node == NULL ||
return 1;
/* Get sum of nodes in left and right subtrees */

ls = sum(node->left);
rs = sum(node->right);

if((node->data == ls + rs)&&
isSumTree(node->left) &&
isSumTree(node->right))
return 1;
return 0;
}
/*

pointers.
*/
{
struct node* node =
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(10);
root->right
= newNode(3);
root->left->left
= newNode(4);
if(isSumTree(root))
else
getchar();
return 0;
}
TimeComplexity:O(n^2)inworstcase.Worstcaseoccursforaskewedtree.
Method2(Tricky)
TheMethod1usessum()togetthesumofnodesinleftandrightsubtrees.Themethod2usesfollowing
rulestogetthesumdirectly.
1)Ifthenodeisaleafnodethensumofsubtreerootedwiththisnodeisequaltovalueofthisnode.
2)Ifthenodeisnotaleafnodethensumofsubtreerootedwiththisnodeistwicethevalueofthisnode
(AssumingthatthetreerootedwiththisnodeisSumTree).
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node* left;
struct node* right;
};
/* Utillity function to check if the given node is leaf or not */

int isLeaf(struct node *node)
{
if(node == NULL)
return 0;
if(node->left == NULL && node->right == NULL)
return 1;
return 0;
}
/* returns 1 if SumTree property holds for the given

tree */
{
int ls; // for sum of nodes in left subtree
int rs; // for sum of nodes in right subtree

return true */
if(node == NULL || isLeaf(node))
return 1;
if( isSumTree(node->left) && isSumTree(node->right))

{
// Get the sum of nodes in left subtree
if(node->left == NULL)
ls = 0;
else if(isLeaf(node->left))
ls = node->left->data;
else
ls = 2*(node->left->data);
// Get the sum of nodes in right subtree

if(node->right == NULL)
rs = 0;
else if(isLeaf(node->right))
rs = node->right->data;
else
rs = 2*(node->right->data);
/* If root's data is equal to sum of nodes in left

and right subtrees then return 1 else return 0*/
return(node->data == ls + rs);
}
return 0;
}

pointers.
*/
{
struct node* node =
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
int main()
{
root->left
= newNode(10);
root->right
= newNode(3);
root->left->left
= newNode(4);
if(isSumTree(root))
else
getchar();
return 0;
}
TimeComplexity:O(n)
================================================================
=======
TournamentTree(WinnerTree)and
BinaryHeap
GivenateamofNplayers.Howmanyminimumgamesarerequiredtofindsecondbestplayer?
Wecanuseadversaryargumentsbasedontournamenttree(BinaryHeap).
Tournamenttreeisaformofmin(max)heapwhichisacompletebinarytree.Everyexternalnode
representsaplayerandinternalnoderepresentswinner.Inatournamenttreeeveryinternalnodecontains
winnerandeveryleafnodecontainsoneplayer.
TherewillbeN1internalnodesinabinarytreewithNleaf(external)nodes.Fordetailsseethispost(put
n=2inequationgiveninthepost).
ItisobviousthattoselectthebestplayeramongNplayers,(N1)playerstobeeliminated,i.e.weneed
minimumof(N1)games(comparisons).Mathematicallywecanproveit.InabinarytreeI=E1,whereI
isnumberofinternalnodesandEisnumberofexternalnodes.Itmeanstofindmaximumorminimum
elementofanarray,weneedN1(internalnodes)comparisons.
SecondBestPlayer
Theinformationexploredduringbestplayerselectioncanbeusedtominimizethenumberofcomparisons
intracingthenextbestplayers.Forexample,wecanpicksecondbestplayerin(N+log2N2)
comparisons.Fordetailsreadthiscomment.
Thefollowingdiagramdisplaysatournamenttree(winnertree)asamaxheap.Notethattheconceptof
losertreeisdifferent.
Theabovetreecontains4leafnodesthatrepresentplayersandhave3levels0,1and2.Initially2games
areconductedatlevel2,onebetween5and3andanotheronebetween7and8.Inthenextmove,one
moregameisconductedbetween5and8toconcludethefinalwinner.Overallweneed3comparisons.For
secondbestplayerweneedtotracethecandidatesparticipatedwithfinalwinner,thatleadsto7assecond
best.
MedianofSortedArrays
Tournamenttreecaneffectivelybeusedtofindmedianofsortedarrays.Assume,givenMsortedarraysof
equalsizeL(forsimplicity).Wecanattachallthesesortedarraystothetournamenttree,onearrayperleaf.
WeneedatreeofheightCEIL(log2M)tohaveatleastMexternalnodes.
Consideranexample.Given3(M=3)sortedintegerarraysofmaximumsize5elements.
{2,5,7,11,15}Array1
{1,3,4}Array2
{6,8,12,13,14}Array3
Whatshouldbetheheightoftournamenttree?Weneedtoconstructatournamenttreeofheightlog23.=
1.585=2roundedtonextinteger.Abinarytreeofheight2willhave4leavestowhichwecanattachthe
arraysasshowninthebelowfigure.

Afterthefirsttournament,thetreeappearsasbelow,
WecanobservethatthewinnerisfromArray2.HencethenextelementfromArray2willdiveinandgames
willbeplayedalongthewinnerpathofprevioustournament.
Notethatinfinityisusedassentinelelement.Basedondatabeingholdinnodeswecanselectthesentinel
character.Forexampleweusuallystorethepointersinnodesratherthankeys,soNULLcanserveas
sentinel.Ifanyofthearrayexhaustswewillfillthecorrespondingleafandupcominginternalnodeswith
sentinel.
Afterthesecondtournament,thetreeappearsasbelow,
ThenextwinnerisfromArray1,sonextelementofArray1arraywhichis5willdiveintothenextround,and
nexttournamentplayedalongthepathof2.
Thetournamentscanbecontinuedtillwegetmedianelementwhichis(5+3+5)/2=7thelement.Notethat
thereareevenbetteralgorithmsforfindingmedianofunionofsortedarrays,fordetailsseetherelatedlinks
givenbelow.
IngeneralwithMsortedlistsofsizeL1,L2LmrequirestimecomplexityofO((L1+L2++Lm)*logM)to
mergeallthearrays,andO(m*logM)timetofindmedian,wheremismedianposition.
Selectsmallestonemillionelementsfromonebillionunsortedelements:
Asasimplesolution,wecansortthebillionnumbersandselectfirstonemillion.
Onalimitedmemorysystemsortingbillionelementsandpickingthefirstonemillionseemstobe
impractical.Wecanusetournamenttreeapproach.Atanytimeonlyelementsoftreetobeinmemory.
Splitthelargearray(perhapsstoredondisk)intosmallersizearraysofsizeonemillioneach(oreven
smallerthatcanbesortedbythemachine).Sortthese1000smallsizearraysandstorethemondiskas
individualfiles.Constructatournamenttreewhichcanhaveatleast1000leafnodes(treetobeofheight10
since29<1000<210,iftheindividualfilesizeisevensmallerwewillneedmoreleafnodes).Everyleafnode
willhaveanenginethatpicksnextelementfromthesortedfilestoredondisk.Wecanplaythetournament
treegametoextractfirstonemillionelements.
Totalcost=sorting1000listsofonemillioneach+treeconstruction+tournaments
Implementation
Weneedtobuildthetree(heap)inbottomupmanner.Alltheleafnodesfilledfirst.Startattheleftextreme
oftreeandfillalongthebreadth(i.e.from2k1to2k1wherekisdepthoftree)andplaythegame.After
practicingwithfewexamplesitwillbeeasytowritecode.Wewillhavecodeinanupcomingarticle.
================================================================
=======
PrintBSTkeysinthegivenrange
Giventwovaluesk1andk2(wherek1<k2)andarootpointertoaBinarySearchTree.Printallthekeysof
treeinrangek1tok2.i.e.printallxsuchthatk1<=x<=k2andxisakeyofgivenBST.Printallthekeysin
increasingorder.
Forexample,ifk1=10andk2=22,thenyourfunctionshouldprint12,20and22.
Thankstobhaskerforsuggestingthefollowingsolution.
Algorithm:
1)Ifvalueofrootskeyisgreaterthank1,thenrecursivelycallinleftsubtree.
2)Ifvalueofrootskeyisinrange,thenprinttherootskey.
3)Ifvalueofrootskeyissmallerthank2,thenrecursivelycallinrightsubtree.
Implementation:
#include<stdio.h>

struct node
{
int data;
struct node *left;
struct node *right;

};
/* The functions prints all the keys which in the given range [k1..k2].
The function assumes than k1 < k2 */
void Print(struct node *root, int k1, int k2)
{
/* base case */
if ( NULL == root )
return;
/* Since the desired o/p is sorted, recurse for left subtree first
If root->data is greater than k1, then only we can get o/p keys
in left subtree */
if ( k1 < root->data )
Print(root->left, k1, k2);
/* if root's data lies in range, then prints root's data */

if ( k1 <= root->data && k2 >= root->data )
printf("%d ", root->data );
/* If root->data is smaller than k2, then only we can get o/p keys
in right subtree */
if ( k2 > root->data )
Print(root->right, k1, k2);
}

{
temp->data = data;
temp->left = NULL;
temp->right = NULL;
return temp;
}

int main()
{
struct node *root = new struct node;
int k1 = 10, k2 = 25;

root = newNode(20);
Print(root, k1, k2);
getchar();
return 0;
}
Output:
122022
TimeComplexity:O(n)wherenisthetotalnumberofkeysintree.
=======================================================================
PrintAncestorsofagivennodein
BinaryTree
GivenaBinaryTreeandakey,writeafunctionthatprintsalltheancestorsofthekeyinthegivenbinary
tree.
Forexample,ifthegiventreeisfollowingBinaryTreeandkeyis7,thenyourfunctionshouldprint4,2and
1.
1
/\
23
/\
45
/
7
ThankstoMike,Sambasivaandwgpshashankfortheircontribution.
#include<iostream>
#include<stdio.h>
#include<stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
/* If target is present in tree, then prints the ancestors

and returns true, otherwise returns false. */
bool printAncestors(struct node *root, int target)
{
/* base cases */
if (root == NULL)
return false;
if (root->data == target)
return true;
/* If target is present in either left or right subtree of this node,

then print this node */
if ( printAncestors(root->left, target) ||
printAncestors(root->right, target) )
{
return true;
}
/* Else return false */

return false;
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
/* Construct the following binary tree

1
/
2 3
/ \
4
/
7
*/
root->left
= newnode(2);
root->right
= newnode(3);
root->left->left->left = newnode(7);
printAncestors(root, 7);
getchar();
return 0;
}
Output:
421
=======================================================================
GetLevelofanodeinaBinaryTree
GivenaBinaryTreeandakey,writeafunctionthatreturnslevelofthekey.
Forexample,considerthefollowingtree.Iftheinputkeyis3,thenyourfunctionshouldreturn1.Iftheinput
keyis4,thenyourfunctionshouldreturn3.Andforkeywhichisnotpresentinkey,thenyourfunction
shouldreturn0.
Thankstoprandeeyforsuggestingthefollowingsolution.
Theideaistostartfromtherootandlevelas1.Ifthekeymatcheswithrootsdata,returnlevel.Else
recursivelycallforleftandrightsubtreeswithlevelaslevel+1.
#include<stdio.h>

struct node
{
int data;
struct node *left;
struct node *right;
};
/* Helper function for getLevel(). It returns level of the data if data is

present in tree, otherwise returns 0.*/
int getLevelUtil(struct node *node, int data, int level)
{
if (node == NULL)
return 0;
if (node->data == data)
return level;
int downlevel = getLevelUtil(node->left, data, level+1);

if (downlevel != 0)
return downlevel;
downlevel = getLevelUtil(node->right, data, level+1);

return downlevel;
}
/* Returns level of given data value */

int getLevel(struct node *node, int data)
{
return getLevelUtil(node,data,1);
}

{
temp->data = data;
temp->left = NULL;
temp->right = NULL;
return temp;
}

int main()
{
struct node *root = new struct node;
int x;

root = newNode(3);
for (x = 1; x <=5; x++)

{
int level = getLevel(root, x);
if (level)
printf(" Level of %d is %d\n", x, getLevel(root, x));
else
printf(" %d is not present in tree \n", x);
getchar();
return 0;
}
Output:
Levelof1is3
Levelof2is2
Levelof3is1
Levelof4is3
Levelof5is2
================================================================
=======
FindkthsmallestelementinBST
(OrderStatisticsinBST)
GivenrootofbinarysearchtreeandKasinput,findKthsmallestelementinBST.
Forexample,inthefollowingBST,ifk=3,thenoutputshouldbe10,andifk=5,thenoutputshouldbe14.
Method1:UsingInorderTraversal.
InordertraversalofBSTretrieveselementsoftreeinthesortedorder.Theinordertraversalusesstackto
storetobeexplorednodesoftree(threadedtreeavoidsstackandrecursionfortraversal,seethispost).The
ideaistokeeptrackofpoppedelementswhichparticipateintheorderstatics.Hypotheticalalgorithmis
providedbelow,
Timecomplexity:O(n)wherenistotalnodesintree..
Algorithm:
/*initialization*/
pCrawl=root
setinitialstackelementasNULL(sentinal)
/*traverseuptoleftextreme*/
while(pCrawlisvalid)
stack.push(pCrawl)
pCrawl=pCrawl.left
/*processothernodes*/
while(pCrawl=stack.pop()isvalid)
stopifsufficientnumberofelementsarepopped.
if(pCrawl.rightisvalid)
pCrawl=pCrawl.right
while(pCrawlisvalid)
stack.push(pCrawl)
pCrawl=pCrawl.left
Implementation:
#include <stdio.h>
#include <stdlib.h>
#define ARRAY_SIZE(arr) sizeof(arr)/sizeof(arr[0])
/* just add elements to test */

/* NOTE: A sorted array results in skewed tree */
int ele[] = { 20, 8, 22, 4, 12, 10, 14 };
/* same alias */
typedef struct node_t node_t;
/* Binary tree node */

struct node_t
{
int data;
node_t* left;
node_t* right;
};
/* simple stack that stores node addresses */

typedef struct stack_t stack_t;
/* initial element always NULL, uses as sentinal */
struct stack_t
{
node_t* base[ARRAY_SIZE(ele) + 1];
int
stackIndex;
};
/* pop operation of stack */

node_t *pop(stack_t *st)
{
node_t *ret = NULL;
if( st && st->stackIndex > 0 )

{
ret = st->base[st->stackIndex];
st->stackIndex--;
}
return ret;
}
/* push operation of stack */

void push(stack_t *st, node_t *node)
{
if( st )
{
st->stackIndex++;
st->base[st->stackIndex] = node;
}
}
/* Iterative insertion
Recursion is least preferred unless we gain something
*/
node_t *insert_node(node_t *root, node_t* node)
{
/* A crawling pointer */
node_t *pTraverse = root;
node_t *currentParent = root;
// Traverse till appropriate node

while(pTraverse)
{
currentParent = pTraverse;
if( node->data < pTraverse->data )

{
/* left subtree */
pTraverse = pTraverse->left;
}
else
{
/* right subtree */
pTraverse = pTraverse->right;
}
}
/* If the tree is empty, make it as root node */

if( !root )
{
root = node;
}
else if( node->data < currentParent->data )
{
/* Insert on left side */
currentParent->left = node;
}
else
{
/* Insert on right side */
currentParent->right = node;
}
return root;
}
/* Elements are in an array. The function builds binary tree */

node_t* binary_search_tree(node_t *root, int keys[], int const size)
{
int iterator;
node_t *new_node = NULL;
for(iterator = 0; iterator < size; iterator++)

{
new_node = (node_t *)malloc( sizeof(node_t) );
/* initialize */
new_node->data
= keys[iterator];
new_node->left
= NULL;
new_node->right = NULL;
/* insert into BST */

root = insert_node(root, new_node);
}
return root;
}
node_t *k_smallest_element_inorder(stack_t *stack, node_t *root, int k)

{
stack_t *st = stack;
node_t *pCrawl = root;
/* move to left extremen (minimum) */

while( pCrawl )
{
push(st, pCrawl);
pCrawl = pCrawl->left;
}
/* pop off stack and process each node */

while( pCrawl = pop(st) )
{
/* each pop operation emits one element
in the order
*/
if( !--k )
{
/* loop testing */
st->stackIndex = 0;
break;
}
/* there is right subtree */

if( pCrawl->right )
{
/* push the left subtree of right subtree */
pCrawl = pCrawl->right;
while( pCrawl )
{
push(st, pCrawl);
pCrawl = pCrawl->left;
}
/* pop off stack and repeat */

}
}
/* node having k-th element or NULL node */

return pCrawl;
}

int main(void)
{
node_t* root = NULL;
stack_t stack = { {0}, 0 };
node_t *kNode = NULL;
int k = 5;
/* Creating the tree given in the above diagram */

root = binary_search_tree(root, ele, ARRAY_SIZE(ele));
kNode = k_smallest_element_inorder(&stack, root, k);
if( kNode )
{
printf("kth smallest elment for k = %d is %d", k, kNode->data);
}
else
{
printf("There is no such element");
}
getchar();
return 0;
}
Method2:AugmentedTreeDataStructure.
Theideaistomaintainrankofeachnode.Wecankeeptrackofelementsinasubtreeofanynodewhile
buildingthetree.SinceweneedKthsmallestelement,wecanmaintainnumberofelementsofleftsubtree
ineverynode.
AssumethattherootishavingNnodesinitsleftsubtree.IfK=N+1,rootisKthnode.IfK<N,wewill
continueoursearch(recursion)fortheKthsmallestelementintheleftsubtreeofroot.IfK>N+1,we
continueoursearchintherightsubtreeforthe(KN1)thsmallestelement.Notethatweneedthecount
ofelementsinleftsubtreeonly.
Timecomplexity:O(n)wherenistotalnodesintree.
Algorithm:
start:
ifK=root.leftElement+1
rootnodeistheKthnode.
gotostop
elseifK>root.leftElements
K=K(root.leftElements+1)
root=root.right
gotostart
else
root=root.left
gotosrart
stop:
Implementation:
#include <stdio.h>
#include <stdlib.h>
#define ARRAY_SIZE(arr) sizeof(arr)/sizeof(arr[0])
typedef struct node_t node_t;
/* Binary tree node */

struct node_t
{
int data;
int lCount;
node_t* left;
node_t* right;
};
/* Iterative insertion
Recursion is least preferred unless we gain something
*/
node_t *insert_node(node_t *root, node_t* node)
{
node_t *currentParent = root;
// Traverse till appropriate node

while(pTraverse)
{
currentParent = pTraverse;
if( node->data < pTraverse->data )

{
/* We are branching to left subtree
increment node count */
pTraverse->lCount++;
/* left subtree */
}
else
{
/* right subtree */
}
}
/* If the tree is empty, make it as root node */

if( !root )
{
root = node;
}
else if( node->data < currentParent->data )
{
/* Insert on left side */
currentParent->left = node;
}
else
{
/* Insert on right side */
currentParent->right = node;
}
return root;
}
/* Elements are in an array. The function builds binary tree */

node_t* binary_search_tree(node_t *root, int keys[], int const size)
{
int iterator;
node_t *new_node = NULL;
for(iterator = 0; iterator < size; iterator++)

{
new_node = (node_t *)malloc( sizeof(node_t) );
/* initialize */
new_node->data
= keys[iterator];
new_node->lCount = 0;
new_node->left
= NULL;
new_node->right = NULL;
/* insert into BST */

root = insert_node(root, new_node);
}
return root;
}
int k_smallest_element(node_t *root, int k)

{
int ret = -1;
if( root )
{
/* Go to k-th smallest */
while(pTraverse)
{
if( (pTraverse->lCount + 1) == k )
{
ret = pTraverse->data;
break;
}
else if( k > pTraverse->lCount )

{
/* There are less nodes on left subtree
Go to right subtree */
k = k - (pTraverse->lCount + 1);
}
else
{
/* The node is on left subtree */
}
}
}
return ret;
}

int main(void)
{
/* just add elements to test */
/* NOTE: A sorted array results in skewed tree */
int ele[] = { 20, 8, 22, 4, 12, 10, 14 };
int i;
node_t* root = NULL;
/* Creating the tree given in the above diagram */

root = binary_search_tree(root, ele, ARRAY_SIZE(ele));
/* It should print the sorted array */

for(i = 1; i <= ARRAY_SIZE(ele); i++)
{
printf("\n kth smallest elment for k = %d is %d",
i, k_smallest_element(root, i));
}
getchar();
return 0;
}
=======================================================================
InorderSuccessorinBinarySearch
Tree
InBinaryTree,InordersuccessorofanodeisthenextnodeinInordertraversaloftheBinaryTree.Inorder
SuccessorisNULLforthelastnodeinInoordertraversal.
InBinarySearchTree,InorderSuccessorofaninputnodecanalsobedefinedasthenodewiththesmallest
keygreaterthanthekeyofinputnode.So,itissometimesimportanttofindnextnodeinsortedorder.
Intheabovediagram,inordersuccessorof8is10,inordersuccessorof10is12andinordersuccessorof
14is20.
Method1(UsesParentPointer)
Inthismethod,weassumethateverynodehasparentpointer.
TheAlgorithmisdividedintotwocasesonthebasisofrightsubtreeoftheinputnodebeingemptyornot.
Input:node,root//nodeisthenodewhoseInordersuccessorisneeded.
output:succ//succisInordersuccessorofnode.
1)IfrightsubtreeofnodeisnotNULL,thensuccliesinrightsubtree.Dofollowing.
Gotorightsubtreeandreturnthenodewithminimumkeyvalueinrightsubtree.
2)IfrightsbtreeofnodeisNULL,thensuccisoneoftheancestors.Dofollowing.
Travelupusingtheparentpointeruntilyouseeanodewhichisleftchildofitsparent.Theparentofsucha
nodeisthesucc.
Implementation
NotethatthefunctiontofindInOrderSuccessorishighlighted(withgraybackground)inbelowcode.
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
struct node* parent;
};
struct node * minValue(struct node* node);
struct node * inOrderSuccessor(struct node *root, struct node *n)

{
// step 1 of the above algorithm
if( n->right != NULL )
return minValue(n->right);

struct node *p = n->parent;
while(p != NULL && n == p->right)
{
n = p;
p = p->parent;
}
return p;
}
/* Given a non-empty binary search tree, return the minimum data
value found in that tree. Note that the entire tree does not need
to be searched. */
struct node * minValue(struct node* node) {

while (current->left != NULL) {
}
return current;
}
{
node->data
= data;
node->left
= NULL;
node->right = NULL;
node->parent = NULL;
return(node);
}
/* Give a binary search tree and a number, inserts a new node with
the given number in the correct place in the tree. Returns the new
root pointer which the caller should then use (the standard trick to
avoid using reference parameters). */
{
/* 1. If the tree is empty, return a new,
single node */
if (node == NULL)
return(newNode(data));
else
{
struct node *temp;
/* 2. Otherwise, recur down the tree */

{
temp = insert(node->left, data);
node->left = temp;
temp->parent= node;
}
else
{
temp = insert(node->right, data);
node->right = temp;
temp->parent = node;
}

return node;
}
}

int main()
{
struct node* root = NULL, *temp, *succ, *min;
//creating the tree given in the above diagram

temp = root->left->right->right;
succ = inOrderSuccessor(root, temp);

if(succ != NULL)
printf("\n Inorder Successor of %d is %d ", temp->data, succ->data);
else
printf("\n Inorder Successor doesn't exit");
getchar();
return 0;
}
Outputoftheaboveprogram:
InorderSuccessorof14is20
TimeComplexity:O(h)wherehisheightoftree.
Method2(Searchfromroot)
ParentpointerisNOTneededinthisalgorithm.TheAlgorithmisdividedintotwocasesonthebasisofright
subtreeoftheinputnodebeingemptyornot.
2)IfrightsbtreeofnodeisNULL,thenstartfromrootandussearchliketechnique.Dofollowing.
Traveldownthetree,ifanodesdataisgreaterthanrootsdatathengorightside,otherwisegotoleftside.
{
struct node *succ = NULL;
// Start from root and search for successor down the tree
{
if (n->data < root->data)
{
succ = root;
root = root->left;
}
else if (n->data > root->data)
root = root->right;
else
break;
}
return succ;
}
=======================================================================
Applicationsoftreedatastructure
DifficultyLevel:Rookie
WhyTree?
UnlikeArrayandLinkedList,whicharelineardatastructures,treeishierarchical(ornonlinear)data
structure.
1)
Onereasontousetreesmightbebecauseyouwanttostoreinformationthatnaturallyformsa
hierarchy.Forexample,thefilesystemonacomputer:
filesystem
/<root
/\
...home
/\
ugradcourse
//|\
...cs101cs112cs113
2)
Ifweorganizekeysinformofatree(withsomeorderinge.g.,BST),wecansearchforagivenkey
inmoderatetime(quickerthanLinkedListandslowerthanarrays).SelfbalancingsearchtreeslikeAVLand
RedBlacktreesguaranteeanupperboundofO(Logn)forsearch.
3)
Wecaninsert/deletekeysinmoderatetime(quickerthanArraysandslowerthanUnorderedLinked
Lists).SelfbalancingsearchtreeslikeAVLandRedBlacktreesguaranteeanupperboundofO(Logn)for
insertion/deletion.
4)
LikeLinkedListsandunlikeArrays,Pointerimplementationoftreesdonthaveanupperlimiton
numberofnodesasnodesarelinkedusingpointers.
AsperWikipedia,followingarethecommonusesoftree.
1.
Manipulatehierarchicaldata.
2.
Makeinformationeasytosearch(seetreetraversal).
3.
Manipulatesortedlistsofdata.
4.
Asaworkflowforcompositingdigitalimagesforvisualeffects.
5.
Routeralgorithms
=======================================================================
Sortedorderprintingofagiven
arraythatrepresentsaBST
GivenanarraythatstoresacompleteBinarySearchTree,writeafunctionthatefficientlyprintsthegiven
arrayinascendingorder.
Forexample,givenanarray[4,2,5,1,3],thefunctionshouldprint1,2,3,4,5
Solution:
InordertraversalofBSTprintsitinascendingorder.Theonlytrickistomodifyrecursiontermination
conditioninstandardInorderTreeTraversal.
Implementation:
#include<stdio.h>
void printSorted(int arr[], int start, int end)

{
if(start > end)
return;
// print left subtree

printSorted(arr, start*2 + 1, end);
// print root
printf("%d ", arr[start]);
// print right subtree

printSorted(arr, start*2 + 2, end);
}
int main()
{
int arr[] = {4, 2, 5, 1, 3};
int arr_size = sizeof(arr)/sizeof(int);
printSorted(arr, 0, arr_size-1);
getchar();
return 0;
}
TimeComplexity:O(n)
=======================================================================
Printnodesatkdistancefromroot
Givenarootofatree,andanintegerk.Printallthenodeswhichareatkdistancefromroot.
Forexample,inthebelowtree,4,5&8areatdistance2fromroot.
1
/\
23
/\/
458
Theproblemcanbesolvedusingrecursion.Thankstoeldhoforsuggestingthesolution.
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
void printKDistant(node *root , int k)

{
if(root == NULL)
return;
if( k == 0 )
{
printf( "%d ", root->data );
return ;
}
else
{
printKDistant( root->left, k-1 ) ;
printKDistant( root->right, k-1 ) ;
}
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{

1
/
/ \
4 5 8
*/
root->left
= newNode(2);
root->right
= newNode(3);
printKDistant(root, 2);
getchar();
return 0;
}
Theaboveprogramprints4,5and8.
TimeComplexity:O(n)wherenisnumberofnodesinthegivenbinarytree.
=======================================================================
FoldableBinaryTrees
Question:Givenabinarytree,findoutifthetreecanbefoldedornot.
Atreecanbefoldedifleftandrightsubtreesofthetreearestructurewisemirrorimageofeachother.An
emptytreeisconsideredasfoldable.
Considerthebelowtrees:
(a)and(b)canbefolded.
(c)and(d)cannotbefolded.
(a)
10
/\
715
\/
911
(b)
10
/\
715
/\
911
(c)
10
/\
715
//
511
(d)
10
/\
715
/\/
91012
Method1(ChangeLeftsubtreetoitsMirrorandcompareitwithRightsubtree)
Algorithm:isFoldable(root)
1)Iftreeisempty,thenreturntrue.
2)Converttheleftsubtreetoitsmirrorimage
mirror(root>left)/*Seethispost*/
3)Checkifthestructureofleftsubtreeandrightsubtreeissame
andstoretheresult.
res=isStructSame(root>left,root>right)/*isStructSame()
recursivelycomparesstructuresoftwosubtreesandreturns
trueifstructuresaresame*/
4)Revertthechangesmadeinstep(2)togettheoriginaltree.
mirror(root>left)
5)Returnresultresstoredinstep2.
Thankstoajaymforsuggestingthisapproach.
#include<stdio.h>
#include<stdlib.h>
/* You would want to remove below 3 lines if your compiler

supports bool, true and false */
#define bool int
#define true 1
#define false 0

struct node
{
int data;
struct node* left;
struct node* right;
};
/* converts a tree to its mrror image */

void mirror(struct node* node);
/* returns true if structure of two trees a and b is same

Only structure is considered for comparison, not data! */
bool isStructSame(struct node *a, struct node *b);
/* Returns true if the given tree is foldable */

bool isFoldable(struct node *root)
{
bool res;
/* base case */
if(root == NULL)
return true;
/* convert left subtree to its mirror */

mirror(root->left);
/* Compare the structures of the right subtree and mirrored

left subtree */
res = isStructSame(root->left, root->right);
/* Get the originial tree back */

mirror(root->left);
return res;
}
bool isStructSame(struct node *a, struct node *b)

{
if (a == NULL && b == NULL)
{ return true; }
if ( a != NULL && b != NULL &&
isStructSame(a->left, b->left) &&
isStructSame(a->right, b->right)
)
{ return true; }
return false;
}
/* Change a tree so that the roles of the left and
right pointers are swapped at every node.
See http://geeksforgeeks.org/?p=662 for details */
void mirror(struct node* node)
{
if (node==NULL)
return;
else
{
struct node* temp;
/* do the subtrees */
mirror(node->left);
mirror(node->right);
/* swap the pointers in this node */

temp
= node->left;
node->left = node->right;
node->right = temp;
}
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Driver program to test mirror() */

int main(void)
{
/* The constructed binary tree is
1
/
4 5
*/
root->left
= newNode(2);
root->right
= newNode(3);
if(isFoldable(root) == 1)
{ printf("\n tree is foldable"); }
else
{ printf("\n tree is not foldable"); }
getchar();
return 0;
}
Timecomplexity:O(n)
Method2(CheckifLeftandRightsubtreesareMirror)
Therearemainlytwofunctions:
//Checksiftreecanbefoldedornot
IsFoldable(root)
1)Iftreeisemptythenreturntrue
2)Elsecheckifleftandrightsubtreesarestructurewisemirrorsof
eachother.UseutilityfunctionIsFoldableUtil(root>left,
root>right)forthis.
//Checksifn1andn2aremirrorofeachother.
IsFoldableUtil(n1,n2)
1)Ifbothtreesareemptythenreturntrue.
2)Ifoneofthemisemptyandotherisnotthenreturnfalse.
3)Returntrueiffollowingconditionsaremet
a)n1>leftismirrorofn2>right
b)n1>rightismirrorofn2>left
#include<stdio.h>
#include<stdlib.h>
/* You would want to remove below 3 lines if your compiler

supports bool, true and false */
#define bool int
#define true 1
#define false 0

struct node
{
int data;
struct node* left;
struct node* right;
};
/* A utility function that checks if trees with roots as n1 and n2

are mirror of each other */
bool IsFoldableUtil(struct node *n1, struct node *n2);
/* Returns true if the given tree can be folded */

bool IsFoldable(struct node *root)
{
if (root == NULL)
{ return true; }
return IsFoldableUtil(root->left, root->right);

}
/* A utility function that checks if trees with roots as n1 and n2
are mirror of each other */

bool IsFoldableUtil(struct node *n1, struct node *n2)
{
/* If both left and right subtrees are NULL,
then return true */
if (n1 == NULL && n2 == NULL)
{ return true; }
/* If one of the trees is NULL and other is not,

then return false */
if (n1 == NULL || n2 == NULL)
{ return false; }
/* Otherwise check if left and right subtrees are mirrors of

their counterparts */
return IsFoldableUtil(n1->left, n2->right) &&
IsFoldableUtil(n1->right, n2->left);
}
/*UTILITY FUNCTIONS */
{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main(void)
{
/* The constructed binary tree is
1
/
4 5
*/
root->left
= newNode(2);
root->right
= newNode(3);
if(IsFoldable(root) == true)
{ printf("\n tree is foldable"); }
else
{ printf("\n tree is not foldable"); }
getchar();
return 0;
}
================================================================
=======
TotalnumberofpossibleBinary
SearchTreeswithnkeys
TotalnumberofpossibleBinarySearchTreeswithndifferentkeys=CatalannumberCn=(2n)!/(n+1)!*n!
Seereferencesforproofandexamples.
References:
http://en.wikipedia.org/wiki/Catalan_number
=======================================================================
Maximumwidthofabinarytree
Givenabinarytree,writeafunctiontogetthemaximumwidthofthegiventree.Widthofatreeismaximum
ofwidthsofalllevels.
Letusconsiderthebelowexampletree.
1
/\
23
/\\
458
/\
67
Fortheabovetree,
widthoflevel1is1,
widthoflevel2is2,
widthoflevel3is3
widthoflevel4is2.
Sothemaximumwidthofthetreeis3.
Method1(UsingLevelOrderTraversal)
Thismethodmainlyinvolvestwofunctions.Oneistocountnodesatagivenlevel(getWidth),andotheristo
getthemaximumwidthofthetree(getMaxWidth).getMaxWidth()makesuseofgetWidth()togetthewidthof
alllevelsstartingfromroot.
/*Functiontoprintlevelordertraversaloftree*/
getMaxWidth(tree)
maxWdth=0
fori=1toheight(tree)
width=getWidth(tree,i)
if(width>maxWdth)
maxWdth=width
returnwidth
/*Functiontogetwidthofagivenlevel*/
getWidth(tree,level)
iftreeisNULLthenreturn0
iflevelis1,thenreturn1
elseiflevelgreaterthan1,then
returngetWidth(tree>left,level1)+
getWidth(tree>right,level1)
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
int getWidth(struct node* root, int level);
/* Function to get the maximum width of a binary tree*/

int getMaxWidth(struct node* root)
{
int maxWidth = 0;
int width;
int i;
/* Get width of each level and compare

the width with maximum width so far */
for(i=1; i<=h; i++)
{
width = getWidth(root, i);
if(width > maxWidth)
maxWidth = width;
}
return maxWidth;
}
/* Get width of a given level */

int getWidth(struct node* root, int level)
{
if(root == NULL)
return 0;
if(level == 1)
return 1;
else if (level > 1)

return getWidth(root->left, level-1) +
getWidth(root->right, level-1);
}
{
if (node==NULL)
return 0;
else
{
int lHeight = height(node->left);
int rHeight = height(node->right);
return (lHeight > rHeight)? (lHeight+1): (rHeight+1);

}
}
{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
/*
Constructed bunary tree is:
1
/ \
2
/ \
8
/ \
6
*/
printf("Maximum width is %d \n", getMaxWidth(root));
getchar();
return 0;
}
TimeComplexity:O(n^2)intheworstcase.
WecanuseQueuebasedlevelordertraversaltooptimizethetimecomplexityofthismethod.TheQueue
basedlevelordertraversalwilltakeO(n)timeinworstcase.ThankstoNitish,DivyaCandtech.login.id2for
suggestingthisoptimization.Seetheircommentsforimplementationusingqueuebasedtraversal.
Method2(UsingPreorderTraversal)
Inthismethodwecreateatemporaryarraycount[]ofsizeequaltotheheightoftree.Weinitializeallvalues
incountas0.Wetraversethetreeusingpreordertraversalandfilltheentriesincountsothatthecount
arraycontainscountofnodesateachlevelinBinaryTree.
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
// A utility function to get height of a binary tree

// A utility function to allocate a new node with given data

// A utility function that returns maximum value in arr[] of size n

int getMax(int arr[], int n);
// A function that fills count array with count of nodes at every

// level of given binary tree
void getMaxWidthRecur(struct node *root, int count[], int level);
/* Function to get the maximum width of a binary tree*/

int getMaxWidth(struct node* root)
{
int width;
// Create an array that will store count of nodes at each level

int *count = (int *)calloc(sizeof(int), h);
int level = 0;
// Fill the count array using preorder traversal

getMaxWidthRecur(root, count, level);
// Return the maximum value from count array

return getMax(count, h);
}
// A function that fills count array with count of nodes at every

// level of given binary tree
void getMaxWidthRecur(struct node *root, int count[], int level)
{
if(root)
{
count[level]++;
getMaxWidthRecur(root->left, count, level+1);
getMaxWidthRecur(root->right, count, level+1);
}
}

{
if (node==NULL)
return 0;
else
{
int lHeight = height(node->left);
int rHeight = height(node->right);
return (lHeight > rHeight)? (lHeight+1): (rHeight+1);

}
}
{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
// Return the maximum value from count array

int getMax(int arr[], int n)
{
int max = arr[0];
int i;
for (i = 0; i < n; i++)
{
if (arr[i] > max)
max = arr[i];
}
return max;

int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
/*
Constructed bunary tree is:
1
/ \
2
/ \
8
/ \
6
*/
printf("Maximum width is %d \n", getMaxWidth(root));
getchar();
return 0;
}
TimeComplexity:O(n)
=======================================================================
DoubleTree
WriteaprogramthatconvertsagiventreetoitsDoubletree.TocreateDoubletreeofthegiventree,create
anewduplicateforeachnode,andinserttheduplicateastheleftchildoftheoriginalnode.
Sothetree
2
/\
13
ischangedto
2
/\
23
//
13
/
1
Andthetree
1
/\
23
/\
45
ischangedto
1
/\
13
//
23
/\
25
//
45
/
4
Algorithm:
Recursivelyconvertthetreetodoubletreeinpostorderfashion.Foreachnode,firstconverttheleftsubtree
ofthenode,thenrightsubtree,finallycreateaduplicatenodeofthenodeandfixtheleftchildofthenode
andleftchildofleftchild.
Implementation:
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
/* function to create a new node of tree and returns pointer */

/* Function to convert a tree to double tree */

void doubleTree(struct node* node)
{
struct node* oldLeft;
doubleTree(node->left);
doubleTree(node->right);
/* duplicate this node to its left */

oldLeft = node->left;
node->left = newNode(node->data);
node->left->left = oldLeft;
}
/* UTILITY FUNCTIONS TO TEST doubleTree() FUNCTION */

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Given a binary tree, print its nodes in inorder*/

{
if (node == NULL)
return;
}

int main()
{

1
/
2
/ \
4 5
*/
\
3

root->left
= newNode(2);
root->right
= newNode(3);
printf("Inorder traversal of the original tree is \n");

printInorder(root);
doubleTree(root);
printf("\n Inorder traversal of the double tree is \n");

printInorder(root);
getchar();
return 0;
}
TimeComplexity:O(n)wherenisthenumberofnodesinthetree.
=======================================================================
Givenabinarytree,printall
roottoleafpaths
Forthebelowexampletree,allroottoleafpathsare:
10>8>3
10>8>5
10>2>2

Algorithm:
Useapatharraypath[]tostorecurrentroottoleafpath.Traversefromroottoallleavesintopdownfashion.
Whiletraversing,storedataofallnodesincurrentpathinarraypath[].Whenwereachaleafnode,printthe
patharray.
#include<stdio.h>
#include<stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
/* Prototypes for funtions needed in printPaths() */

void printPathsRecur(struct node* node, int path[], int pathLen);
void printArray(int ints[], int len);
/*Given a binary tree, print out all of its root-to-leaf

paths, one per line. Uses a recursive helper to do the work.*/
void printPaths(struct node* node)
{
int path[1000];
printPathsRecur(node, path, 0);
}
/* Recursive helper function -- given a node, and an array containing

the path from the root node up to but not including this node,
print out all the root-leaf paths.*/
void printPathsRecur(struct node* node, int path[], int pathLen)
{
if (node==NULL)
return;
/* append this node to the path array */

path[pathLen] = node->data;
pathLen++;
/* it's a leaf, so print the path that led to here */

if (node->left==NULL && node->right==NULL)
{
printArray(path, pathLen);
}
else
{
/* otherwise try both subtrees */
printPathsRecur(node->left, path, pathLen);
printPathsRecur(node->right, path, pathLen);
}
}
/* Utility that prints out an array on a line. */
void printArray(int ints[], int len)
{
int i;
for (i=0; i<len; i++)
{
printf("%d ", ints[i]);
}
printf("\n");
/* utility that allocates a new node with the

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{

10
/
/ \
3 5 2
*/
root->left
= newnode(8);
root->right
= newnode(2);
root->right->left = newnode(2);
printPaths(root);
getchar();
return 0;
}
TimeComplexity:O(n)
=======================================================================
ConstructTreefromgivenInorder
andPreordertraversals
Letusconsiderthebelowtraversals:
Inordersequence:DBEAFC
Preordersequence:ABDECF
InaPreordersequence,leftmostelementistherootofthetree.SoweknowAisrootforgivensequences.
BysearchingAinInordersequence,wecanfindoutallelementsonleftsideofAareinleftsubtreeand
elementsonrightareinrightsubtree.Soweknowbelowstructurenow.
A
/\
/\
DBEFC
Werecursivelyfollowabovestepsandgetthefollowingtree.
A
/\
/\
BC
/\/
/\/
DEF
Algorithm:buildTree()
1)PickanelementfromPreorder.IncrementaPreorderIndexVariable(preIndexinbelowcode)topicknext
elementinnextrecursivecall.
2)CreateanewtreenodetNodewiththedataaspickedelement.
3)FindthepickedelementsindexinInorder.LettheindexbeinIndex.
4)CallbuildTreeforelementsbeforeinIndexandmakethebuilttreeasleftsubtreeoftNode.
5)CallbuildTreeforelementsafterinIndexandmakethebuilttreeasrightsubtreeoftNode.
6)returntNode.
ThankstoRohiniandTusharforsuggestingthecode.
/* program to construct tree using inorder and preorder traversals */
#include<stdio.h>
#include<stdlib.h>

struct node
{
char data;
struct node* left;
struct node* right;
};
/* Prototypes for utility functions */

int search(char arr[], int strt, int end, char value);
struct node* newNode(char data);
/* Recursive function to construct binary of size len from

Inorder traversal in[] and Preorder traversal pre[]. Initial values
of inStrt and inEnd should be 0 and len -1. The function doesn't
do any error checking for cases where inorder and preorder
do not form a tree */
struct node* buildTree(char in[], char pre[], int inStrt, int inEnd)
{
static int preIndex = 0;
if(inStrt > inEnd)

return NULL;
/* Pick current node from Preorder traversal using preIndex

and increment preIndex */
struct node *tNode = newNode(pre[preIndex++]);
/* If this node has no children then return */

if(inStrt == inEnd)
return tNode;
/* Else find the index of this node in Inorder traversal */

int inIndex = search(in, inStrt, inEnd, tNode->data);
/* Using index in Inorder traversal, construct left and

right subtress */
tNode->left = buildTree(in, pre, inStrt, inIndex-1);
tNode->right = buildTree(in, pre, inIndex+1, inEnd);
return tNode;
}
/* Function to find index of value in arr[start...end]
The function assumes that value is present in in[] */
int search(char arr[], int strt, int end, char value)
{
int i;
for(i = strt; i <= end; i++)
{
if(arr[i] == value)
return i;
}
}

struct node* newNode(char data)
{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* This funtcion is here just to test buildTree() */

{
if (node == NULL)
return;


printf("%c ", node->data);

}

int main()
{
char in[] = {'D', 'B', 'E', 'A', 'F', 'C'};
char pre[] = {'A', 'B', 'D', 'E', 'C', 'F'};
int len = sizeof(in)/sizeof(in[0]);
struct node *root = buildTree(in, pre, 0, len - 1);

printInorder(root);
getchar();
}
TimeComplexity:O(n^2).Worstcaseoccurswhentreeisleftskewed.ExamplePreorderandInorder
traversalsforworstcaseare{A,B,C,D}and{D,C,B,A}.
=======================================================================
Roottoleafpathsumequaltoa
givennumber
Givenabinarytreeandanumber,returntrueifthetreehasaroottoleafpathsuchthataddingupallthe
valuesalongthepathequalsthegivennumber.Returnfalseifnosuchpathcanbefound.
Forexample,intheabovetreeroottoleafpathsexistwithfollowingsums.
21>1083
23>1085
14>1022
Sothereturnedvalueshouldbetrueonlyfornumbers21,23and14.Foranyothernumber,returnedvalue
shouldbefalse.
Algorithm:
Recursivelycheckifleftorrightchildhaspathsumequalto(numbervalueatcurrentnode)
Implementation:
#include<stdio.h>
#include<stdlib.h>
#define bool int

struct node
{
int data;
struct node* left;
struct node* right;

};
/*
Given a tree and a sum, return true if there is a path from the root
down to a leaf, such that adding up all the values along the path
equals the given sum.
Strategy: subtract the node value from the sum when recurring down,
and check to see if the sum is 0 when you run out of tree.
*/
bool hasPathSum(struct node* node, int sum)
{
/* return true if we run out of tree and sum==0 */
if (node == NULL)
{
return (sum == 0);
}
else
{
bool ans = 0;
/* otherwise check both subtrees */

int subSum = sum - node->data;
/* If we reach a leaf node and sum becomes 0 then return true*/

if ( subSum == 0 && node->left == NULL && node->right == NULL )
return 1;
if(node->left)
ans = ans || hasPathSum(node->left, subSum);
if(node->right)
ans = ans || hasPathSum(node->right, subSum);
return ans;
}
{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
int sum = 21;

10
/
/ \
3 5 2
*/
root->left
= newnode(8);
root->right
= newnode(2);
root->right->left = newnode(2);
if(hasPathSum(root, sum))
printf("There is a root-to-leaf path with sum %d", sum);

else
printf("There is no root-to-leaf path with sum %d", sum);
getchar();
return 0;
}
TimeComplexity:O(n)
References:
=======================================================================
InorderTreeTraversalwithout
recursionandwithoutstack!
UsingMorrisTraversal,wecantraversethetreewithoutusingstackandrecursion.TheideaofMorris
TraversalisbasedonThreadedBinaryTree.Inthistraversal,wefirstcreatelinkstoInordersuccessorand
printthedatausingtheselinks,andfinallyrevertthechangestorestoreoriginaltree.
1.Initializecurrentasroot
2.WhilecurrentisnotNULL
Ifcurrentdoesnothaveleftchild
a)Printcurrentsdata
b)Gototheright,i.e.,current=current>right
Else
a)Makecurrentasrightchildoftherightmostnodeincurrent'sleftsubtree
b)Gotothisleftchild,i.e.,current=current>left
Althoughthetreeismodifiedthroughthetraversal,itisrevertedbacktoitsoriginalshapeafterthe
completion.UnlikeStackbasedtraversal,noextraspaceisrequiredforthistraversal.
#include<stdio.h>
#include<stdlib.h>
/* A binary tree tNode has data, pointer to left child

struct tNode
{
int data;
struct tNode* left;
struct tNode* right;
};
/* Function to traverse binary tree without recursion and

without stack */
void MorrisTraversal(struct tNode *root)
{
struct tNode *current,*pre;
if(root == NULL)
return;
current = root;
while(current != NULL)
{
if(current->left == NULL)
{
printf(" %d ", current->data);
}
else
{
/* Find the inorder predecessor of current */
pre = current->left;
while(pre->right != NULL && pre->right != current)
pre = pre->right;
/* Make current as right child of its inorder predecessor */

if(pre->right == NULL)
{
pre->right = current;
}
/* Revert the changes made in if part to restore the original
tree i.e., fix the right child of predecssor */

else
{
pre->right = NULL;
printf(" %d ",current->data);
} /* End of if condition pre->right == NULL */
} /* End of if condition current->left == NULL*/
} /* End of while */
}
/* Helper function that allocates a new tNode with the
struct tNode* newtNode(int data)
{
struct tNode* tNode = (struct tNode*)
malloc(sizeof(struct tNode));
tNode->data = data;
tNode->left = NULL;
tNode->right = NULL;
return(tNode);
}

int main()
{

1
/
2
\
3
/ \
4 5
*/
struct tNode *root = newtNode(1);
root->left
= newtNode(2);
root->right
= newtNode(3);
root->left->left = newtNode(4);
root->left->right = newtNode(5);
MorrisTraversal(root);
getchar();
return 0;
}
================================================================
=======
InorderTreeTraversalwithout
Recursion
UsingStackistheobviouswaytotraversetreewithoutrecursion.Belowisanalgorithmfortraversingbinary
treeusingstack.Seethisforstepwisestepexecutionofthealgorithm.
1)CreateanemptystackS.
2)Initializecurrentnodeasroot
3)PushthecurrentnodetoSandsetcurrent=current>leftuntilcurrentisNULL
4)IfcurrentisNULLandstackisnotemptythen
a)Popthetopitemfromstack.
b)Printthepoppeditem,setcurrent=popped_item>right
c)Gotostep3.
5)IfcurrentisNULLandstackisemptythenwearedone.
Letusconsiderthebelowtreeforexample
1
/\
23
/\
45
Step1Createsanemptystack:S=NULL
Step2setscurrentasaddressofroot:current>1
Step3Pushesthecurrentnodeandsetcurrent=current>leftuntilcurrentisNULL
current>1
push1:StackS>1
current>2
push2:StackS>2,1
current>4
push4:StackS>4,2,1
current=NULL
Step4popsfromS
a)Pop4:StackS>2,1
b)print"4"
c)current=NULL/*rightof4*/andgotostep3
SincecurrentisNULLstep3doesn'tdoanything.
Step4popsagain.
a)Pop2:StackS>1
b)print"2"
c)current>5/*rightof2*/andgotostep3
Step3pushes5tostackandmakescurrentNULL
StackS>5,1
current=NULL
Step4popsfromS
a)Pop5:StackS>1
b)print"5"
c)current=NULL/*rightof5*/andgotostep3
SincecurrentisNULLstep3doesn'tdoanything
Step4popsagain.
a)Pop1:StackS>NULL
b)print"1"
c)current>3/*rightof5*/
Step3pushes3tostackandmakescurrentNULL
StackS>3
current=NULL
Step4popsfromS
a)Pop3:StackS>NULL
b)print"3"
c)current=NULL/*rightof3*/
TraversalisdonenowasstackSisemptyandcurrentisNULL.
Implementation:
#include<stdio.h>
#include<stdlib.h>
#define bool int
/* A binary tree tNode has data, pointer to left child

struct tNode
{
int data;
struct tNode* left;
struct tNode* right;
};
/* Structure of a stack node. Linked List implementation is used for

stack. A stack node contains a pointer to tree node and a pointer to
next stack node */
struct sNode
{
struct tNode *t;
struct sNode *next;
};
/* Stack related functions */
void push(struct sNode** top_ref, struct tNode *t);

struct tNode *pop(struct sNode** top_ref);
bool isEmpty(struct sNode *top);
/* Iterative function for inorder tree traversal */

void inOrder(struct tNode *root)
{
/* set current to root of binary tree */
struct tNode *current = root;
struct sNode *s = NULL; /* Initialize stack s */
bool done = 0;
while (!done)
{
/* Reach the left most tNode of the current tNode */
if(current != NULL)
{
/* place pointer to a tree node on the stack before traversing
the node's left subtree */
push(&s, current);
}
/* backtrack from the empty subtree and visit the tNode

at the top of the stack; however, if the stack is empty,
you are done */
else
{
if (!isEmpty(s))
{
current = pop(&s);
printf("%d ", current->data);
/* we have visited the node and its left subtree.

Now, it's right subtree's turn */
}
else
done = 1;
}
} /* end of while */
}
/* Function to push an item to sNode*/
void push(struct sNode** top_ref, struct tNode *t)
{
/* allocate tNode */
struct sNode* new_tNode =
(struct sNode*) malloc(sizeof(struct sNode));
if(new_tNode == NULL)
{
printf("Stack Overflow \n");
getchar();
exit(0);
}
/* put in the data */

new_tNode->t = t;
/* link the old list off the new tNode */

new_tNode->next = (*top_ref);
/* move the head to point to the new tNode */

(*top_ref) = new_tNode;
}
/* The function returns true if stack is empty, otherwise false */

bool isEmpty(struct sNode *top)
{
return (top == NULL)? 1 : 0;
}
/* Function to pop an item from stack*/

struct tNode *pop(struct sNode** top_ref)
{
struct tNode *res;
struct sNode *top;
/*If sNode is empty then error */

if(isEmpty(*top_ref))
{
printf("Stack Underflow \n");
getchar();
exit(0);
}
else
{
top = *top_ref;
res = top->t;
*top_ref = top->next;
free(top);
return res;
}
}
/* Helper function that allocates a new tNode with the

struct tNode* newtNode(int data)
{
struct tNode* tNode = (struct tNode*)
malloc(sizeof(struct tNode));
tNode->data = data;
tNode->left = NULL;
tNode->right = NULL;
return(tNode);
}
int main()
{

1
/
2
\
3
/ \
4 5
*/
struct tNode *root = newtNode(1);
root->left
= newtNode(2);
root->right
= newtNode(3);
root->left->left = newtNode(4);
root->left->right = newtNode(5);
inOrder(root);
getchar();
return 0;
}
TimeComplexity:O(n)
=======================================================================
Howtodetermineifabinarytreeis
heightbalanced?
Atreewherenoleafismuchfartherawayfromtherootthananyotherleaf.Differentbalancingschemes
allowdifferentdefinitionsofmuchfartheranddifferentamountsofworktokeepthembalanced.
Consideraheightbalancingschemewherefollowingconditionsshouldbecheckedtodetermineifabinary
treeisbalanced.
Anemptytreeisheightbalanced.AnonemptybinarytreeTisbalancedif:
1)LeftsubtreeofTisbalanced
2)RightsubtreeofTisbalanced
3)Thedifferencebetweenheightsofleftsubtreeandrightsubtreeisnotmorethan1.
TheaboveheightbalancingschemeisusedinAVLtrees.Thediagrambelowshowstwotrees,oneofthem
isheightbalancedandotherisnot.Thesecondtreeisnotheightbalancedbecauseheightofleftsubtreeis
2morethanheightofrightsubtree.
Tocheckifatreeisheightbalanced,gettheheightofleftandrightsubtrees.Returntrueifdifference
betweenheightsisnotmorethan1andleftandrightsubtreesarebalanced,otherwisereturnfalse.
/* program to check if a tree is height-balanced or not */
#include<stdio.h>
#include<stdlib.h>
#define bool int

struct node
{
int data;
struct node* left;
struct node* right;
};
/* Returns the height of a binary tree */

/* Returns true if binary tree with root as root is height-balanced */

bool isBalanced(struct node *root)
{
int lh; /* for height of left subtree */
int rh; /* for height of right subtree */
/* If tree is empty then return true */

if(root == NULL)
return 1;
/* Get the height of left and right sub trees */

lh = height(root->left);
rh = height(root->right);
if( abs(lh-rh) <= 1 &&

isBalanced(root->left) &&
isBalanced(root->right))
return 1;
/* If we reach here then tree is not height-balanced */

return 0;
}
/* UTILITY FUNCTIONS TO TEST isBalanced() FUNCTION */
/* returns maximum of two integers */

{
return (a >= b)? a: b;
}
/* The function Compute the "height" of a tree. Height is the

number of nodes along the longest path from the root node
{
/* base case tree is empty */
if(node == NULL)
return 0;
/* If tree is not empty then height = 1 + max of left

height and right heights */
return 1 + max(height(node->left), height(node->right));
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
int main()
{
if(isBalanced(root))
printf("Tree is balanced");
else
printf("Tree is not balanced");
getchar();
return 0;
}
TimeComplexity:O(n^2)Worstcaseoccursincaseofskewedtree.
Optimizedimplementation:Aboveimplementationcanbeoptimizedbycalculatingtheheightinthesame
recursionratherthancallingaheight()functionseparately.ThankstoAmarforsuggestingthisoptimized
version.ThisoptimizationreducestimecomplexitytoO(n).
/* program to check if a tree is height-balanced or not */
#include<stdio.h>
#include<stdlib.h>
#define bool int

struct node
{
int data;
struct node* left;
struct node* right;
};
/* The function returns true if root is balanced else false

The second parameter is to store the height of tree.
Initially, we need to pass a pointer to a location with value
as 0. We can also write a wrapper over this function */
bool isBalanced(struct node *root, int* height)
{
/* lh --> Height of left subtree
rh --> Height of right subtree */
int lh = 0, rh = 0;
/* l will be true if left subtree is balanced

and r will be true if right subtree is balanced */
int l = 0, r = 0;
if(root == NULL)
{
*height = 0;
return 1;
/* Get the heights of left and right subtrees in lh and rh

And store the returned values in l and r */
l = isBalanced(root->left, &lh);
r = isBalanced(root->right,&rh);
/* Height of current node is max of heights of left and

right subtrees plus 1*/
*height = (lh > rh? lh: rh) + 1;
/* If difference between heights of left and right

subtrees is more than 2 then this node is not balanced
so return 0 */
if((lh - rh >= 2) || (rh - lh >= 2))
return 0;
/* If this node is balanced and left and right subtrees

are balanced then return true */
else return l&&r;
}
/* UTILITY FUNCTIONS TO TEST isBalanced() FUNCTION */

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
int main()
{
int height = 0;

1
/
2
\
3
/ \ /
4
5 6
/
7
*/
if(isBalanced(root, &height))
printf("Tree is balanced");
else
printf("Tree is not balanced");
getchar();
return 0;
}
TimeComplexity:O(n)
=======================================================================
DiameterofaBinaryTree
Thediameterofatree(sometimescalledthewidth)isthenumberofnodesonthelongestpathbetweentwo
leavesinthetree.Thediagrambelowshowstwotreeseachwithdiameternine,theleavesthatformthe
endsofalongestpathareshaded(notethatthereismorethanonepathineachtreeoflengthnine,butno
pathlongerthanninenodes).
ThediameterofatreeTisthelargestofthefollowingquantities:
*thediameterofTsleftsubtree
*thediameterofTsrightsubtree
*thelongestpathbetweenleavesthatgoesthroughtherootofT(thiscanbecomputedfromtheheightsof
thesubtreesofT)
Implementation:
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
/* function to create a new node of tree and returns pointer */
/* returns max of two integers */

/* function to Compute height of a tree. */

/* Function to get diameter of a binary tree */

int diameter(struct node * tree)
{
/* base case where tree is empty */
if (tree == 0)
return 0;
/* get the height of left and right sub-trees */

int lheight = height(tree->left);
int rheight = height(tree->right);
/* get the diameter of left and right sub-trees */

int ldiameter = diameter(tree->left);
int rdiameter = diameter(tree->right);
/* Return max of following three

1) Diameter of left subtree
2) Diameter of right subtree
3) Height of left subtree + height of right subtree + 1 */
return max(lheight + rheight + 1, max(ldiameter, rdiameter));
}
/* UTILITY FUNCTIONS TO TEST diameter() FUNCTION */
/* The function Compute the "height" of a tree. Height is the

number f nodes along the longest path from the root node
{
/* base case tree is empty */

if(node == NULL)
return 0;
/* If tree is not empty then height = 1 + max of left

height and right heights */
return 1 + max(height(node->left), height(node->right));
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* returns maximum of two integers */

{
return (a >= b)? a: b;
}

int main()
{

1
/
2
/ \
\
3
4 5
*/
root->left
= newNode(2);
root->right
= newNode(3);
printf("Diameter of the given binary tree is %d\n", diameter(root));
getchar();
return 0;
}
TimeComplexity:O(n^2)
Optimizedimplementation:Theaboveimplementationcanbeoptimizedbycalculatingtheheightinthe
samerecursionratherthancallingaheight()separately.ThankstoAmarforsuggestingthisoptimized
version.ThisoptimizationreducestimecomplexitytoO(n).
/*The second parameter is to store the height of tree.
Initially, we need to pass a pointer to a location with value
as 0. So, function should be used as follows:
int height = 0;
struct node *root = SomeFunctionToMakeTree();
int diameter = diameterOpt(root, &height); */
int diameterOpt(struct node *root, int* height)
{
/* lh --> Height of left subtree
rh --> Height of right subtree */
int lh = 0, rh = 0;
/* ldiameter --> diameter of left subtree

rdiameter --> Diameter of right subtree */
int ldiameter = 0, rdiameter = 0;
if(root == NULL)
{
*height = 0;
return 0; /* diameter is also 0 */
}
/* Get the heights of left and right subtrees in lh and rh

And store the returned values in ldiameter and ldiameter */
ldiameter = diameterOpt(root->left, &lh);
rdiameter = diameterOpt(root->right, &rh);
/* Height of current node is max of heights of left and

right subtrees plus 1*/
*height = max(lh, rh) + 1;
return max(lh + rh + 1, max(ldiameter, rdiameter));

}
TimeComplexity:O(n)
================================================================
=======
ConvertanarbitraryBinaryTreeto
atreethatholdsChildrenSum
Property
Question:Givenanarbitrarybinarytree,convertittoabinarytreethatholdsChildrenSumProperty.You
canonlyincrementdatavaluesinanynode(Youcannotchangestructureoftreeandcannotdecrement
valueofanynode).
Forexample,thebelowtreedoesntholdthechildrensumproperty,convertittoatreethatholdsthe
property.
50
/\
/\
72
/\/\
/\/\
35130
Algorithm:
Traversegiventreeinpostordertoconvertit,i.e.,firstchangeleftandrightchildrentoholdthechildrensum
propertythenchangetheparentnode.
Letdifferencebetweennodesdataandchildrensumbediff.
diff=nodeschildrensumnodesdata
Ifdiffis0thennothingneedstobedone.
Ifdiff>0(nodesdataissmallerthannodeschildrensum)incrementthenodesdatabydiff.
Ifdiff<0(nodesdataisgreaterthanthenode'schildrensum)thenincrementonechildsdata.Wecan
choosetoincrementeitherleftorrightchildiftheybotharenotNULL.Letusalwaysfirstincrementtheleft
child.Incrementingachildchangesthesubtreeschildrensumpropertysoweneedtochangeleftsubtree
also.Sowerecursivelyincrementtheleftchild.Ifleftchildisemptythenwerecursivelycallincrement()for
rightchild.
Letusrunthealgorithmforthegivenexample.
Firstconverttheleftsubtree(increment7to8).
50
/\
/\
82
/\/\
/\/\
35130
Thenconverttherightsubtree(increment2to31)
50
/\
/\
831
/\/\
/\/\
35130
Nowconverttheroot,wehavetoincrementleftsubtreeforconvertingtheroot.
50
/\
/\
1931
/\/\
/\/\
145130
Pleasenotethelaststepwehaveincremented8to19,andtofixthesubtreewehaveincremented3to
14.
Implementation:
/* Program to convert an aribitary binary tree to
a tree that holds children sum property */
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node* left;
struct node* right;
};
/* This function is used to increment left subtree */

void increment(struct node* node, int diff);

pointers. */
/* This function changes a tree to to hold children sum

property */
void convertTree(struct node* node)
{
int left_data = 0, right_data = 0, diff;
/* If tree is empty or it's a leaf node then

return true */
if (node == NULL ||
return;
else
{
/* convert left and right subtrees */
convertTree(node->left);
convertTree(node->right);
/* If left child is not present then 0 is used

as data of left child */
left_data = node->left->data;
/* If right child is not present then 0 is used

as data of right child */
right_data = node->right->data;
/* get the diff of node's data and children sum */

diff = left_data + right_data - node->data;
/* If node's children sum is greater than the node's data */

if (diff > 0)
node->data = node->data + diff;
/* THIS IS TRICKY --> If node's data is greater than children sum,

then increment subtree by diff */
if (diff < 0)
increment(node, -diff); // -diff is used to make diff positive
}
}
/* This function is used to increment subtree by diff */

void increment(struct node* node, int diff)
{
/* IF left child is not NULL then increment it */
if(node->left != NULL)
{
node->left->data = node->left->data + diff;
// Recursively call to fix the descendants of node->left

increment(node->left, diff);
}
else if (node->right != NULL) // Else increment right child
{
node->right->data = node->right->data + diff;
// Recursively call to fix the descendants of node->right

increment(node->right, diff);
}
}
/* Given a binary tree, printInorder() prints out its

inorder traversal*/
{
if (node == NULL)
return;



}

pointers. */
{
struct node* node =
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(7);
root->right
= newNode(2);
printf("\n Inorder traversal before conversion ");

printInorder(root);
convertTree(root);
printf("\n Inorder traversal after conversion ");

printInorder(root);
getchar();
return 0;
}
TimeComplexity:O(n^2),Worstcasecomplexityisforaskewedtreesuchthatnodesareindecreasing
orderfromroottoleaf.
=======================================================================
CheckforChildrenSumPropertyin
aBinaryTree.
Givenabinarytree,writeafunctionthatreturnstrueifthetreesatisfiesbelowproperty.
Foreverynode,datavaluemustbeequaltosumofdatavaluesinleftandrightchildren.Considerdata
valueas0forNULLchildren.Belowtreeisanexample
Algorithm:
Traversethegivenbinarytree.Foreachnodecheck(recursively)ifthenodeandbothitschildrensatisfythe
ChildrenSumProperty,ifsothenreturntrueelsereturnfalse.
Implementation:
/* Program to check children sum property */
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
struct node* left;
struct node* right;
};
/* returns 1 if children sum property holds for the given

node and both of its children*/
int isSumProperty(struct node* node)
{
/* left_data is left child data and right_data is for right child data*/
int left_data = 0, right_data = 0;

return true */
if(node == NULL ||
return 1;
else
{
/* If left child is not present then 0 is used
as data of left child */
if(node->left != NULL)
left_data = node->left->data;
/* If right child is not present then 0 is used

as data of right child */
if(node->right != NULL)
right_data = node->right->data;

if((node->data == left_data + right_data)&&
isSumProperty(node->left) &&
isSumProperty(node->right))
return 1;
else
return 0;
}
}
/*

pointers.
*/
{
struct node* node =
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(8);
root->right
= newNode(2);
root->left->left
= newNode(3);
if(isSumProperty(root))
printf("The given tree satisfies the children sum property ");
else
printf("The given tree does not satisfy the children sum property ");
getchar();
return 0;
}
TimeComplexity:O(n),wearedoingacompletetraversalofthetree.
=======================================================================
Levelordertraversalinspiralform
Writeafunctiontoprintspiralordertraversalofatree.Forbelowtree,functionshouldprint1,2,3,4,5,6,7.
Method1(Recursive)
Thisproblemcanbeeseenasanextensionofthelevelordertraversalpost.
Toprintthenodesinspiralorder,nodesatdifferentlevelsshouldbeprintedinalternatingorder.An
additionalBooleanvariableltrisusedtochangeprintingorderoflevels.Ifltris1thenprintGivenLevel()
printsnodesfromlefttorightelsefromrighttoleft.Valueofltrisflippedineachiterationtochangethe
order.
Functiontoprintlevelordertraversaloftree
printSpiral(tree)
boolltr=0
ford=1toheight(tree)
printGivenLevel(tree,d,ltr)
ltr~=ltr/*flipltr*/
Functiontoprintallnodesatagivenlevel
printGivenLevel(tree,level,ltr)
iflevelis1,then
print(tree>data)
if(ltr)
printGivenLevel(tree>left,level1,ltr)
printGivenLevel(tree>right,level1,ltr)
else
printGivenLevel(tree>right,level1,ltr)
printGivenLevel(tree>left,level1,ltr)
FollowingisCimplementationofabovealgorithm.
#include <stdio.h>
#include <stdlib.h>
#include <stdbool.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
/* Function protoypes */
void printGivenLevel(struct node* root, int level, int ltr);
/* Function to print spiral traversal of a tree*/

void printSpiral(struct node* root)
{
int i;
/*ltr -> Left to Right. If this variable is set,

then the given level is traverseed from left to right. */
bool ltr = false;
for(i=1; i<=h; i++)
{
printGivenLevel(root, i, ltr);
/*Revert ltr to traverse next level in oppposite order*/

ltr = !ltr;
}
}

void printGivenLevel(struct node* root, int level, int ltr)
{
if(root == NULL)
return;
if(level == 1)
else if (level > 1)
{
if(ltr)
{
printGivenLevel(root->left, level-1, ltr);
printGivenLevel(root->right, level-1, ltr);
}
else
{
printGivenLevel(root->right, level-1, ltr);
printGivenLevel(root->left, level-1, ltr);
}
}
}

{
if (node==NULL)
return 0;
else
{

return(lheight+1);
}
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
printf("Spiral Order traversal of binary tree is \n");
printSpiral(root);
return 0;
}
Output:
SpiralOrdertraversalofbinarytreeis
1234567
TimeComplexity:WorstcasetimecomplexityoftheabovemethodisO(n^2).Worstcaseoccursincaseof
skewedtrees.
Method2(Iterative)
WecanprintspiralordertraversalinO(n)timeandO(n)extraspace.Theideaistousetwostacks.Wecan
useonestackforprintingfromlefttorightandotherstackforprintingfromrighttoleft.Ineveryiteration,we
havenodesofonelevelinoneofthestacks.Weprintthenodes,andpushnodesofnextlevelinother
stack.
// C++ implementation of a O(n) time method for spiral order traversal
#include <iostream>
#include <stack>
// Binary Tree node

struct node
{
int data;
};
void printSpiral(struct node *root)

{
// NULL check
// Create two stacks to store alternate levels

stack<struct node*> s1; // For levels to be printed from right to left
stack<struct node*> s2; // For levels to be printed from left to right
// Push first level to first stack 's1'

s1.push(root);
// Keep ptinting while any of the stacks has some nodes

while (!s1.empty() || !s2.empty())
{
// Print nodes of current level from s1 and push nodes of

// next level to s2
while (!s1.empty())
{
struct node *temp = s1.top();
s1.pop();
cout << temp->data << " ";
// Note that is right is pushed before left

if (temp->right)
s2.push(temp->right);
if (temp->left)
s2.push(temp->left);
}
// Print nodes of current level from s2 and push nodes of

// next level to s1
while (!s2.empty())
{
struct node *temp = s2.top();
s2.pop();
cout << temp->data << " ";
// Note that is left is pushed before right

if (temp->left)
s1.push(temp->left);
if (temp->right)
s1.push(temp->right);
}
}
}
// A utility functiont to create a new node

{
struct node* node = new struct node;
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
cout << "Spiral Order traversal of binary tree is \n";
printSpiral(root);
return 0;
}
Output:
SpiralOrdertraversalofbinarytreeis
1234567
=======================================================================
Aprogramtocheckifabinarytree
isBSTornot
Abinarysearchtree(BST)isanodebasedbinarytreedatastructurewhichhasthefollowingproperties.
Theleftsubtreeofanodecontainsonlynodeswithkeyslessthanthenodeskey.
Therightsubtreeofanodecontainsonlynodeswithkeysgreaterthanthenodeskey.
Boththeleftandrightsubtreesmustalsobebinarysearchtrees.
Fromtheabovepropertiesitnaturallyfollowsthat:
Eachnode(iteminthetree)hasadistinctkey.
METHOD1(SimplebutWrong)
Followingisasimpleprogram.Foreachnode,checkifleftnodeofitissmallerthanthenodeandrightnode
ofitisgreaterthanthenode.
int isBST(struct node* node)
{
if (node == NULL)
return 1;
/* false if left is > than node */

if (node->left != NULL && node->left->data > node->data)
return 0;
/* false if right is < than node */

if (node->right != NULL && node->right->data < node->data)
return 0;
/* false if, recursively, the left or right is not a BST */

if (!isBST(node->left) || !isBST(node->right))
return 0;
/* passing all that, it's a BST */

return 1;
}
Thisapproachiswrongasthiswillreturntrueforbelowbinarytree(andbelowtreeisnotaBST
because4isinleftsubtreeof3)
METHOD2(Correctbutnotefficient)
Foreachnode,checkifmaxvalueinleftsubtreeissmallerthanthenodeandminvalueinrightsubtree
greaterthanthenode.
/* Returns true if a binary tree is a binary search tree */
{
if (node == NULL)
return(true);
/* false if the max of the left is > than us */

if (node->left!=NULL && maxValue(node->left) > node->data)
return(false);
/* false if the min of the right is <= than us */

if (node->right!=NULL && minValue(node->right) < node->data)
return(false);

return(false);
return(true);
}
ItisassumedthatyouhavehelperfunctionsminValue()andmaxValue()thatreturntheminormaxintvalue
fromanonemptytree
METHOD3(CorrectandEfficient)
Method2aboverunsslowlysinceittraversesoversomepartsofthetreemanytimes.Abettersolution
looksateachnodeonlyonce.ThetrickistowriteautilityhelperfunctionisBSTUtil(structnode*node,int
min,intmax)thattraversesdownthetreekeepingtrackofthenarrowingminandmaxallowedvaluesasit
goes,lookingateachnodeonlyonce.TheinitialvaluesforminandmaxshouldbeINT_MINandINT_MAX
theynarrowfromthere.
/*Returnstrueifthegiventreeisabinarysearchtree
(efficientversion).*/
intisBST(structnode*node)
{
return(isBSTUtil(node,INT_MIN,INT_MAX))
}
/*ReturnstrueifthegiventreeisaBSTandits
valuesare>=minand<=max.*/
intisBSTUtil(structnode*node,intmin,intmax)
Implementation:
#include <stdio.h>
#include <stdlib.h>
#include <limits.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
int isBSTUtil(struct node* node, int min, int max);
/* Returns true if the given tree is a binary search tree

{
return(isBSTUtil(node, INT_MIN, INT_MAX));
}
/* Returns true if the given tree is a BST and its

values are >= min and <= max. */
int isBSTUtil(struct node* node, int min, int max)
{
/* an empty tree is BST */

if (node==NULL)
return 1;
/* false if this node violates the min/max constraint */

if (node->data < min || node->data > max)
return 0;
/* otherwise check the subtrees recursively,

tightening the min or max constraint */
return
isBSTUtil(node->left, min, node->data-1) && // Allow only distinct values
isBSTUtil(node->right, node->data+1, max); // Allow only distinct values
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(2);
root->right
= newNode(5);
if(isBST(root))
printf("Is BST");
else
printf("Not a BST");
getchar();
return 0;
}
TimeComplexity:O(n)
AuxiliarySpace:O(1)ifFunctionCallStacksizeisnotconsidered,otherwiseO(n)
METHOD4(UsingInOrderTraversal)
ThankstoLJW489forsuggestingthismethod.
1)DoInOrderTraversalofthegiventreeandstoretheresultinatemparray.
3)Checkifthetemparrayissortedinascendingorder,ifitis,thenthetreeisBST.
TimeComplexity:O(n)
WecanavoidtheuseofAuxiliaryArray.WhiledoingInOrdertraversal,wecankeeptrackofpreviously
visitednode.Ifthevalueofthecurrentlyvisitednodeislessthanthepreviousvalue,thentreeisnotBST.
Thankstoygosforthisspaceoptimization.
bool isBST(struct node* root)
{
static struct node *prev = NULL;
// traverse the tree in inorder fashion and keep track of prev node
if (root)
{
if (!isBST(root->left))
return false;
// Allows only distinct valued nodes

if (prev != NULL && root->data <= prev->data)
return false;
prev = root;
return isBST(root->right);
}
return true;
}
Theuseofstaticvariablecanalsobeavoidedbyusingreferencetoprevnodeasaparameter(Similarto
thispost).
=======================================================================
Programtocountleafnodesina
binarytree
AnodeisaleafnodeifbothleftandrightchildnodesofitareNULL.
Hereisanalgorithmtogettheleafnodecount.
getLeafCount(node)
1)IfnodeisNULLthenreturn0.
2)ElseIfleftandrightchildnodesareNULLreturn1.
3)Elserecursivelycalculateleafcountofthetreeusingbelowformula.
Leafcountofatree=Leafcountofleftsubtree+
Leafcountofrightsubtree

ExampleTree
Leafcountfortheabovetreeis3.
Implementation:
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
/* Function to get the count of leaf nodes in a binary tree*/

unsigned int getLeafCount(struct node* node)
{
if(node == NULL)
return 0;
if(node->left == NULL && node->right==NULL)
return 1;
else
return getLeafCount(node->left)+
getLeafCount(node->right);
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/*Driver program to test above functions*/

int main()
{
/*create a tree*/
root->left
= newNode(2);
root->right
= newNode(3);
/*get leaf count of the above created tree*/

printf("Leaf count of the tree is %d", getLeafCount(root));
getchar();
return 0;
}
Time&SpaceComplexities:Sincethisprogramissimilartotraversaloftree,timeandspacecomplexities
willbesameasTreetraversal(PleaseseeourTreeTraversalpostfordetails)
=======================================================================
LevelOrderTreeTraversal
Levelordertraversalofatreeisbreadthfirsttraversalforthetree.
ExampleTree
Levelordertraversaloftheabovetreeis12345
METHOD1(Usefunctiontoprintagivenlevel)
Algorithm:
Therearebasicallytwofunctionsinthismethod.Oneistoprintallnodesatagivenlevel(printGivenLevel),
andotheristoprintlevelordertraversalofthetree(printLevelorder).printLevelordermakesuseof
printGivenLeveltoprintnodesatalllevelsonebyonestartingfromroot.
/*Functiontoprintlevelordertraversaloftree*/
printLevelorder(tree)
ford=1toheight(tree)
printGivenLevel(tree,d)
/*Functiontoprintallnodesatagivenlevel*/
printGivenLevel(tree,level)
iflevelis1,then
print(tree>data)
printGivenLevel(tree>left,level1)
printGivenLevel(tree>right,level1)
Implementation:
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
void printGivenLevel(struct node* root, int level);
/* Function to print level order traversal a tree*/

void printLevelOrder(struct node* root)
{
int i;
for(i=1; i<=h; i++)
printGivenLevel(root, i);
}

void printGivenLevel(struct node* root, int level)
{
if(root == NULL)
return;
if(level == 1)
else if (level > 1)
{
printGivenLevel(root->left, level-1);
printGivenLevel(root->right, level-1);
}
}

{
if (node==NULL)
return 0;
else
{

return(lheight+1);
}
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
int main()
{
root->left
= newNode(2);
root->right
= newNode(3);

printLevelOrder(root);
getchar();
return 0;
}
TimeComplexity:O(n^2)inworstcase.Foraskewedtree,printGivenLevel()takesO(n)timewherenisthe
numberofnodesintheskewedtree.SotimecomplexityofprintLevelOrder()isO(n)+O(n1)+O(n2)+..+
O(1)whichisO(n^2).
METHOD2(UseQueue)
Algorithm:
Foreachnode,firstthenodeisvisitedandthenitschildnodesareputinaFIFOqueue.
printLevelorder(tree)
1)Createanemptyqueueq
2)temp_node=root/*startfromroot*/
3)Loopwhiletemp_nodeisnotNULL
a)printtemp_node>data.
b)Enqueuetemp_nodeschildren(firstleftthenrightchildren)toq
c)Dequeueanodefromqandassignitsvaluetotemp_node
Implementation:
Hereisasimpleimplementationoftheabovealgorithm.Queueisimplementedusinganarraywith
maximumsizeof500.Wecanimplementqueueaslinkedlistalso.
#include <stdio.h>
#include <stdlib.h>
#define MAX_Q_SIZE 500

struct node
{
int data;
struct node* left;
struct node* right;
};
/* frunction prototypes */
struct node** createQueue(int *, int *);
void enQueue(struct node **, int *, struct node *);
struct node *deQueue(struct node **, int *);
/* Given a binary tree, print its nodes in level order

using array for implementing queue */
void printLevelOrder(struct node* root)
{
int rear, front;
struct node **queue = createQueue(&front, &rear);
struct node *temp_node = root;
while(temp_node)
{
printf("%d ", temp_node->data);
/*Enqueue left child */

if(temp_node->left)
enQueue(queue, &rear, temp_node->left);
/*Enqueue right child */

if(temp_node->right)
enQueue(queue, &rear, temp_node->right);
/*Dequeue node and make it temp_node*/

temp_node = deQueue(queue, &front);
}
}
/*UTILITY FUNCTIONS*/
struct node** createQueue(int *front, int *rear)
{
struct node **queue =
(struct node **)malloc(sizeof(struct node*)*MAX_Q_SIZE);
*front = *rear = 0;
return queue;
}
void enQueue(struct node **queue, int *rear, struct node *new_node)

{
queue[*rear] = new_node;
(*rear)++;
}
struct node *deQueue(struct node **queue, int *front)

{
(*front)++;
return queue[*front - 1];
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(2);
root->right
= newNode(3);

printLevelOrder(root);
getchar();
return 0;
}
TimeComplexity:O(n)wherenisnumberofnodesinthebinarytree
================================================================
=======
Findthenodewithminimumvalue
inaBinarySearchTree
Thisisquitesimple.JusttraversethenodefromroottoleftrecursivelyuntilleftisNULL.Thenodewhose
leftisNULListhenodewithminimumvalue.

Fortheabovetree,westartwith20,thenwemoveleft8,wekeeponmovingtoleftuntilweseeNULL.
Sinceleftof4isNULL,4isthenodewithminimumvalue.
#include <stdio.h>
#include<stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

pointers. */
{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Give a binary search tree and a number,

inserts a new node with the given number in
the correct place in the tree. Returns the new
root pointer which the caller should then use
(the standard trick to avoid using reference
parameters). */
{
single node */
if (node == NULL)
else
{
else

return node;
}
}
/* Given a non-empty binary search tree,

return the minimum data value found in that
tree. Note that the entire tree does not need
to be searched. */
int minValue(struct node* node) {

}
return(current->data);
}
/* Driver program to test sameTree function*/

int main()
{
insert(root, 2);
insert(root, 1);
insert(root, 3);
insert(root, 6);
insert(root, 5);
printf("\n Minimum value in BST is %d", minValue(root));

getchar();
return 0;
}
TimeComplexity:O(n)Worstcasehappensforleftskewedtrees.
Similarlywecangetthemaximumvaluebyrecursivelytraversingtherightnodeofabinarysearchtree.
=======================================================================
TheGreatTreeListRecursion
Problem.
Question:
WritearecursivefunctiontreeToList(Noderoot)thattakesanorderedbinarytreeandrearrangesthe
internalpointerstomakeacirculardoublylinkedlistoutofthetreenodes.Thepreviouspointersshouldbe
storedinthesmallfieldandthenextpointersshouldbestoredinthelargefield.Thelistshouldbe
arrangedsothatthenodesareinincreasingorder.Returntheheadpointertothenewlist.
Thisisverywellexplainedandimplementedathttp://cslibrary.stanford.edu/109/TreeListRecursion.html
=======================================================================
LowestCommonAncestorinaBinarySearchTree
GivenvaluesoftwonodesinaBinarySearchTree,writeacprogramtofindtheLowestCommonAncestor
(LCA).Youmayassumethatboththevaluesexistinthetree.
Thefunctionprototypeshouldbeasfollows:
structnode*lca(node*root,intn1,intn2)
n1andn2aretwogivenvaluesinthetreewithgivenroot.
Forexample,considertheBSTindiagram,LCAof10and14is12andLCAof8and14is8.
Solutions:
IfwearegivenaBSTwhereeverynodehasparentpointer,thenLCAcanbeeasilydeterminedby
traversingupusingparentpointerandprintingthefirstintersectingnode.
WecansolvethisproblemusingBSTproperties.WecanrecursivelytraversetheBSTfromroot.Themain
ideaofthesolutionis,whiletraversingfromtoptobottom,thefirstnodenweencounterwithvaluebetween
n1andn2,i.e.,n1<n<n2orsameasoneofthen1orn2,isLCAofn1andn2(assumingthatn1<n2).So
justrecursivelytraversetheBSTin,ifnode'svalueisgreaterthanbothn1andn2thenourLCAliesinleft
sideofthenode,ifit'sissmallerthanbothn1andn2,thenLCAliesonrightside.OtherwiserootisLCA
(assumingthatbothn1andn2arepresentinBST)
// A recursive C program to find LCA of two nodes n1 and n2.
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
};
/* Function to find LCA of n1 and n2. The function assumes that both
n1 and n2 are present in BST */
struct node *lca(struct node* root, int n1, int n2)
{
// If both n1 and n2 are smaller than root, then LCA lies in left
if (root->data > n1 && root->data > n2)
return lca(root->left, n1, n2);
// If both n1 and n2 are greater than root, then LCA lies in right
if (root->data < n1 && root->data < n2)
return lca(root->right, n1, n2);
return root;
}
/* Helper function that allocates a new node with the given data.*/
{
node->data = data;
return(node);

int main()
{
struct node *root
= newNode(20);
root->left
= newNode(8);
root->right
= newNode(22);
root->left->left
= newNode(4);
root->left->right
= newNode(12);
int n1 = 10, n2 = 14;

struct node *t = lca(root, n1, n2);
printf("LCA of %d and %d is %d \n", n1, n2, t->data);
n1 = 14, n2 = 8;
t = lca(root, n1, n2);
n1 = 10, n2 = 22;
getchar();
return 0;
}
Output:
LCAof10and14is12
LCAof14and8is8
LCAof10and22is20
TimecomplexityofabovesolutionisO(h)wherehisheightoftree.Also,theabovesolutionrequiresO(h)
extraspaceinfunctioncallstackforrecursivefunctioncalls.Wecanavoidextraspaceusingiterative
solution.
{
{
root = root->left;
else if (root->data < n1 && root->data < n2)
root = root->right;
else break;
}
return root;
}
Seethisforcompleteprogram.
YoumayliketoseeLowestCommonAncestorinaBinaryTreealso.
Exercise
Theabovefunctionsassumethatn1andn2bothareinBST.Ifn1andn2arenotpresent,thentheymay
returnincorrectresult.ExtendtheabovesolutionstoreturnNULLifn1orn2orbothnotpresentinBST.
=======================================================================
Givenabinarytree,printoutallof
itsroottoleafpathsoneperline.
Hereisthesolution.
Algorithm:
initialize:pathlen=0,path[1000]
/*1000issomemaxlimitforpaths,itcanchange*/
/*printPathsRecurtraversesnodesoftreeinpreorder*/
printPathsRecur(tree,path[],pathlen)
1)IfnodeisnotNULLthen
a)pushdatatopatharray:
path[pathlen]=node>data.
b)incrementpathlen
pathlen++
2)Ifnodeisaleafnodethenprintthepatharray.
3)Else
a)CallprintPathsRecurforleftsubtree
printPathsRecur(node>left,path,pathLen)
b)CallprintPathsRecurforrightsubtree.
printPathsRecur(node>right,path,pathLen)
Example:
ExampleTree
Outputfortheaboveexamplewillbe
124
125
13
Implementation:
/*program to print all of its root-to-leaf paths for a tree*/
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
void printArray(int [], int);

void printPathsRecur(struct node*, int [], int);
struct node* newNode(int );
void printPaths(struct node*);
/* Given a binary tree, print out all of its root-to-leaf

paths, one per line. Uses a recursive helper to do the work.*/
void printPaths(struct node* node)
{
int path[1000];
printPathsRecur(node, path, 0);
}
/* Recursive helper function -- given a node, and an array containing

the path from the root node up to but not including this node,
print out all the root-leaf paths. */
void printPathsRecur(struct node* node, int path[], int pathLen)
{
/* append this node to the path array */

path[pathLen] = node->data;
pathLen++;
/* it's a leaf, so print the path that led to here */

if (node->left==NULL && node->right==NULL)
{
printArray(path, pathLen);
}
else
{
/* otherwise try both subtrees */
printPathsRecur(node->left, path, pathLen);
printPathsRecur(node->right, path, pathLen);
}
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Utility that prints out an array on a line */

void printArray(int ints[], int len)
{
int i;
for (i=0; i<len; i++) {
printf("%d ", ints[i]);
}
printf("\n");
}

int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
/* Print all root-to-leaf paths of the input tree */

printPaths(root);
getchar();
return 0;
}
=======================================================================
Ifyouaregiventwotraversal
sequences,canyouconstructthe
binarytree?
Itdependsonwhattraversalsaregiven.IfoneofthetraversalmethodsisInorderthenthetreecanbe
constructed,otherwisenot.
Therefore,followingcombinationcanuniquelyidentifyatree.
InorderandPreorder.
InorderandPostorder.
InorderandLevelorder.
Andfollowingdonot.
PostorderandPreorder.
PreorderandLevelorder.
PostorderandLevelorder.
Forexample,Preorder,LevelorderandPostordertraversalsaresameforthetreesgiveninabovediagram.
PreorderTraversal=AB
PostorderTraversal=BA
LevelOrderTraversal=AB
So,evenifthreeofthem(Pre,PostandLevel)aregiven,thetreecannotbeconstructed.
=======================================================================
WriteanEfficientCFunctionto
ConvertaBinaryTreeintoits
MirrorTree
MirrorofaTree:MirrorofaBinaryTreeTisanotherBinaryTreeM(T)withleftandrightchildrenofall
nonleafnodesinterchanged.
Treesinthebelowfigurearemirrorofeachother
AlgorithmMirror(tree):
(1)CallMirrorforleftsubtreei.e.,Mirror(leftsubtree)
(2)CallMirrorforrightsubtreei.e.,Mirror(leftsubtree)
(3)Swapleftandrightsubtrees.
temp=leftsubtree
leftsubtree=rightsubtree
rightsubtree=temp
Program:
#include<stdio.h>
#include<stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Change a tree so that the roles of the left and

right pointers are swapped at every node.
So the tree...
4
/\
2
/\
1
is changed to...
4
/\
5
2
/\
3
*/
void mirror(struct node* node)
{
if (node==NULL)
return;
else
{
struct node* temp;
mirror(node->left);
mirror(node->right);
/* swap the pointers in this node */

temp
= node->left;
node->left = node->right;
node->right = temp;
}
}
/* Helper function to test mirror(). Given a binary

search tree, print out its data elements in
increasing sorted order.*/
void inOrder(struct node* node)
{
if (node == NULL)
return;
inOrder(node->left);
inOrder(node->right);
}

int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
/* Print inorder traversal of the input tree */

inOrder(root);
/* Convert tree to its mirror */

mirror(root);
/* Print inorder traversal of the mirror tree */

printf("\n Inorder traversal of the mirror tree is \n");
inOrder(root);
getchar();
return 0;
}
Time&SpaceComplexities:Thisprogramissimilartotraversaloftreespaceandtimecomplexitieswillbe
sameasTreetraversal(PleaseseeourTreeTraversalpostfordetails)
=======================================================================
WriteaCprogramtoDeleteaTree.
Todeleteatreewemusttraverseallthenodesofthetreeanddeletethemonebyone.Sowhichtraversal
weshoulduseInorderorPreorderorPostorder.AnswerissimplePostorder,becausebeforedeleting
theparentnodeweshoulddeleteitschildrennodesfirst
Wecandeletetreewithothertraversalsalsowithextraspacecomplexitybutwhyshouldwegoforother
traversalsifwehavePostorderavailablewhichdoestheworkwithoutstoringanythinginsametime
complexity.
Forthefollowingtreenodesaredeletedinorder4,5,2,3,1
ExampleTree
Program
#include<stdio.h>
#include<stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* This function traverses tree in post order to

to delete each and every node of the tree */
void deleteTree(struct node* node)
{
/* first delete both subtrees */

deleteTree(node->left);
deleteTree(node->right);
/* then delete the node */

printf("\n Deleting node: %d", node->data);
free(node);
}
/* Driver program to test deleteTree function*/

int main()
{

root->left
= newNode(2);
root->right
= newNode(3);
root->left->left
= newNode(4);
root->left->right
= newNode(5);
deleteTree(root);
root = NULL;
printf("\n Tree deleted ");
getchar();
return 0;
}
TheabovedeleteTree()functiondeletesthetree,butdoesntchangeroottoNULLwhichmaycause
problemsiftheuserofdeleteTree()doesntchangeroottoNULLandtirestoaccessvaluesusingroot
pointer.WecanmodifythedeleteTree()functiontotakereferencetotherootnodesothatthisproblem
doesntoccur.Seethefollowingcode.
#include<stdio.h>
#include<stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* This function is same as deleteTree() in the previous program */

void _deleteTree(struct node* node)
{
/* first delete both subtrees */

_deleteTree(node->left);
_deleteTree(node->right);
/* then delete the node */

printf("\n Deleting node: %d", node->data);
free(node);
}
/* Deletes a tree and sets the root as NULL */

void deleteTree(struct node** node_ref)
{
_deleteTree(*node_ref);
*node_ref = NULL;
}
/* Driver program to test deleteTree function*/

int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
root->left->left
= newNode(4);
root->left->right
= newNode(5);
// Note that we pass the address of root here
deleteTree(&root);
printf("\n Tree deleted ");
getchar();
return 0;
}
TimeComplexity:O(n)
SpaceComplexity:IfwedontconsidersizeofstackforfunctioncallsthenO(1)otherwiseO(n)
=======================================================================
WriteaCProgramtoFindthe
MaximumDepthorHeightofaTree
Maximumdepthorheightofthebelowtreeis3.
ExampleTree
Recursivelycalculateheightofleftandrightsubtreesofanodeandassignheighttothenodeasmaxofthe
heightsoftwochildrenplus1.Seebelowpseudocodeandprogramfordetails.
Algorithm:
maxDepth()
1.Iftreeisemptythenreturn0
2.Else
(a)Getthemaxdepthofleftsubtreerecursivelyi.e.,
callmaxDepth(tree>leftsubtree)
(a)Getthemaxdepthofrightsubtreerecursivelyi.e.,
callmaxDepth(tree>rightsubtree)
(c)Getthemaxofmaxdepthsofleftandright
subtreesandadd1toitforthecurrentnode.
max_depth=max(maxdeptofleftsubtree,
maxdepthofrightsubtree)
+1
(d)Returnmax_depth
SeethebelowdiagramformoreclarityaboutexecutionoftherecursivefunctionmaxDepth()for
aboveexampletree.
maxDepth('1')=max(maxDepth('2'),maxDepth('3'))+1
=2+1
/\
/\
/\
/\
/\
maxDepth('1')maxDepth('3')=1
=max(maxDepth('4'),maxDepth('5'))+1
=1+1=2
/\
/\
/\
/\
/\
maxDepth('4')=1maxDepth('5')=1
Implementation:
#include<stdio.h>
#include<stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
/* Compute the "maxDepth" of a tree -- the number of

int maxDepth(struct node* node)
{
if (node==NULL)
return 0;
else
{
/* compute the depth of each subtree */
int lDepth = maxDepth(node->left);
int rDepth = maxDepth(node->right);

if (lDepth > rDepth)
return(lDepth+1);
else return(rDepth+1);
}
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
int main()
{
printf("Hight of tree is %d", maxDepth(root));
getchar();
return 0;
}
TimeComplexity:O(n)(PleaseseeourpostTreeTraversalfordetails)
=======================================================================
WriteCCodetoDetermineifTwo
TreesareIdentical
Twotreesareidenticalwhentheyhavesamedataandarrangementofdataisalsosame.
Toidentifyiftwotreesareidentical,weneedtotraversebothtreessimultaneously,andwhiletraversingwe
needtocomparedataandchildrenofthetrees.
Algorithm:
sameTree(tree1,tree2)
1.Ifbothtreesareemptythenreturn1.
2.ElseIfbothtreesarenonempty
(a)Checkdataoftherootnodes(tree1>data==tree2>data)
(b)Checkleftsubtreesrecursivelyi.e.,callsameTree(
tree1>left_subtree,tree2>left_subtree)
(c)Checkrightsubtreesrecursivelyi.e.,callsameTree(
tree1>right_subtree,tree2>right_subtree)
(d)Ifa,bandcaretruethenreturn1.
3Elsereturn0(oneisemptyandotherisnot)
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Given two trees, return true if they are

structurally identical */
int identicalTrees(struct node* a, struct node* b)
{
/*1. both empty */
if (a==NULL && b==NULL)
return 1;
/* 2. both non-empty -> compare them */

if (a!=NULL && b!=NULL)
{
return
(
a->data == b->data &&
identicalTrees(a->left, b->left) &&
identicalTrees(a->right, b->right)
);
}
/* 3. one empty, one not -> false */

return 0;
}
/* Driver program to test identicalTrees function*/

int main()
{
root1->left->left = newNode(4);
root2->left->left = newNode(4);
if(identicalTrees(root1, root2))
printf("Both tree are identical.");
else
printf("Trees are not identical.");
getchar();
return 0;
}
TimeComplexity:
ComplexityoftheidenticalTree()willbeaccordingtothetreewithlessernumberofnodes.Letnumberof
nodesintwotreesbemandnthencomplexityofsameTree()isO(m)wherem<n.
=======================================================================
WriteaCprogramtoCalculateSize
ofatree
Sizeofatreeisthenumberofelementspresentinthetree.Sizeofthebelowtreeis5.
ExampleTree
Size()functionrecursivelycalculatesthesizeofatree.Itworksasfollows:
Sizeofatree=Sizeofleftsubtree+1+Sizeofrightsubtree
Algorithm:
size(tree)
1.Iftreeisemptythenreturn0
2.Else
(a)Getthesizeofleftsubtreerecursivelyi.e.,call
size(tree>leftsubtree)
(a)Getthesizeofrightsubtreerecursivelyi.e.,call
size(tree>rightsubtree)
(c)Calculatesizeofthetreeasfollowing:
tree_size=size(leftsubtree)+size(right
subtree)+1
(d)Returntree_size
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Computes the number of nodes in a tree. */

int size(struct node* node)
{
if (node==NULL)
return 0;
else
return(size(node->left) + 1 + size(node->right));
}
/* Driver program to test size function*/

int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
printf("Size of the tree is %d", size(root));

getchar();
return 0;
}
Time&SpaceComplexities:Sincethisprogramissimilartotraversaloftree,timeandspacecomplexities
willbesameasTreetraversal(PleaseseeourTreeTraversalpostfordetails)
=======================================================================
TreeTraversals
Unlikelineardatastructures(Array,LinkedList,Queues,Stacks,etc)whichhaveonlyonelogicalwayto
traversethem,treescanbetraversedindifferentways.Followingarethegenerallyusedwaysfortraversing
trees.
ExampleTree
DepthFirstTraversals:
(a)Inorder
(b)Preorder
(c)Postorder
BreadthFirstorLevelOrderTraversal
PleaseseethispostforBreadthFirstTraversal.
InorderTraversal:
AlgorithmInorder(tree)
1.Traversetheleftsubtree,i.e.,callInorder(leftsubtree)
2.Visittheroot.
3.Traversetherightsubtree,i.e.,callInorder(rightsubtree)
UsesofInorder
Incaseofbinarysearchtrees(BST),Inordertraversalgivesnodesinnondecreasingorder.Togetnodesof
BSTinnonincreasingorder,avariationofInordertraversalwhereInorderitraversalsreversed,canbe
used.
Example:Inordertraversalfortheabovegivenfigureis42513.
PreorderTraversal:
AlgorithmPreorder(tree)
1.Visittheroot.
2.Traversetheleftsubtree,i.e.,callPreorder(leftsubtree)
3.Traversetherightsubtree,i.e.,callPreorder(rightsubtree)
UsesofPreorder
Preordertraversalisusedtocreateacopyofthetree.Preordertraversalisalsousedtogetprefix
expressiononofanexpressiontree.Pleaseseehttp://en.wikipedia.org/wiki/Polish_notationtoknowwhy
prefixexpressionsareuseful.
Example:Preordertraversalfortheabovegivenfigureis12453.
PostorderTraversal:
AlgorithmPostorder(tree)
1.Traversetheleftsubtree,i.e.,callPostorder(leftsubtree)
2.Traversetherightsubtree,i.e.,callPostorder(rightsubtree)
3.Visittheroot.
UsesofPostorder
Postordertraversalisusedtodeletethetree.Pleaseseethequestionfordeletionoftreefordetails.
Postordertraversalisalsousefultogetthepostfixexpressionofanexpressiontree.Pleasesee
http://en.wikipedia.org/wiki/Reverse_Polish_notationtofortheusageofpostfixexpression.
Example:Postordertraversalfortheabovegivenfigureis45231.
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Given a binary tree, print its nodes according to the

"bottom-up" postorder traversal. */
void printPostorder(struct node* node)
{
if (node == NULL)
return;
// first recur on left subtree
printPostorder(node->left);
// then recur on right subtree

printPostorder(node->right);
// now deal with the node

}

{
if (node == NULL)
return;



}

void printPreorder(struct node* node)
{
if (node == NULL)
return;
/* first print data of node */

/* then recur on left sutree */

printPreorder(node->left);

printPreorder(node->right);
}

int main()
{
root->left
= newNode(2);
root->right
= newNode(3);
root->left->left
= newNode(4);
root->left->right
= newNode(5);
printf("\n Preorder traversal of binary tree is \n");

printPreorder(root);
printf("\n Inorder traversal of binary tree is \n");

printInorder(root);
printf("\n Postorder traversal of binary tree is \n");

printPostorder(root);
getchar();
return 0;
}
TimeComplexity:O(n)
Letusproveit:
ComplexityfunctionT(n)forallproblemwheretreetraversalisinvolvedcanbedefinedas:
T(n)=T(k)+T(nk1)+c
Wherekisthenumberofnodesononesideofrootandnk1ontheotherside.
Letsdoanalysisofboundaryconditions
Case1:Skewedtree(Oneofthesubtreesisemptyandothersubtreeisnonempty)
kis0inthiscase.
T(n)=T(0)+T(n1)+c
T(n)=2T(0)+T(n2)+2c
T(n)=3T(0)+T(n3)+3c
T(n)=4T(0)+T(n4)+4c
.
T(n)=(n1)T(0)+T(1)+(n1)c
T(n)=nT(0)+(n)c
ValueofT(0)willbesomeconstantsayd.(traversingaemptytreewilltakesomeconstantstime)
T(n)=n(c+d)
T(n)=()(n)(Thetaofn)
Case2:Bothleftandrightsubtreeshaveequalnumberofnodes.
T(n)=2T(|_n/2_|)+c
Thisrecursivefunctionisinthestandardform(T(n)=aT(n/b)+()(n))formaster
methodhttp://en.wikipedia.org/wiki/Master_theorem.Ifwesolveitbymastermethodweget()(n)
AuxiliarySpace:IfwedontconsidersizeofstackforfunctioncallsthenO(1)otherwiseO(n).
=======================================================================
InorderSuccessorinBinarySearch
Tree
InBinaryTree,InordersuccessorofanodeisthenextnodeinInordertraversaloftheBinaryTree.Inorder
SuccessorisNULLforthelastnodeinInoordertraversal.
InBinarySearchTree,InorderSuccessorofaninputnodecanalsobedefinedasthenodewiththesmallest
keygreaterthanthekeyofinputnode.So,itissometimesimportanttofindnextnodeinsortedorder.

Intheabovediagram,inordersuccessorof8is10,inordersuccessorof10is12andinordersuccessorof
14is20.
Method1(UsesParentPointer)
Inthismethod,weassumethateverynodehasparentpointer.
TheAlgorithmisdividedintotwocasesonthebasisofrightsubtreeoftheinputnodebeingemptyornot.
2)IfrightsbtreeofnodeisNULL,thensuccisoneoftheancestors.Dofollowing.
Travelupusingtheparentpointeruntilyouseeanodewhichisleftchildofitsparent.Theparentofsucha
nodeisthesucc.
Implementation
NotethatthefunctiontofindInOrderSuccessorishighlighted(withgraybackground)inbelowcode.
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
struct node* parent;

};
struct node * minValue(struct node* node);

{

struct node *p = n->parent;
while(p != NULL && n == p->right)
{
n = p;
p = p->parent;
}
return p;
}
/* Given a non-empty binary search tree, return the minimum data

value found in that tree. Note that the entire tree does not need
to be searched. */
struct node * minValue(struct node* node) {

}
return current;
}
{
node->data
= data;
node->left
= NULL;
node->right = NULL;
node->parent = NULL;
return(node);
}
/* Give a binary search tree and a number, inserts a new node with
the given number in the correct place in the tree. Returns the new
root pointer which the caller should then use (the standard trick to
avoid using reference parameters). */
{
single node */
if (node == NULL)
else
{
struct node *temp;

{
temp = insert(node->left, data);
node->left = temp;
temp->parent= node;
}
else
{
temp = insert(node->right, data);
node->right = temp;
temp->parent = node;

return node;
}
}

int main()
{
struct node* root = NULL, *temp, *succ, *min;
//creating the tree given in the above diagram

temp = root->left->right->right;
succ = inOrderSuccessor(root, temp);

if(succ != NULL)
printf("\n Inorder Successor of %d is %d ", temp->data, succ->data);
else
printf("\n Inorder Successor doesn't exit");
getchar();
return 0;
}
Outputoftheaboveprogram:
InorderSuccessorof14is20
Method2(Searchfromroot)
ParentpointerisNOTneededinthisalgorithm.TheAlgorithmisdividedintotwocasesonthebasisofright
subtreeoftheinputnodebeingemptyornot.
2)IfrightsbtreeofnodeisNULL,thenstartfromrootandussearchliketechnique.Dofollowing.
Traveldownthetree,ifanodesdataisgreaterthanrootsdatathengorightside,otherwisegotoleftside.
{
struct node *succ = NULL;
// Start from root and search for successor down the tree
{
if (n->data < root->data)
{
succ = root;
root = root->left;
}
else if (n->data > root->data)
root = root->right;
else
break;
}
return succ;
}
=======================================================================
BinaryTreetoBinarySearchTree
Conversion
GivenaBinaryTree,convertittoaBinarySearchTree.Theconversionmustbedoneinsuchawaythat
keepstheoriginalstructureofBinaryTree.
Examples.
Example1
Input:
10
/\
27
/\
84
Output:
8
/\
410
/\
27
Example2
Input:
10
/\
3015
/\
205
Output:
15
/\
1020
/\
530
Solution
Followingisa3stepsolutionforconvertingBinarytreetoBinarySearchTree.
1)Createatemparrayarr[]thatstoresinordertraversalofthetree.ThissteptakesO(n)time.
2)Sortthetemparrayarr[].Timecomplexityofthisstepdependsuponthesortingalgorithm.Inthefollowing
implementation,QuickSortisusedwhichtakes(n^2)time.ThiscanbedoneinO(nLogn)timeusingHeap
SortorMergeSort.
3)Againdoinordertraversaloftreeandcopyarrayelementstotreenodesonebyone.ThissteptakesO(n)
time.
FollowingisCimplementationoftheaboveapproach.Themainfunctiontoconvertishighlightedinthe
followingcode.
/* A program to convert Binary Tree to Binary Search Tree */
#include<stdio.h>
#include<stdlib.h>
/* A binary tree node structure */

struct node
{
int data;
struct node *left;
struct node *right;
};
/* A helper function that stores inorder traversal of a tree rooted

with node */
void storeInorder (struct node* node, int inorder[], int *index_ptr)
{
// Base Case
if (node == NULL)
return;
/* first store the left subtree */

storeInorder (node->left, inorder, index_ptr);
/* Copy the root's data */

/* finally store the right subtree */

storeInorder (node->right, inorder, index_ptr);
}
/* A helper function to count nodes in a Binary Tree */

int countNodes (struct node* root)
{
if (root == NULL)
return 0;
return countNodes (root->left) +
countNodes (root->right) + 1;
}
// Following function is needed for library function qsort()

int compare (const void * a, const void * b)
{
return ( *(int*)a - *(int*)b );
}
/* A helper function that copies contents of arr[] to Binary Tree.

This functon basically does Inorder traversal of Binary Tree and
one by one copy arr[] elements to Binary Tree nodes */
void arrayToBST (int *arr, struct node* root, int *index_ptr)
{
// Base Case
if (root == NULL)
return;
/* first update the left subtree */

arrayToBST (arr, root->left, index_ptr);
/* Now update root's data and increment index */

root->data = arr[*index_ptr];
(*index_ptr)++;
/* finally update the right subtree */

arrayToBST (arr, root->right, index_ptr);
// This function converts a given Binary Tree to BST

void binaryTreeToBST (struct node *root)
{
// base case: tree is empty
if(root == NULL)
return;
/* Count the number of nodes in Binary Tree so that

we know the size of temporary array to be created */
int n = countNodes (root);
// Create a temp array arr[] and store inorder traversal of tree in arr[]
int *arr = new int[n];
int i = 0;
storeInorder (root, arr, &i);
// Sort the array using library function for quick sort

qsort (arr, n, sizeof(arr[0]), compare);
// Copy array elements back to Binary Tree

i = 0;
arrayToBST (arr, root, &i);
// delete dynamically allocated memory to avoid meory leak

delete [] arr;
}

{
temp->data = data;
temp->left = NULL;
temp->right = NULL;
return temp;
/* Utility function to print inorder traversal of Binary Tree */

{
if (node == NULL)
return;



}

int main()
{
/* Constructing tree given in the above figure

10
/ \
30
15
20
*/
root = newNode(10);
// convert Binary Tree to BST

binaryTreeToBST (root);
printf("Following is Inorder Traversal of the converted BST: \n");

printInorder (root);
return 0;
}
Output:
FollowingisInorderTraversaloftheconvertedBST:
510152030
WewillbecoveringanothermethodforthisproblemwhichconvertsthetreeusingO(heightoftree)extra
space.
=======================================================================
Aprogramtocheckifabinarytree
isBSTornot
Abinarysearchtree(BST)isanodebasedbinarytreedatastructurewhichhasthefollowingproperties.
Theleftsubtreeofanodecontainsonlynodeswithkeyslessthanthenodeskey.
Therightsubtreeofanodecontainsonlynodeswithkeysgreaterthanthenodeskey.
Boththeleftandrightsubtreesmustalsobebinarysearchtrees.
Fromtheabovepropertiesitnaturallyfollowsthat:
Eachnode(iteminthetree)hasadistinctkey.
METHOD1(SimplebutWrong)
Followingisasimpleprogram.Foreachnode,checkifleftnodeofitissmallerthanthenodeandrightnode
ofitisgreaterthanthenode.
{
if (node == NULL)
return 1;
/* false if left is > than node */

if (node->left != NULL && node->left->data > node->data)
return 0;
/* false if right is < than node */

if (node->right != NULL && node->right->data < node->data)
return 0;

return 0;

return 1;
}
Thisapproachiswrongasthiswillreturntrueforbelowbinarytree(andbelowtreeisnotaBST
because4isinleftsubtreeof3)
METHOD2(Correctbutnotefficient)
Foreachnode,checkifmaxvalueinleftsubtreeissmallerthanthenodeandminvalueinrightsubtree
greaterthanthenode.
/* Returns true if a binary tree is a binary search tree */
{
if (node == NULL)
return(true);
/* false if the max of the left is > than us */

if (node->left!=NULL && maxValue(node->left) > node->data)
return(false);
/* false if the min of the right is <= than us */

if (node->right!=NULL && minValue(node->right) < node->data)
return(false);

return(false);

return(true);
}
ItisassumedthatyouhavehelperfunctionsminValue()andmaxValue()thatreturntheminormaxintvalue
fromanonemptytree
METHOD3(CorrectandEfficient)
Method2aboverunsslowlysinceittraversesoversomepartsofthetreemanytimes.Abettersolution
looksateachnodeonlyonce.ThetrickistowriteautilityhelperfunctionisBSTUtil(structnode*node,int
min,intmax)thattraversesdownthetreekeepingtrackofthenarrowingminandmaxallowedvaluesasit
goes,lookingateachnodeonlyonce.TheinitialvaluesforminandmaxshouldbeINT_MINandINT_MAX
theynarrowfromthere.
/*Returnstrueifthegiventreeisabinarysearchtree
(efficientversion).*/
intisBST(structnode*node)
{
return(isBSTUtil(node,INT_MIN,INT_MAX))
}
/*ReturnstrueifthegiventreeisaBSTandits
valuesare>=minand<=max.*/
intisBSTUtil(structnode*node,intmin,intmax)
Implementation:
#include <stdio.h>
#include <stdlib.h>
#include <limits.h>

struct node
{
int data;
struct node* left;
struct node* right;
};
int isBSTUtil(struct node* node, int min, int max);
/* Returns true if the given tree is a binary search tree

{
return(isBSTUtil(node, INT_MIN, INT_MAX));
}
/* Returns true if the given tree is a BST and its

values are >= min and <= max. */
int isBSTUtil(struct node* node, int min, int max)
/* an empty tree is BST */

if (node==NULL)
return 1;
/* false if this node violates the min/max constraint */

if (node->data < min || node->data > max)
return 0;
/* otherwise check the subtrees recursively,

tightening the min or max constraint */
return
isBSTUtil(node->left, min, node->data-1) && // Allow only distinct values
isBSTUtil(node->right, node->data+1, max); // Allow only distinct values
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}

int main()
{
root->left
= newNode(2);
root->right
= newNode(5);
if(isBST(root))
printf("Is BST");
else
printf("Not a BST");
getchar();
return 0;
}
TimeComplexity:O(n)
AuxiliarySpace:O(1)ifFunctionCallStacksizeisnotconsidered,otherwiseO(n)
METHOD4(UsingInOrderTraversal)
ThankstoLJW489forsuggestingthismethod.
1)DoInOrderTraversalofthegiventreeandstoretheresultinatemparray.
3)Checkifthetemparrayissortedinascendingorder,ifitis,thenthetreeisBST.
TimeComplexity:O(n)
WecanavoidtheuseofAuxiliaryArray.WhiledoingInOrdertraversal,wecankeeptrackofpreviously
visitednode.Ifthevalueofthecurrentlyvisitednodeislessthanthepreviousvalue,thentreeisnotBST.
Thankstoygosforthisspaceoptimization.
bool isBST(struct node* root)
{
static struct node *prev = NULL;
// traverse the tree in inorder fashion and keep track of prev node
if (root)
{
if (!isBST(root->left))
return false;
// Allows only distinct valued nodes

if (prev != NULL && root->data <= prev->data)
return false;
prev = root;
return isBST(root->right);
return true;
}
=======================================================================
LowestCommonAncestorina
BinarySearchTree.
GivenvaluesoftwonodesinaBinarySearchTree,writeacprogramtofindtheLowestCommonAncestor
(LCA).Youmayassumethatboththevaluesexistinthetree.
Thefunctionprototypeshouldbeasfollows:
structnode*lca(node*root,intn1,intn2)
n1andn2aretwogivenvaluesinthetreewithgivenroot.
Forexample,considertheBSTindiagram,LCAof10and14is12andLCAof8and14is8.
Solutions:
IfwearegivenaBSTwhereeverynodehasparentpointer,thenLCAcanbeeasilydeterminedby
traversingupusingparentpointerandprintingthefirstintersectingnode.
WecansolvethisproblemusingBSTproperties.WecanrecursivelytraversetheBSTfromroot.Themain
ideaofthesolutionis,whiletraversingfromtoptobottom,thefirstnodenweencounterwithvaluebetween
n1andn2,i.e.,n1<n<n2orsameasoneofthen1orn2,isLCAofn1andn2(assumingthatn1<n2).So
justrecursivelytraversetheBSTin,ifnode'svalueisgreaterthanbothn1andn2thenourLCAliesinleft
sideofthenode,ifit'sissmallerthanbothn1andn2,thenLCAliesonrightside.OtherwiserootisLCA
(assumingthatbothn1andn2arepresentinBST)
// A recursive C program to find LCA of two nodes n1 and n2.
#include <stdio.h>
#include <stdlib.h>
struct node
{
int data;
};
{
return lca(root->left, n1, n2);
if (root->data < n1 && root->data < n2)
return lca(root->right, n1, n2);
return root;
}
/* Helper function that allocates a new node with the given data.*/

{
node->data = data;
return(node);
}

int main()
{
struct node *root
= newNode(20);
root->left
= newNode(8);
root->right
= newNode(22);
root->left->left
= newNode(4);
root->left->right
= newNode(12);
int n1 = 10, n2 = 14;

struct node *t = lca(root, n1, n2);
n1 = 14, n2 = 8;
n1 = 10, n2 = 22;
getchar();
return 0;
}
Output:
LCAof10and14is12
LCAof14and8is8
LCAof10and22is20
TimecomplexityofabovesolutionisO(h)wherehisheightoftree.Also,theabovesolutionrequiresO(h)
extraspaceinfunctioncallstackforrecursivefunctioncalls.Wecanavoidextraspaceusingiterative
solution.
{
{
root = root->left;
else if (root->data < n1 && root->data < n2)
root = root->right;
else break;
}
return root;
}
=======================================================================
MergeTwoBalancedBinarySearch
Trees
Youaregiventwobalancedbinarysearchtreese.g.,AVLorRedBlackTree.Writeafunctionthatmerges
thetwogivenbalancedBSTsintoabalancedbinarysearchtree.Lettherebemelementsinfirsttreeandn
elementsintheothertree.YourmergefunctionshouldtakeO(m+n)time.
Inthefollowingsolutions,itisassumedthatsizesoftreesarealsogivenasinput.Ifthesizeisnotgiven,
thenwecangetthesizebytraversingthetree(Seethis).
Method1(Insertelementsoffirsttreetosecond)
TakeallelementsoffirstBSTonebyone,andinsertthemintothesecondBST.Insertinganelementtoa
selfbalancingBSTtakesLogntime(Seethis)wherenissizeoftheBST.Sotimecomplexityofthismethod
isLog(n)+Log(n+1)Log(m+n1).ThevalueofthisexpressionwillbebetweenmLognandmLog(m+n1).
Asanoptimization,wecanpickthesmallertreeasfirsttree.
Method2(MergeInorderTraversals)
1)Doinordertraversaloffirsttreeandstorethetraversalinonetemparrayarr1[].ThissteptakesO(m)
time.
2)Doinordertraversalofsecondtreeandstorethetraversalinanothertemparrayarr2[].Thissteptakes
O(n)time.
3)Thearrayscreatedinstep1and2aresortedarrays.Mergethetwosortedarraysintoonearrayofsizem
+n.ThissteptakesO(m+n)time.
4)Constructabalancedtreefromthemergedarrayusingthetechniquediscussedinthispost.Thisstep
takesO(m+n)time.
TimecomplexityofthismethodisO(m+n)whichisbetterthanmethod1.ThismethodtakesO(m+n)time
eveniftheinputBSTsarenotbalanced.
FollowingisC++implementationofthismethod.
#include <stdio.h>
#include <stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

int *merge(int arr1[], int arr2[], int m, int n);
// A helper function that stores inorder traversal of a tree in inorder array

void storeInorder(struct node* node, int inorder[], int *index_ptr);
struct node* sortedArrayToBST(int arr[], int start, int end);
/* This function merges two balanced BSTs with roots as root1 and root2.
m and n are the sizes of the trees respectively */
struct node* mergeTrees(struct node *root1, struct node *root2, int m, int n)
{
// Store inorder traversal of first tree in an array arr1[]
int *arr1 = new int[m];
int i = 0;
storeInorder(root1, arr1, &i);
// Store inorder traversal of second tree in another array arr2[]

int *arr2 = new int[n];
int j = 0;
storeInorder(root2, arr2, &j);
// Merge the two sorted array into one

int *mergedArr = merge(arr1, arr2, m, n);
// Construct a tree from the merged array and return root of the tree
return sortedArrayToBST (mergedArr, 0, m+n-1);
}

{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
// A utility function to print inorder traversal of a given binary tree

{
if (node == NULL)
return;


}

int *merge(int arr1[], int arr2[], int m, int n)
{
// mergedArr[] is going to contain result
int *mergedArr = new int[m + n];
int i = 0, j = 0, k = 0;
// Traverse through both arrays

while (i < m && j < n)
{
// Pick the smaler element and put it in mergedArr
if (arr1[i] < arr2[j])
{
i++;
}
else
{
j++;
}
k++;
}
// If there are more elements in first array

while (i < m)
{
i++; k++;
}
// If there are more elements in second array

while (j < n)
{
j++; k++;
}
return mergedArr;
}
// A helper function that stores inorder traversal of a tree rooted with node
void storeInorder(struct node* node, int inorder[], int *index_ptr)
{
if (node == NULL)
return;

storeInorder(node->left, inorder, index_ptr);

storeInorder(node->right, inorder, index_ptr);
}
struct node* sortedArrayToBST(int arr[], int start, int end)
{
/* Base Case */
if (start > end)
return NULL;

struct node *root = newNode(arr[mid]);


return root;
}

int main()
{
/* Create following tree as first balanced BST
100
/ \
50
300
/\
20 70
*/
root1->left
= newNode(50);
root1->right
= newNode(300);
root1->left->left
= newNode(20);
/* Create following tree as second balanced BST
80
/ \
40
120
*/
root2->left
= newNode(40);
root2->right
= newNode(120);
struct node *mergedTree = mergeTrees(root1, root2, 5, 3);
printf ("Following is Inorder traversal of the merged tree \n");

printInorder(mergedTree);
getchar();
return 0;
}
Output:
FollowingisInordertraversalofthemergedtree
2040507080100120300
Method3(InPlaceMergeusingDLL)
WecanuseaDoublyLinkedListtomergetreesinplace.Followingarethesteps.
1)ConvertthegiventwoBinarySearchTreesintodoublylinkedlistinplace(Referthispostforthisstep).
2)MergethetwosortedLinkedLists(Referthispostforthisstep).
3)BuildaBalancedBinarySearchTreefromthemergedlistcreatedinstep2.(Referthispostforthisstep)
TimecomplexityofthismethodisalsoO(m+n)andthismethoddoesconversioninplace.
=======================================================================
Findthenodewithminimumvalue
inaBinarySearchTree
Thisisquitesimple.JusttraversethenodefromroottoleftrecursivelyuntilleftisNULL.Thenodewhose
leftisNULListhenodewithminimumvalue.
Fortheabovetree,westartwith20,thenwemoveleft8,wekeeponmovingtoleftuntilweseeNULL.
Sinceleftof4isNULL,4isthenodewithminimumvalue.
#include <stdio.h>
#include<stdlib.h>

struct node
{
int data;
struct node* left;
struct node* right;
};

pointers. */
{
node->data = data;
node->left = NULL;
node->right = NULL;
return(node);
}
/* Give a binary search tree and a number,

inserts a new node with the given number in
the correct place in the tree. Returns the new
root pointer which the caller should then use
(the standard trick to avoid using reference
parameters). */
{
single node */
if (node == NULL)
else
{
else

return node;
}
}
/* Given a non-empty binary search tree,

return the minimum data value found in that
tree. Note that the entire tree does not need
to be searched. */
int minValue(struct node* node) {

}
return(current->data);
}
/* Driver program to test sameTree function*/

int main()
{
insert(root, 2);
insert(root, 1);
insert(root, 3);
insert(root, 6);
insert(root, 5);
printf("\n Minimum value in BST is %d", minValue(root));

getchar();
return 0;
}
TimeComplexity:O(n)Worstcasehappensforleftskewedtrees.
=======================================================================
DynamicProgramming|Set24
(OptimalBinarySearchTree)
Givenasortedarraykeys[0..n1]ofsearchkeysandanarrayfreq[0..n1]offrequencycounts,wherefreq[i]
isthenumberofsearchestokeys[i].Constructabinarysearchtreeofallkeyssuchthatthetotalcostofall
thesearchesisassmallaspossible.
LetusfirstdefinethecostofaBST.ThecostofaBSTnodeislevelofthatnodemultipliedbyitsfrequency.
Levelofrootis1.
Example1
Input:keys[]={10,12},freq[]={34,50}
TherecanbefollowingtwopossibleBSTs
1012
\/
1210
III
Frequencyofsearchesof10and12are34and50respectively.
ThecostoftreeIis34*1+50*2=134
ThecostoftreeIIis50*1+34*2=118
Example2
Input:keys[]={10,12,20},freq[]={34,8,50}
TherecanbefollowingpossibleBSTs
1012201020
\/\/\/
121020122010
\//\
20101212
IIIIIIIVV
AmongallpossibleBSTs,costofthefifthBSTisminimum.
CostofthefifthBSTis1*50+2*34+3*8=142
1)OptimalSubstructure:
Theoptimalcostforfreq[i..j]canberecursivelycalculatedusingfollowingformula.

WeneedtocalculateoptCost(0,n1)tofindtheresult.
Theideaofaboveformulaissimple,weonebyonetryallnodesasroot(rvariesfromitojinsecondterm).
Whenwemakerthnodeasroot,werecursivelycalculateoptimalcostfromitor1andr+1toj.
Weaddsumoffrequenciesfromitoj(seefirsttermintheaboveformula),thisisaddedbecauseevery
searchwillgothroughrootandonecomparisonwillbedoneforeverysearch.
2)OverlappingSubproblems
Followingisrecursiveimplementationthatsimplyfollowstherecursivestructurementionedabove.
// A naive recursive implementation of optimal binary search tree problem
#include <stdio.h>
#include <limits.h>
// A utility function to get sum of array elements freq[i] to freq[j]

int sum(int freq[], int i, int j);
// A recursive function to calculate cost of optimal binary search tree

int optCost(int freq[], int i, int j)
{
// Base cases
if (j < i)
// If there are no elements in this subarray
return 0;
if (j == i)
// If there is one element in this subarray
return freq[i];
// Get sum of freq[i], freq[i+1], ... freq[j]

int fsum = sum(freq, i, j);
// Initialize minimum value

int min = INT_MAX;
// One by one consider all elements as root and recursively find cost
// of the BST, compare the cost with min and update min if needed
for (int r = i; r <= j; ++r)
{
int cost = optCost(freq, i, r-1) + optCost(freq, r+1, j);
if (cost < min)

min = cost;
}
// Return minimum value

return min + fsum;
}
// The main function that calculates minimum cost of a Binary Search Tree.
// It mainly uses optCost() to find the optimal cost.
int optimalSearchTree(int keys[], int freq[], int n)
{
// Here array keys[] is assumed to be sorted in increasing order.
// If keys[] is not sorted, then add code to sort keys, and rearrange
// freq[] accordingly.
return optCost(freq, 0, n-1);
}

int sum(int freq[], int i, int j)
{
int s = 0;
for (int k = i; k <=j; k++)
s += freq[k];
return s;
}

int main()
{
int keys[] = {10, 12, 20};
int freq[] = {34, 8, 50};
int n = sizeof(keys)/sizeof(keys[0]);
printf("Cost of Optimal BST is %d ", optimalSearchTree(keys, freq, n));
return 0;
}
Output:
CostofOptimalBSTis142
Timecomplexityoftheabovenaiverecursiveapproachisexponential.Itshouldbenotedthattheabove
functioncomputesthesamesubproblemsagainandagain.Wecanseemanysubproblemsbeingrepeated
inthefollowingrecursiontreeforfreq[1..4].
Sincesamesuproblemsarecalledagain,thisproblemhasOverlappingSubprolemsproperty.Sooptimal
BSTproblemhasbothproperties(seethisandthis)ofadynamicprogrammingproblem.Likeothertypical
DynamicProgramming(DP)problems,recomputationsofsamesubproblemscanbeavoidedbyconstructing
atemporaryarraycost[][]inbottomupmanner.
DynamicProgrammingSolution
FollowingisC/C++implementationforoptimalBSTproblemusingDynamicProgramming.Weusean
auxiliaryarraycost[n][n]tostorethesolutionsofsubproblems.cost[0][n1]willholdthefinalresult.The
challengeinimplementationis,alldiagonalvaluesmustbefilledfirst,thenthevalueswhichlieontheline
justabovethediagonal.Inotherwords,wemustfirstfillallcost[i][i]values,thenallcost[i][i+1]values,then
allcost[i][i+2]values.Sohowtofillthe2Darrayinsuchmanner>Theideausedintheimplementationis
sameasMatrixChainMultiplicationproblem,weuseavariableLforchainlengthandincrementL,oneby
one.WecalculatecolumnnumberjusingthevaluesofiandL.
// Dynamic Programming code for Optimal Binary Search Tree Problem
#include <stdio.h>
#include <limits.h>

int sum(int freq[], int i, int j);
/* A Dynamic Programming based function that calculates minimum cost of

a Binary Search Tree. */
int optimalSearchTree(int keys[], int freq[], int n)
{
/* Create an auxiliary 2D matrix to store results of subproblems */
int cost[n][n];
/* cost[i][j] = Optimal cost of binary search tree that can be

formed from keys[i] to keys[j].
cost[0][n-1] will store the resultant cost */
// For a single key, cost is equal to frequency of the key

for (int i = 0; i < n; i++)
cost[i][i] = freq[i];
// Now we need to consider chains of length 2, 3, ... .

// L is chain length.
for (int L=2; L<=n; L++)
{
// i is row number in cost[][]
for (int i=0; i<=n-L+1; i++)
{
// Get column number j from row number i and chain length L
int j = i+L-1;
cost[i][j] = INT_MAX;
// Try making all keys in interval keys[i..j] as root

for (int r=i; r<=j; r++)
{
// c = cost when keys[r] becomes root of this subtree
int c = ((r > i)? cost[i][r-1]:0) +
((r < j)? cost[r+1][j]:0) +
sum(freq, i, j);
if (c < cost[i][j])
cost[i][j] = c;
}
}
}
return cost[0][n-1];
}

int sum(int freq[], int i, int j)
{
int s = 0;
for (int k = i; k <=j; k++)
s += freq[k];
return s;
}

int main()
{
int keys[] = {10, 12, 20};
int freq[] = {34, 8, 50};
int n = sizeof(keys)/sizeof(keys[0]);
printf("Cost of Optimal BST is %d ", optimalSearchTree(keys, freq, n));
return 0;
}
Output:
CostofOptimalBSTis142
Notes
1)ThetimecomplexityoftheabovesolutionisO(n^4).ThetimecomplexitycanbeeasilyreducedtoO(n^3)
byprecalculatingsumoffrequenciesinsteadofcallingsum()againandagain.
2)Intheabovesolutions,wehavecomputedoptimalcostonly.Thesolutionscanbeeasilymodifiedtostore
thestructureofBSTsalso.Wecancreateanotherauxiliaryarrayofsizentostorethestructureoftree.All
weneedtodois,storethechosenrintheinnermostloop.
=======================================================================

Trees

Uploaded by

Copyright:

Available Formats

Trees

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Trees

Uploaded by

Copyright:

Available Formats

FindtheclosestleafinaBinaryTree

/* Helper function that allocates a new node with the

// A utility function to find minimum of x and y

// A utility function to find distance of closest leaf of the tree

// Return minimum of left and right, plus one

// Traverse all ancestors and update result if any parent node

return findClosestUtil(root, k, ancestors, 0);

/* Driver program to test above functions*/

// Tree node constructor

// Diagonal sum method

// Map to store sum of node's data lying diagonally

// Assign the root's vertical distance as 0.

// Add root node to the queue

// Loop while the queue is not empty

// Get the vertical distance of the dequeued node.

// Sum over this node's right-child, right-of-right-child

// If for any node the left child is not null add

// Move to the current node's right child.

// Make an entry set from map.

// Traverse the map elements using the iterator.

// Tree node class

// Constructor of tree node

// Parameterized tree constructor

// Method that prints the bottom view.

// Initialize a variable 'hd' with 0 for the root element.

// TreeMap which stores key value pair sorted on key value

// Queue to store tree nodes in level order traversal

// Assign initialized horizontal distance value to root

// Loop until the queue is empty (standard level order loop)

// Extract the horizontal distance value from the

// Put the dequeued tree node to TreeMap having key

// If the dequeued node has a left child add it to the

// Extract the entries of map into a set to traverse

// Traverse the map elements using the iterator.

// Main driver class

root.right = new TreeNode(22);

n --> No. of elements present in input array.

// Returns sum of arr[0..index]. This function assumes

// index in BITree[] is 1 more than the index in arr[]

// Traverse ancestors of BITree[index]

// Move index to parent node

// Updates a node in Binary Index Tree (BITree) at given index

// index in BITree[] is 1 more than the index in arr[]

// Traverse all ancestors and add 'val'

// Update index to that of parent

// Constructs and returns a Binary Indexed Tree for given

// Store the actual values in BITree[] using update()

// Uncomment below lines to see contents of BITree[]

cout << BITree[i] << " ";

// Driver program to test above functions

// Let use test the update operation

cout << "\nSum of elements in arr[0..5] after update is "

// There are atmost 11 different chars in a valid IP address

// Maximum length of a valid IP address

// A utility function to find index of child for a given character 'c'

// A utility function to find character for a given child index.

// Function to create a new trie node.

// This method inserts an ip address and the corresponding

// Traversing over the length of the ip address.