给定一个笛卡尔树的有序遍历,任务是从中构建整个树。
例子:
Input: arr[] = {1, 5, 3}
Output: 1 5 3
5
/ \
1 3
Input: arr[] = {3, 7, 4, 8}
Output: 3 7 4 8
8
/
7
/ \
3 4
方法:我们已经在这里看到了一个算法,它平均花费 O(NlogN) 时间,但在最坏的情况下可以达到 O(N 2 )。
在本文中,我们将看到如何在O(Nlog(N)) 的最坏情况下运行时间构建笛卡尔函数。为此,我们将使用段树来回答范围最大查询。
下面将是我们在范围 {L, R} 上的递归算法:
- 使用段树上的范围最大查询找到此范围 {L, R} 中的最大值。假设’M’是范围内的最大值。
- 选取 ‘arr[M]’ 作为当前节点的值并使用该值创建一个节点。
- 求解范围 {L, M-1} 和 {M+1, R}。
- 将{L, M-1}返回的节点设置为当前节点的左孩子,{M+1, R}为右孩子。
下面是上述方法的实现:
C++
// C++ implementation of the approach
#include
using namespace std;
#define maxLen 30
// Node of the BST
struct node {
int data;
node* left;
node* right;
node(int data)
{
left = NULL;
right = NULL;
this->data = data;
}
};
// Array to store segment tree
int segtree[maxLen * 3];
// Function to create segment-tree to answer
// range-max query
int buildTree(int l, int r, int i, int* arr)
{
// Base case
if (l == r) {
segtree[i] = l;
return l;
}
// Maximum index in left range
int l1 = buildTree(l, (l + r) / 2,
2 * i + 1, arr);
// Maximum index in right range
int r1 = buildTree((l + r) / 2 + 1,
r, 2 * i + 2, arr);
// If value at l1 > r1
if (arr[l1] > arr[r1])
segtree[i] = l1;
// Else
else
segtree[i] = r1;
// Returning the maximum in range
return segtree[i];
}
// Function to answer range max query
int rangeMax(int l, int r, int rl,
int rr, int i, int* arr)
{
// Base cases
if (r < rl || l > rr)
return -1;
if (l >= rl and r <= rr)
return segtree[i];
// Maximum in left range
int l1 = rangeMax(l, (l + r) / 2, rl,
rr, 2 * i + 1, arr);
// Maximum in right range
int r1 = rangeMax((l + r) / 2 + 1, r,
rl, rr, 2 * i + 2, arr);
// l1 = -1 means left range
// was out-side required range
if (l1 == -1)
return r1;
if (r1 == -1)
return l1;
// Returning the maximum
// among two ranges
if (arr[l1] > arr[r1])
return l1;
else
return r1;
}
// Function to print the inorder
// traversal of the binary tree
void inorder(node* curr)
{
// Base case
if (curr == NULL)
return;
// Traversing the left sub-tree
inorder(curr->left);
// Printing current node
cout << curr->data << " ";
// Traversing the right sub-tree
inorder(curr->right);
}
// Function to build cartesian tree
node* createCartesianTree(int l, int r, int* arr, int n)
{
// Base case
if (r < l)
return NULL;
// Maximum in the range
int m = rangeMax(0, n - 1, l, r, 0, arr);
// Creating current node
node* curr = new node(arr[m]);
// Creating left sub-tree
curr->left = createCartesianTree(l, m - 1, arr, n);
// Creating right sub-tree
curr->right = createCartesianTree(m + 1, r, arr, n);
// Returning current node
return curr;
}
// Driver code
int main()
{
// In-order traversal of cartesian tree
int arr[] = { 8, 11, 21, 100, 5, 70, 55 };
// Size of the array
int n = sizeof(arr) / sizeof(int);
// Building the segment tree
buildTree(0, n - 1, 0, arr);
// Building and printing cartesian tree
inorder(createCartesianTree(0, n - 1, arr, n));
}
Java
// Java implementation of the approach
import java.util.*;
class GFG
{
static int maxLen = 30;
// Node of the BST
static class node
{
int data;
node left;
node right;
node(int data)
{
left = null;
right = null;
this.data = data;
}
};
// Array to store segment tree
static int segtree[] = new int[maxLen * 3];
// Function to create segment-tree to answer
// range-max query
static int buildTree(int l, int r,
int i, int[] arr)
{
// Base case
if (l == r)
{
segtree[i] = l;
return l;
}
// Maximum index in left range
int l1 = buildTree(l, (l + r) / 2,
2 * i + 1, arr);
// Maximum index in right range
int r1 = buildTree((l + r) / 2 + 1,
r, 2 * i + 2, arr);
// If value at l1 > r1
if (arr[l1] > arr[r1])
segtree[i] = l1;
// Else
else
segtree[i] = r1;
// Returning the maximum in range
return segtree[i];
}
// Function to answer range max query
static int rangeMax(int l, int r, int rl,
int rr, int i, int[] arr)
{
// Base cases
if (r < rl || l > rr)
return -1;
if (l >= rl && r <= rr)
return segtree[i];
// Maximum in left range
int l1 = rangeMax(l, (l + r) / 2, rl,
rr, 2 * i + 1, arr);
// Maximum in right range
int r1 = rangeMax((l + r) / 2 + 1, r,
rl, rr, 2 * i + 2, arr);
// l1 = -1 means left range
// was out-side required range
if (l1 == -1)
return r1;
if (r1 == -1)
return l1;
// Returning the maximum
// among two ranges
if (arr[l1] > arr[r1])
return l1;
else
return r1;
}
// Function to print the inorder
// traversal of the binary tree
static void inorder(node curr)
{
// Base case
if (curr == null)
return;
// Traversing the left sub-tree
inorder(curr.left);
// Printing current node
System.out.print(curr.data + " ");
// Traversing the right sub-tree
inorder(curr.right);
}
// Function to build cartesian tree
static node createCartesianTree(int l, int r,
int[] arr, int n)
{
// Base case
if (r < l)
return null;
// Maximum in the range
int m = rangeMax(0, n - 1, l, r, 0, arr);
// Creating current node
node curr = new node(arr[m]);
// Creating left sub-tree
curr.left = createCartesianTree(l, m - 1, arr, n);
// Creating right sub-tree
curr.right = createCartesianTree(m + 1, r, arr, n);
// Returning current node
return curr;
}
// Driver code
public static void main(String args[])
{
// In-order traversal of cartesian tree
int arr[] = { 8, 11, 21, 100, 5, 70, 55 };
// Size of the array
int n = arr.length;
// Building the segment tree
buildTree(0, n - 1, 0, arr);
// Building && printing cartesian tree
inorder(createCartesianTree(0, n - 1, arr, n));
}
}
// This code is contributed by Arnab Kundu
Python3
# Python3 implementation of the approach
# Node of a linked list
class Node:
def __init__(self, data = None, left = None,
right = None ):
self.data = data
self.right = right
self.left = left
maxLen = 30
# Array to store segment tree
segtree = [0]*(maxLen * 3)
# Function to create segment-tree to answer
# range-max query
def buildTree(l , r ,i , arr):
global segtree
global maxLen
# Base case
if (l == r) :
segtree[i] = l
return l
# Maximum index in left range
l1 = buildTree(l, int((l + r) / 2),
2 * i + 1, arr)
# Maximum index in right range
r1 = buildTree(int((l + r) / 2) + 1,r,
2 * i + 2, arr)
# If value at l1 > r1
if (arr[l1] > arr[r1]):
segtree[i] = l1
# Else
else:
segtree[i] = r1
# Returning the maximum in range
return segtree[i]
# Function to answer range max query
def rangeMax(l, r, rl, rr, i, arr):
global segtree
global maxLen
# Base cases
if (r < rl or l > rr):
return -1
if (l >= rl and r <= rr):
return segtree[i]
# Maximum in left range
l1 = rangeMax(l, int((l + r) / 2), rl,
rr, 2 * i + 1, arr)
# Maximum in right range
r1 = rangeMax(int((l + r) / 2) + 1, r, rl,
rr, 2 * i + 2, arr)
# l1 = -1 means left range
# was out-side required range
if (l1 == -1):
return r1
if (r1 == -1):
return l1
# Returning the maximum
# among two ranges
if (arr[l1] > arr[r1]):
return l1
else:
return r1
# Function to print the inorder
# traversal of the binary tree
def inorder(curr):
# Base case
if (curr == None):
return
# Traversing the left sub-tree
inorder(curr.left)
# Printing current node
print(curr.data, end= " ")
# Traversing the right sub-tree
inorder(curr.right)
# Function to build cartesian tree
def createCartesianTree(l , r , arr, n):
# Base case
if (r < l):
return None
# Maximum in the range
m = rangeMax(0, n - 1, l, r, 0, arr)
# Creating current node
curr = Node(arr[m])
# Creating left sub-tree
curr.left = createCartesianTree(l, m - 1, arr, n)
# Creating right sub-tree
curr.right = createCartesianTree(m + 1, r, arr, n)
# Returning current node
return curr
# Driver code
# In-order traversal of cartesian tree
arr = [ 8, 11, 21, 100, 5, 70, 55 ]
# Size of the array
n = len(arr)
# Building the segment tree
buildTree(0, n - 1, 0, arr)
# Building && printing cartesian tree
inorder(createCartesianTree(0, n - 1, arr, n))
# This code is contributed by Arnab Kundu
C#
// C# implementation of the approach
using System;
class GFG
{
static int maxLen = 30;
// Node of the BST
public class node
{
public int data;
public node left;
public node right;
public node(int data)
{
left = null;
right = null;
this.data = data;
}
};
// Array to store segment tree
static int []segtree = new int[maxLen * 3];
// Function to create segment-tree to answer
// range-max query
static int buildTree(int l, int r,
int i, int[] arr)
{
// Base case
if (l == r)
{
segtree[i] = l;
return l;
}
// Maximum index in left range
int l1 = buildTree(l, (l + r) / 2,
2 * i + 1, arr);
// Maximum index in right range
int r1 = buildTree((l + r) / 2 + 1,
r, 2 * i + 2, arr);
// If value at l1 > r1
if (arr[l1] > arr[r1])
segtree[i] = l1;
// Else
else
segtree[i] = r1;
// Returning the maximum in range
return segtree[i];
}
// Function to answer range max query
static int rangeMax(int l, int r, int rl,
int rr, int i, int[] arr)
{
// Base cases
if (r < rl || l > rr)
return -1;
if (l >= rl && r <= rr)
return segtree[i];
// Maximum in left range
int l1 = rangeMax(l, (l + r) / 2, rl,
rr, 2 * i + 1, arr);
// Maximum in right range
int r1 = rangeMax((l + r) / 2 + 1, r,
rl, rr, 2 * i + 2, arr);
// l1 = -1 means left range
// was out-side required range
if (l1 == -1)
return r1;
if (r1 == -1)
return l1;
// Returning the maximum
// among two ranges
if (arr[l1] > arr[r1])
return l1;
else
return r1;
}
// Function to print the inorder
// traversal of the binary tree
static void inorder(node curr)
{
// Base case
if (curr == null)
return;
// Traversing the left sub-tree
inorder(curr.left);
// Printing current node
Console.Write(curr.data + " ");
// Traversing the right sub-tree
inorder(curr.right);
}
// Function to build cartesian tree
static node createCartesianTree(int l, int r,
int[] arr, int n)
{
// Base case
if (r < l)
return null;
// Maximum in the range
int m = rangeMax(0, n - 1, l, r, 0, arr);
// Creating current node
node curr = new node(arr[m]);
// Creating left sub-tree
curr.left = createCartesianTree(l, m - 1,
arr, n);
// Creating right sub-tree
curr.right = createCartesianTree(m + 1, r,
arr, n);
// Returning current node
return curr;
}
// Driver code
public static void Main()
{
// In-order traversal of cartesian tree
int []arr = { 8, 11, 21, 100, 5, 70, 55 };
// Size of the array
int n = arr.Length;
// Building the segment tree
buildTree(0, n - 1, 0, arr);
// Building && printing cartesian tree
inorder(createCartesianTree(0, n - 1, arr, n));
}
}
// This code is contributed by AnkitRai01
Javascript
输出:
8 11 21 100 5 70 55
时间复杂度: O(NlogN)
如果您希望与专家一起参加现场课程,请参阅DSA 现场工作专业课程和学生竞争性编程现场课程。