数据结构 — AVL tree(平衡二叉树)
发布日期:2021-06-30 19:49:42 浏览次数:2 分类:技术文章

本文共 4480 字,大约阅读时间需要 14 分钟。

# BST(二叉搜索树、二叉查找树、二叉排序树)

定义:

1、要么是一棵空树

2、如果不为空,那么其左子树节点的值都小于根节点的值;右子树节点的值都大于根节点的值

3、其左右子树也是二叉搜索树

# AVL tree(平衡二叉树)

定义:

 平衡二叉树(Balanced Binary Tree)又被称为AVL树(有别于AVL算法),且具有以下性质:它是一 棵空树或它的左右两个子树的高度差的绝对值不超过1,并且左右两个子树都是一棵平衡二叉树。平衡二叉树的常用算法有红黑树、AVL、Treap、伸展树等。

最小不平衡子树: 以离插入结点最近、且平衡因子绝对值大于 1 的结点作根结点的子树。

调整该子树的分为四种情况:

(1)LL形

(2)LR形

(3)RR形

(4)RL形

代码实现:

LR:

RL:

#include
#include
using namespace std;#define FALSE 0#define TRUE 1typedef struct { int key;} element;typedef struct tree_node { struct tree_node *left_child; element data; short bf; struct tree_node *right_child;} tree_node, *tree_pointer;int unbalanced = FALSE;tree_pointer root = NULL;void left_rotation(tree_pointer *parent, int *unbalanced);void right_rotation(tree_pointer *parent, int *unbalanced);void avl_insert(tree_pointer *parent, element x, int *unbalanced);/* *1.如果要插入的元素的父节点为空则为其分配内存并处理 *2.如果小于父节点的数据域则插入父节点的左孩子,并旋转 *3.如果大于父节点的数据域则插入父节点的右孩子,并旋转 */void avl_insert(tree_pointer *parent, element x, int *unbalanced) { if(!*parent) { *unbalanced = TRUE; *parent = new tree_node(); (*parent)->left_child = (*parent)->right_child = NULL; (*parent)->bf = 0; (*parent)->data = x; } else if(x.key < (*parent)->data.key) { avl_insert(&(*parent)->left_child, x, unbalanced); if(*unbalanced) { /* * unbalanced表示是插完之后就不平衡了 和 判断还用不用处理平衡因子 */ switch((*parent)->bf) { case -1: (*parent)->bf = 0; *unbalanced = FALSE; break; case 0: (*parent)->bf = 1; break; case 1: left_rotation(parent,unbalanced); } } } else if(x.key > (*parent)->data.key) { avl_insert(&(*parent)->right_child, x, unbalanced); if(*unbalanced) { switch((*parent)->bf) { case 1: (*parent)->bf = 0; *unbalanced = FALSE; break; case 0: (*parent)->bf = 1; break; case -1: right_rotation(parent, unbalanced); } } } else { *unbalanced = FALSE; printf("该元素已经存在!\n"); }}void left_rotation(tree_pointer *parent, int *unbalanced) { tree_pointer grand_child, child; child = (*parent)->left_child; if(child->bf == 1) { //LL (*parent)->left_child = child->right_child; child->right_child = *parent; (*parent)->bf = 0; (*parent) = child; } else { //LR grand_child = child->right_child; child->right_child = grand_child->left_child; grand_child->left_child = child; (*parent)->left_child = grand_child->right_child; grand_child->right_child = (*parent); switch(grand_child->bf) { case 1: (*parent)->bf = -1; child->bf = 0; case 0: (*parent)->bf = child->bf = 0; case -1: (*parent)->bf = 0; child->bf = 1; } (*parent) = grand_child; } (*parent)->bf = 0; *unbalanced = FALSE;}void right_rotation(tree_pointer *parent, int *unbalanced) { tree_pointer grand_child, child; child = (*parent)->right_child; if(child->bf == -1) { //RR (*parent)->right_child = child->left_child; child->left_child = (*parent); (*parent) = child; } else { //RL grand_child = child->left_child; child->left_child = grand_child->right_child; grand_child->right_child = child; (*parent)->right_child = grand_child->left_child; grand_child->left_child = (*parent); switch(grand_child->bf) { case 1: (*parent)->bf = 0; child->bf = -1; case 0: (*parent)->bf = child->bf = 0; case -1: (*parent)->bf = 1; child->bf = 0; } (*parent) = grand_child; } (*parent)->bf = 0; *unbalanced = FALSE;}void raverse(tree_pointer root) { if(root) { printf("%d ",root->data.key); raverse(root->left_child); raverse(root->right_child); }}int main() { int arr[11] = {15,6,18,3,7,17,20,2,4,13,9}; element arr_x[11]; for(int i = 0; i<11; i++) { arr_x[i].key = arr[i]; // cout<
<

转载地址:https://lipenglin.blog.csdn.net/article/details/50403071 如侵犯您的版权,请留言回复原文章的地址,我们会给您删除此文章,给您带来不便请您谅解!

上一篇:图 之 MST(最小生成树 — kruskal算法 )并查集实现
下一篇:Java — 面向对象(自测题)

发表评论

最新留言

不错!
[***.144.177.141]2024年04月15日 05时39分52秒