哈夫曼树与哈弗曼编码
- 哈夫曼树
带权路径长度(WPL):设二叉树有n个叶子结点,每个叶子结点带有权值 Wk,从根结点到每个叶子结点的长度为 Lk,则每个叶子结点的带权路径长度之和就是:
WPL =
最优二叉树或哈夫曼树: WPL最小的二叉树
- 哈夫曼树的特点:
①没有度为1的结点
②n个叶子结点的哈夫曼树共有2n-1个结点
③哈夫曼树的任意非叶节点的左右子树交换后仍是哈夫曼树
④对同一组权值{w1 ,w2 , …… , wn},存在不同构的两棵哈夫曼树
- 哈夫曼树的构造
每次把权值最小的两颗二叉树合并.(利用堆)
基本操作
- HuffmanTree的建立
- 带权路径的求解
- HuffmanCoding
#include<limits.h> /* INT_MAX等 */
#include<stdio.h> /* EOF(=^Z或F6),NULL */
typedef struct
{
unsigned int weight;
unsigned int parent,lchild,rchild;
}HTNode,*HuffmanTree; /* 动态分配数组存储赫夫曼树 */
typedef char **HuffmanCode; /* 动态分配数组存储赫夫曼编码表 */
int min1(HuffmanTree t,int i)
{ /* 函数void select()调用 */
int j,flag;
unsigned int k=UINT_MAX; /* 取k为不小于可能的值 */
for(j=1;j<=i;j++)
if(t[j].weight<k&&t[j].parent==0)
k=t[j].weight,flag=j;
t[flag].parent=1;
return flag;
}
void select(HuffmanTree t,int i,int *s1,int *s2)
{ /* s1为最小的两个值中序号小的那个 */
int j;
*s1=min1(t,i);
*s2=min1(t,i);
if(*s1>*s2)
{
j=*s1;
*s1=*s2;
*s2=j;
}
}
void HuffmanCoding(HuffmanTree *HT,HuffmanCode *HC,int *w,int n)
{ /* w存放n个字符的权值(均>0),构造赫夫曼树HT,并求出n个字符的赫夫曼编码HC */
int m,i,s1,s2;
unsigned c,cdlen;
HuffmanTree p;
char *cd;
if(n<=1)
return;
m=2*n-1;
*HT=(HuffmanTree)malloc((m+1)*sizeof(HTNode)); /* 0号单元未用 */
for(p=*HT+1,i=1;i<=n;++i,++p,++w)
{
(*p).weight=*w;
(*p).parent=0;
(*p).lchild=0;
(*p).rchild=0;
}
for(;i<=m;++i,++p)
(*p).parent=0;
for(i=n+1;i<=m;++i) /* 建赫夫曼树 */
{ /* 在HT[1~i-1]中选择parent为0且weight最小的两个结点,其序号分别为s1和s2 */
select(*HT,i-1,&s1,&s2);
(*HT)[s1].parent=(*HT)[s2].parent=i;
(*HT)[i].lchild=s1;
(*HT)[i].rchild=s2;
(*HT)[i].weight=(*HT)[s1].weight+(*HT)[s2].weight;
}
/* 以下为无栈非递归遍历赫夫曼树,求赫夫曼编码*/
*HC=(HuffmanCode)malloc((n+1)*sizeof(char*));
/* 分配n个字符编码的头指针向量([0]不用) */
cd=(char*)malloc(n*sizeof(char)); /* 分配求编码的工作空间 */
c=m;
cdlen=0;
for(i=1;i<=m;++i)
(*HT)[i].weight=0; /* 遍历赫夫曼树时用作结点状态标志 */
while(c)
{
if((*HT)[c].weight==0)
{ /* 向左 */
(*HT)[c].weight=1;
if((*HT)[c].lchild!=0)
{
c=(*HT)[c].lchild;
cd[cdlen++]='0';
}
else if((*HT)[c].rchild==0)
{ /* 登记叶子结点的字符的编码 */
(*HC)[c]=(char *)malloc((cdlen+1)*sizeof(char));
cd[cdlen]='\0';
strcpy((*HC)[c],cd); /* 复制编码(串) */
}
}
else if((*HT)[c].weight==1)
{ /* 向右 */
(*HT)[c].weight=2;
if((*HT)[c].rchild!=0)
{
c=(*HT)[c].rchild;
cd[cdlen++]='1';
}
}
else
{ /* HT[c].weight==2,退回 */
(*HT)[c].weight=0;
c=(*HT)[c].parent;
--cdlen; /* 退到父结点,编码长度减1 */
}
}
free(cd);
}
void main()
{
HuffmanTree HT;
HuffmanCode HC;
int *w,n,i;
printf("请输入权值的个数(>1):");
scanf("%d",&n);
w=(int *)malloc(n*sizeof(int));
printf("请依次输入%d个权值(整型):\n",n);
for(i=0;i<=n-1;i++)
scanf("%d",w+i);
HuffmanCoding(&HT,&HC,w,n);
for(i=1;i<=n;i++)
puts(HC[i]);
}
题目
- Input Specification:
Each input file contains one test case. For each case, the first line gives an integer NN (2\le N\le 632≤N≤63), then followed by a line that contains all the NN distinct characters and their frequencies in the following format:
c[1] f[1] c[2] f[2] ... c[N] f[N]
where c[i] is a character chosen from {'0' - '9', 'a' - 'z', 'A' - 'Z', '_'}, and f[i] is the frequency of c[i] and is an integer no more than 1000. The next line gives a positive integer MM (\le 1000≤1000), then followed by MM student submissions. Each student submission consists of NN lines, each in the format:
c[i] code[i]
where c[i] is the i-th character and code[i] is an non-empty string of no more than 63 '0's and '1's.
- Output Specification:
For each test case, print in each line either "Yes" if the student's submission is correct, or "No" if not.
Note: The optimal solution is not necessarily generated by Huffman algorithm. Any prefix code with code length being optimal is considered correct.
Sample Input:
7
A 1 B 1 C 1 D 3 E 3 F 6 G 6
4
A 00000
B 00001
C 0001
D 001
E 01
F 10
G 11
A 01010
B 01011
C 0100
D 011
E 10
F 11
G 00
A 000
B 001
C 010
D 011
E 100
F 101
G 110
A 00000
B 00001
C 0001
D 001
E 00
F 10
G 1
- Sample Output:
Yes
Yes
No
No
AC代码
#include <iostream>
#include <cstdio>
#include <vector>
#include <string>
using namespace std;
#define MAXSIZE 64
int nodenum;
int c[64];
char f[64];
typedef struct TNode* HuffTree;
struct TNode
{
HuffTree left;
HuffTree right;
int freq;
};
struct heap
{
HuffTree* data;
int size;
int capacity;
};
typedef struct heap* Minheap;
Minheap CreatHeap()
{
Minheap H = new struct heap;
H->data = new HuffTree[MAXSIZE];
H->size = 0;
H->capacity = MAXSIZE;
H->data[0] = new struct TNode;
H->data[0]->freq = -1;
return H;
}
bool Insert(Minheap H, HuffTree f)
{
int i;
if (H->size == H->capacity){
printf("minheap is full");
return false;
}
i = ++H->size;
for (; H->data[i / 2]->freq > f->freq; i /= 2)
H->data[i] = H->data[i / 2];
H->data[i] = f;
return true;
}
bool IsEmpty(Minheap H)
{
return (H->size == 0);
}
HuffTree DeleteMin(Minheap H)
{
int Parent, Child;
HuffTree MinItem, X;
if (IsEmpty(H)) {
printf("minheap is empty");
}
MinItem = H->data[1];
X = H->data[H->size--];
for (Parent = 1; Parent * 2 <= H->size; Parent = Child) {
Child = Parent * 2;
if ((Child != H->size) && (H->data[Child]->freq > H->data[Child + 1]->freq))
Child++;
if (X->freq <= H->data[Child]->freq) break;
else
H->data[Parent] = H->data[Child];
}
H->data[Parent] = X;
return MinItem;
}
HuffTree HuffmanTree()
{
Minheap h = CreatHeap();
for (int i = 0; i < nodenum; i++)
{
HuffTree f = new struct TNode;
f->freq = c[i]; //全局变量赋值
f->left = NULL;
f->right = NULL;
if (!Insert(h, f))
break;
}
for (;;)
{
if (h->size == 1) break;
HuffTree f = new struct TNode;
f->left = DeleteMin(h);
f->right = DeleteMin(h);
f->freq = f->left->freq + f->right->freq;
Insert(h, f);
//printf("fleft=%d,fright=%d,ffreq=%d\n",f->left->freq,f->right->freq,f->freq);
}
return DeleteMin(h);
}
int WPL(HuffTree tree, int depth)
{
if ((!tree->left) && (!tree->right))
{
//printf("depth:%d,leaf->freq:%d\n",depth,tree->freq);
return depth*(tree->freq);
}
else
{
//printf("depth:%d,tree->freq:%d\n",depth,tree->freq);
return WPL(tree->left, depth + 1) + WPL(tree->right, depth + 1);
}
}
bool check(HuffTree tree, string s)
{
bool flag = false;
HuffTree p = tree;
for (int i = 0; i < s.size(); i++)
{
if (s[i] == '0')
{
if (!p->left)
{
p->left = new TNode;
p->left->left = NULL;
p->left->right = NULL;
p = p->left;
flag = true;
}
else
{
p = p->left;
}
}
else if (s[i] == '1')
{
if (!p->right)
{
p->right = new TNode;
p->right->left = NULL;
p->right->right = NULL;
p = p->right;
flag = true;
}
else
{
p = p->right;
}
}
}
return flag;
}
int main()
{
int case_;
cin >> nodenum;
for (int i = 0; i < nodenum; i++)
{
cin >> f[i] >> c[i];
}
HuffTree tree = HuffmanTree();
int wpl = WPL(tree, 0); //权值可以放在建HufumanTree的时候计算
cin >> case_;
for (int j = 0; j < case_; j++)
{
HuffTree root = new TNode;
root->left = NULL;
root->right = NULL;
int s_wpl = 0;
string judge = "";
for (int i = 0; i<nodenum; i++)
{
char ch;
string s;
cin >> ch >> s;
if (s.size()>nodenum - 1){
judge = "No";
break;
}
s_wpl += s.size()*c[i];
if (!check(root, s)) //判断序列是否满足要求
judge = "No";
}
if (judge.empty() && s_wpl == wpl)
judge = "Yes";
else
judge = "No";
cout << judge << endl;
}
return 0;
}
- 补充
//补充,理解思路,时间复杂度O(N*logN)
HuffTree HuffmanTree(Minheap H)
{
//假设H->size个权值已经在H->data->frq里
int i;
HuffTree T;
BuildMinTree(H); //将H->data[]按权值调整为最小堆
for (i = 0; i < H->size;i++)
{
T = malloc(sizeof(struct TNode));
T->left = DeleteMin(H); //从最小堆中删除一个结点,作为新T的左子结点
T->right = DeleteMin(H); //从最小堆中删除一个结点,作为新T的右子节点
T->freq = T->left->freq + T->right->freq; //计算新权值
Insert(H, T);
}
T = DeleteMin(H);
return T;
}
题目来源
Reference
05-树9 Huffman Codes
05-树9 Huffman Codes (用优先队列实现)
05-树9 Huffman Codes及基本操作的更多相关文章
-
PAT 05-树8 Huffman Codes
以现在的生产力,是做不到一天一篇博客了.这题给我难得不行了,花了两天时间在PAT上还有测试点1没过,先写上吧.记录几个做题中的难点:1.本来比较WPL那块我是想用一个函数实现的,无奈我对传字符串数组无 ...
-
05-树9 Huffman Codes
哈夫曼树 Yes 需满足两个条件:1.HuffmanTree 结构不同,但WPL一定.子串WPL需一致 2.判断是否为前缀码 开始判断用的strstr函数,但其传值应为char *,不能用在strin ...
-
pta5-9 Huffman Codes (30分)
5-9 Huffman Codes (30分) In 1953, David A. Huffman published his paper "A Method for the Const ...
-
PTA 05-树9 Huffman Codes (30分)
题目地址 https://pta.patest.cn/pta/test/16/exam/4/question/671 5-9 Huffman Codes (30分) In 1953, David ...
-
数据结构慕课PTA 05-树9 Huffman Codes
题目内容 In 1953, David A. Huffman published his paper "A Method for the Construction of Minimum-Re ...
-
哈夫曼树(Huffman Tree)与哈夫曼编码
哈夫曼树(Huffman Tree)与哈夫曼编码(Huffman coding)
-
05-树9 Huffman Codes (30 分)
In 1953, David A. Huffman published his paper "A Method for the Construction of Minimum-Redunda ...
-
05-树9 Huffman Codes (30 分)
In 1953, David A. Huffman published his paper "A Method for the Construction of Minimum-Redunda ...
-
Huffman codes
05-树9 Huffman Codes(30 分) In 1953, David A. Huffman published his paper "A Method for the Const ...
随机推荐
-
Java工具Eclipse
一.下载Eclipse 从官网渠道下载或从公司共享软件目录下载均可. a) http://www.eclipse.org/downloads/eclipse-packages/ ...
-
安装和使用Linux花生壳(公网版)
一.安装说明 1.下载相应的安装包,安装程序 2.运行程序.配置程序(默认使用/etc/phlinux.conf,如果不存在这个文件则自动进入交互配置) [root@localhost -]# phd ...
-
Android项目实战(八):列表右侧边栏拼音展示效果
之前忙着做项目,好久之前的技术都没有时间总结,而发现自己的博客好多写的技术都比自己掌握的时候晚了很多.不管怎么样,写技术博客一定是一个想成为优秀程序猿或者已经是优秀程序猿必须做的.好吧,下面进行学习阶 ...
-
ExecutorService线程池讲解
ExecutorService 建立多线程的步骤: 1.定义线程类 class Handler implements Runnable{ } 2.建立ExecutorService线程池 Execut ...
-
使用dom4j 读取XML文件
第一次接触dom4j的时候,感觉这个东西很神秘,因为之前虽然知道XML文件吧,但从来没有用过,一直感觉XML肯定不好操作.当得知,dom4j可以很容易的操作读取XML文件时,不免有些好奇,那么,用do ...
-
bookStore第二篇【图书模块、前台页面】
图书模块 分析 在设计图书管理的时候,我们应该想到:图书和分类是有关系的.一个分类可以对应多本图书. 为什么要这样设计?这样更加人性化,用户在购买书籍的时候,用户能够查看相关分类后的图书,而不是全部图 ...
-
liunx 系统调用 getopt() 函数
命令行参数解析函数 -- getopt() getopt()函数声明如下: #include <unistd.h>int getopt(int argc, char * const arg ...
-
2017ecjtu-summer training #6 Gym 100952D
D. Time to go back time limit per test 1 second memory limit per test 256 megabytes input standard i ...
-
ORACLE 本地数据库存储过程 调用远程数据库存储过程
废话少说,直接切入主题 步骤1:建立一个远程数据库的连接服务名 D:\oracle\ora92\network\admin\tnsnames.ora 添加如下代码: SDEC = (DESC ...
-
判断以xx开头的字符串
public static void main(String[] args) { String str = "EAN_13,1534651"; String strHttp = & ...