

ChristmasError-Blog

C++ RBTree红黑树的实现

ChristmasError-Blog

C++ RBTree红黑树的实现

C++ 红黑树 数据结构 RBTree

字数统计: 2.7k阅读时长: 15 min

 2018/11/17   Share

• 
• 
• 
• 
• 

红黑树的性质

1. 每个结点不是红色就是黑色
2. 根节点是黑色的
3. 如果一个根节点是红色的，则它的两个叶子结点是黑色的(没有两个连续的红色结点)
4. 对于每个结点，从该结点到其所有后代叶结点的简单路径上，均包含相同数目的黑色结点(每条路径上黑色结点的数量相等)
5. 所有叶子结点为黑（与平时不同，这里的叶子结点是指空的叶子结点，即为NULL）

红黑树是一种特殊的二叉查找树，这意味着它满足二叉查找树的特征：任意一个节点所包含的键值，大于等于左孩子的键值，小于等于右孩子的键值。（相较于二叉查找树，我们只要对结点的颜色进行判断和调整）。
红黑树虽然复杂，但它的最坏情况运行时间也是非常良好的，它可以在O(log n)时间内做查找，插入和删除（n是树中结点树）。

红黑树结点

//RBTNode.h
enum RBTColor { Black, Red };

template<class KeyType>
struct  RBTNode
{
	KeyType key;
	RBTColor color;
	RBTNode<KeyType> * left;
	RBTNode<KeyType> * right;
	RBTNode<KeyType> * parent;
	RBTNode(KeyType k, RBTColor c, RBTNode* p, RBTNode*l, RBTNode*r) :
		key(k), color(c), parent(p), left(l), right(r) { };
};

红黑树的实现

//RBTree.h
#include"RBTNode.h"
#include <iomanip>

template<class T>
class  RBTree
{
public:
	RBTree();
	~RBTree();

	void insert(T key);    //插入结点，key为键值，外部接口
	void remove(T key); //删除key的节点
	RBTNode<T>* search(T key);
	void print();
	void preOrder();    //前序遍历 打印红黑树
	void inOrder();    //中序遍历
	void postOrder();    //后序遍历		
	


private:
	void leftRotate(RBTNode<T>* &root, RBTNode<T>* x);// 左旋
	void rightRotate(RBTNode<T>* &root, RBTNode<T>* y);// 右旋

	void insert(RBTNode<T>* &root, RBTNode<T>* node);// 插入结点，内部接口
	void InsertFixUp(RBTNode<T>* &root, RBTNode<T>* node);
	void destory(RBTNode<T>* &node);

	void remove(RBTNode<T>*& root, RBTNode<T>*node);//删除为KEY的结点
	void removeFixUp(RBTNode<T>* &root, RBTNode<T>* node, RBTNode<T>*parent);

	RBTNode<T>* search(RBTNode<T>*node, T key) const;
	void print(RBTNode<T>* node)const;
	void preOrder(RBTNode<T>* tree)const;
	void inOrder(RBTNode<T>* tree)const;
	void postOrder(RBTNode<T>* tree)const;
private:
	RBTNode<T>*root;
};

template<class T>		//构造函数
RBTree<T>::RBTree() :root(NULL) {
	root = nullptr;
}
template<class T>		//析构函数
RBTree<T>::~RBTree() {
	destory(root);
}
template<class T>		//左旋
void RBTree<T>::leftRotate(RBTNode<T>* &root, RBTNode<T>* x) {
	RBTNode<T>*y = x->right;
	x->right = y->left;
	if (y->left != NULL)
		y->left->parent = x;

	y->parent = x->parent;
	if (x->parent == NULL)
		root = y;
	else {
		if (x == x->parent->left)
			x->parent->left = y;
		else
			x->parent->right = y;
	}
	y->left = x;
	x->parent = y;
};
template<class T>		//右旋
void RBTree<T>::rightRotate(RBTNode<T>*&root, RBTNode<T>*y) {
	RBTNode<T>*x = y->left;
	y->left = x->right;
	if (x->right != NULL)
		x->right->parent = y;

	x->parent = y->parent;
	if (y->parent == NULL)
		root = x;
	else {
		if  (y == y->parent->right)
			y->parent->right = x;
		else
			y->parent->left = x;
	}
	x->right = y;
	y->parent = x;
};
template<class T>		//插入
void RBTree<T>::insert(T key)
{
	RBTNode<T>*z = new RBTNode<T>(key, Red, NULL, NULL, NULL);
	insert(root, z);
};
template<class T>
void RBTree<T>::insert(RBTNode<T>* &root, RBTNode<T>* node)
{
	RBTNode<T> *x = root;
	RBTNode<T> *y = NULL;
	while (x != NULL)
	{
		y = x;
		if (node->key > x->key)
			x = x->right;
		else
			x = x->left;
	}
	node->parent = y;
	if(y!=NULL)
	{
		if (node->key > y->key)
			y->right = node;
		else
			y->left = node;
	}
	else 
		root = node;
	node->color = Red;
	InsertFixUp(root, node);
};
template<class T>
void RBTree<T>::InsertFixUp(RBTNode<T>* &root, RBTNode<T>* node)
{
	RBTNode<T>*parent;
	parent = node->parent;
	while (node != RBTree::root  && parent->color == Red)
	{
		RBTNode<T>*gparent = parent->parent;
		if (gparent->left == parent)
		{
			RBTNode<T>*uncle = gparent->right;
			if (uncle != NULL && uncle->color == Red)
			{
				parent->color = Black;
				uncle->color = Black;
				gparent->color = Red;
				node = gparent;
				parent = node->parent;
			}
			else
			{
				if (parent->right == node)
				{
					leftRotate(root, parent);
					swap(node, parent);
				}
				rightRotate(root, gparent);
				gparent->color = Red;
				parent->color = Black;
				break;
			}
		}
		else
		{
			RBTNode<T>*uncle = gparent->left;
			if (uncle != NULL && uncle->color == Red)
			{
				gparent->color = Red;
				parent->color = Black;
				uncle->color = Black;

				node = gparent;
				parent = node->parent;
			}
			else
			{
				if (parent->left == node)
				{
					rightRotate(root, parent);
					swap(parent, node);
				}
				leftRotate(root, gparent);
				parent->color = Black;
				gparent->color = Red;
				break;
			}
		}
	}
	root->color = Black;
}
template<class T>
//销毁红黑树
void RBTree<T>::destory(RBTNode<T>* &node) 
{
	if (node == NULL)
		return;
	destory(node->left);
	destory(node->right);
	delete node;
	node = nullptr;
}

template<class T>
void RBTree<T>::remove(T key) 
{
	RBTNode<T>*deletenode = search(root,key);
	if (deletenode != NULL)
		remove(root, deletenode);
}
template<class T>
void RBTree<T>::remove(RBTNode<T>*&root, RBTNode<T>*node)
{
	RBTNode<T> *child, *parent;
	RBTColor color;
	//被删除节点左右结点都不为空(不为叶子结点)
	if (node->left != NULL && node->right != NULL)     
	{
		RBTNode<T> *replace = node;
		//找后继结点（当前节点的右子树的最下层的左结点）
		replace = node->right;
		while (replace->left != NULL)
		{
			replace = replace->left;
		}
		//被删节点不为根节点情况
		if (node->parent != NULL)
		{
			if (node->parent->left == node)
				node->parent->left = replace;
			else
				node->parent->right = replace;
		}
		//根节点情况
		else
			root = replace;
		//child是取代节点的右结点，是需要后续调整的结点
		//因为replace是后继结点，因此他不可能有左子节点
		//同理前驱结点不可能有右子节点
		child = replace->right;
		parent = replace->parent;
		color = replace->color;
		
		//被删结点是取代结点（repalce）的父节点的情况
		if (parent == node)
			parent = replace;
		else
		{
			//child结点存在的情况
			if (child != NULL)
				child->parent = parent;
			parent->left = child;

			replace->right = node->right;
			node->right->parent = replace;
		}
		replace->parent = node->parent;
		replace->color = node->color;
		replace->left = node->left;
		node->left->parent = replace;
		if (color == Black)
			removeFixUp(root, child, parent);

		delete node;
		return;
	}
	//当被删节点只有左(右)结点为空情况，找出被删结点的子节点
	if (node->left != NULL)    
		child = node->left;
	else
		child = node->right;

	parent = node->parent;
	color = node->color;
	if (child) 
	{
		child->parent = parent;
	}
	//被删除节点不为根节点
	if (parent)     
	{
		if (node == parent->left)
			parent->left = child;
		else
			parent->right = child;
	}
	//被删除节点为根节点
	else
		RBTree::root = child;		

	if (color == Black)
	{
		removeFixUp(root, child, parent);
	}
	delete node;

}
template<class T>
void RBTree<T>::removeFixUp(RBTNode<T>* &root, RBTNode<T>* node,RBTNode<T>*parent)
{
	RBTNode<T>*othernode;
	while ((!node) || node->color == Black && node != RBTree::root)
	{
		if (parent->left == node)
		{
			othernode = parent->right;
			if (othernode->color == Red)
			{
				othernode->color = Black;
				parent->color = Red;
				leftRotate(root, parent);
				othernode = parent->right;
			}
			else
			{
				if (!(othernode->right) || othernode->right->color == Black)
				{
					othernode->left->color=Black;
					othernode->color = Red;
					rightRotate(root, othernode);
					othernode = parent->right;
				}
				othernode->color = parent->color;
				parent->color = Black;
				othernode->right->color = Black;
				leftRotate(root, parent);
				node = root;
				break;
			}
		}
		else
		{
			othernode = parent->left;
			if (othernode->color == Red)
			{
				othernode->color = Black;
				parent->color = Red;
				rightRotate(root, parent);
				othernode = parent->left;
			}
			if ((!othernode->left || othernode->left->color == Black) && (!othernode->right || othernode->right->color == Black))
			{
				othernode->color = Red;
				node = parent;
				parent = node->parent;
			}
			else
			{
				if (!(othernode->left) || othernode->left->color == Black)
				{
					othernode->right->color = Black;
					othernode->color = Red;
					leftRotate(root, othernode);
					othernode = parent->left;
				}
				othernode->color = parent->color;
				parent->color = Black;
				othernode->left->color = Black;
				rightRotate(root, parent);
				node = root;
				break;
			}
		}
	}
	if (node)
		node->color = Black;
}

template<class T>
RBTNode<T>* RBTree<T>::search(T key) 
{
	return search(root, key);
}
template<class T>
RBTNode<T>* RBTree<T>::search(RBTNode<T>*node, T key) const
{
	if (node == NULL || node->key == key)
		return node;
	else
		if (key > node->key)
			return search(node->right, key);
		else
			return search(node->left, key);
}
template<class T>		//输出二叉树详细信息
void RBTree<T>::print() {
	if (root == NULL)
		cout << "empty RBtree\n";
	else
		print(root);
}
template<class T>
void RBTree<T>::print(RBTNode<T>* node)const {
	if (node == NULL)
		return;
	if (node->parent == NULL)
		cout << node->key << "(" << node->color << ") is root" << endl;
	else if(node->parent->left==node)
	{
		cout << node->key << "(" << node->color << ") is "<<node->parent->key<<"'s "<<"left child" << endl;
	}
	else
	{
		cout << node->key << "(" << node->color << ") is " << node->parent->key << "'s " << "right child" << endl;
	}
	print(node->left);
	print(node->right);
}
template<class T>		//前序遍历RB树
void RBTree<T>::preOrder() {
	if (root == NULL)
		cout << "empty RBtree\n";
	else
		preOrder(root);
};
template<class T>		 
void RBTree<T>::preOrder(RBTNode<T>* tree)const {
		if (tree != NULL) {
			cout << tree->key << " ";
			preOrder(tree->left);
			preOrder(tree->right);
		}
	}
template<class T>		//中遍历RB树
void RBTree<T>::inOrder() {
	if (root == NULL)
		cout << "empty RBtree\n";
	else
		inOrder(root);
};
template<class T>		 
void RBTree<T>::inOrder(RBTNode<T>* tree)const {
	if (tree != NULL) {
		inOrder(tree->left);
		cout << tree->key << " ";
		inOrder(tree->right);
	}
}
template<class T>      //后遍历RB树
void RBTree<T>::postOrder() {
	if (root == NULL)
		cout << "empty RBtree\n";
	else
		postOrder(root);
};
template<class T>		
void RBTree<T>::postOrder(RBTNode<T>* tree)const {
	if (tree != NULL) {
		postOrder(tree->left);
		postOrder(tree->right);
		cout << tree->key << " ";
	}
}

因为RBTree这里使用模板 ，因此不能将cpp和h文件的RBTree的声明和实现分离
原因：
模板定义很特殊。由template<…> 处理的任何东西都意味着编译器在当时不为它分配存储空间，它一直处于等待状态直到被一个模板实例告知。
在编译器和连接器的某一处，有一机制能去掉指定模板的多重定义。所以为了容易使用，几乎总是在头文件中放置全部的模板声明和定义。——《C++编程思想》第15章300页处

红黑树测试

//test.cpp
#include"RBTree.h"
#include<iostream>
#include<vector>
using namespace std;
int main() 
{
	vector<int> nums{ 10,40,30,60,90,70,20,50,80,100};
	RBTree<int> tree;
	for (auto num : nums)
		tree.insert(num);
	tree.preOrder();
	cout << endl;
	tree.inOrder();
	cout << endl;
	tree.postOrder();
	cout << endl;
	cout << "查找key为30的结点:\n";
	cout << endl<<tree.search(30)->key<<endl;
	cout << "删除key为100的结点\n";
	tree.remove(100);
	tree.preOrder();
	cout << endl;
	cout << "\n红黑树详细信息：\n";
	tree.print();
	cin.get();
	return 0;
}

测试运行结果

60 30 10 20 40 50 80 70 90 100
10 20 30 40 50 60 70 80 90 100
20 10 50 40 30 70 100 90 80 60
查找key为30的结点:

30
删除key为100的结点
60 30 10 20 40 50 80 70 90

红黑树详细信息：
60(0) is root
30(1) is 60's left child
10(0) is 30's left child
20(1) is 10's right child
40(0) is 30's right child
50(1) is 40's right child
80(1) is 60's right child
70(0) is 80's left child
90(0) is 80's right child

• Next Post
  C++ GeneralizedTable 广义表实现
• Previous Post
  C++利用RAII技术让系统为我们管理资源（资源获取即初始化——Resource Acquisition Is Initialization）

Powered by Hexotheme Archer

Visit & Clik:

CATALOG

1. 1. 红黑树的性质
2. 2. 红黑树结点
3. 3. 红黑树的实现
4. 4. 红黑树测试
5. 5. 测试运行结果



• Archive
• Tag
• Cate

Total : 16

2018

• 12/27C++ BST-binary search tree 二叉搜索树的实现
• 12/20C++ GeneralizedTable 广义表实现
• 11/17C++ RBTree红黑树的实现
• 11/04C++利用RAII技术让系统为我们管理资源（资源获取即初始化——Resource Acquisition Is Initialization）
• 10/23C++hash_map 哈希表
• 10/21C++11 Lambda表达式
• 10/20C++智能指针：shared_ptr,unique_ptr,weak_ptr
• 10/18C++中const的“就近原则”
• 10/17《深入了解计算机系统》笔记——程序性能优化
• 10/15《深入了解计算机系统》笔记——汇编语言和机器语言
• 10/15《深入了解计算机系统》笔记——信息的表示和处理
• 10/13《深入了解计算机系统》笔记——计算机系统漫游
• 10/10leetcode刷题笔记：144排序链表——归并排序
• 09/27泛型编程学习，编写一个类似STL库中的简易list的迭代器（iterator）
• 09/23Hexo博客发布—Markdown基础语法
• 09/22Hello World

C++ 数据结构 BST Lambda表达式 Lambda函数 C++ 数据结构 广义表 指针，const C++ 红黑树 数据结构 RBTree hash_map 哈希表 这是建立个人Blog来的第一篇测试博客 RAII 智能指针 很基础的Markdown语法及其效果演示 《深入了解计算机系统》 leetcode刷题笔记 STL学习笔记



缺失模块。
1、请确保node版本大于6.2
2、在博客根目录（注意不是archer根目录）执行以下命令：
npm i hexo-generator-json-content --save
3、在根目录_config.yml里添加配置：

jsonContent:
  meta: false
  pages: false
  posts:
    title: true
    date: true
    path: true
    text: false
    raw: false
    content: false
    slug: false
    updated: false
    comments: false
    link: false
    permalink: false
    excerpt: false
    categories: true
    tags: true

