哈夫曼树及其应用

1.定义

  假设有m个权值w1,w2````wm,构造一颗含有n个叶子结点的二叉树，每个叶子节点的权为wi,其中带权路径长度最短的二叉树成为最优二叉树或者哈夫曼树。

2.哈夫曼树的基本概念

路径：从树中一个结点到另一个结点之间的分支构成这两个结点间的路径。

结点的路径长度：两个结点路径上的分支数。

数的路径长度：从树根到每一个节点的路径长度之和。 记作：TL

结点数目相同的二叉树中，完全二叉树是路径长度最短的二叉树。

权：将树中结点赋给一个存有某种含义的数值，则这个数值称为该结点的权。

结点的带权路径长度：从根节点到该结点之间的路径长度与该结点的权的乘积。

树的带权路径长度：树中所有的叶子结点的带权路径长度之和。

3.创建哈夫曼树代码

Status CreateHuffmanTree(HuffmanTree &HT, int length, NumCount cntarray)
{
	if(length <= 1) return ERROR;
	int s1, s2;
	int m = length * 2 - 1;
	HT = new HTNode[m + 1];
	for(int i = 1; i <= m; i++)
	{
		HT[i].parent = 0;
		HT[i].lchild = 0;
		HT[i].rchild = 0;
	}
	
	for(int i = 1; i <= length; i++)
	{
		HT[i].data = cntarray.count[i - 1].ch;
		HT[i].weight = cntarray.count[i - 1].cnt;
	}
	
	for(int i = length + 1; i <= m; i++)
	{
		select(HT, i - 1, &s1, &s2);
		HT[s1].parent = i;
		HT[s2].parent = i;
		HT[i].lchild = s1;
		HT[i].rchild = s2;
		HT[i].weight = HT[s1].weight + HT[s2].weight;
	}
	return OK;
}//of CreateHuffmanTree

4.创建哈夫曼编码的代码

Status CreateHuffmanCode(HuffmanTree HT, HuffmanCode &HC, int length)
{
	HC = new HCode[length + 1];
	char *cd = new char[length];
	cd[length - 1] = '\0';
	int c, f, start;
	for(int i = 1; i <= length; i++)
	{
		start = length - 1;
		c = i;
		f = HT[c].parent;
		while (f != 0)
		{
			start--;
			if(HT[f].lchild == c)
			{
				cd[start] = '0';
			}
			else
			{
				cd[start] = 1;
			}
			c = f;
			f = HT[c].parent;
		}
		HC[i].str = new char[length - start];
		HC[i].data = HT[i].data;
		strcpy(HC[i].str,&cd[start]); 
	}
	delete cd;
}//of CreateHuffmanCode

5.总代码

#include<iostream>
#include<fstream>
#include<string.h>
using namespace std;

#define MaxSize 1024 
#define OK 1
#define ERROR 0
typedef int Status;

typedef struct wordcnt
{
	char ch;
	int cnt = 0;
}Count;

typedef struct NumCount
{
	Count count[MaxSize];
	int length = 0;
}NumCount;

typedef struct HTree
{
	char data;
	int weight;
	int parent, lchild, rchild;
}HTNode, *HuffmanTree;

typedef struct HCode
{
	char data;
	char* str;
}*HuffmanCode;

Status ReadData(char *source);  // 读入文件 
Status WordCount(char *data,NumCount *paraCnt); // 统计次数 
Status Show(NumCount *paraCnt);   // 展示次数 
Status CreateHuffmanTree(HuffmanTree &HT,int length,NumCount cntarray);  // 创建哈夫曼树 
Status select(HuffmanTree HT,int top,int *s1,int *s2);  // 选择权重最小的两个节点 
Status CreateHuffmanCode(HuffmanTree HT,HuffmanCode &HC,int length);  // 创建哈夫曼编码 
Status Encode(char *data,HuffmanCode HC,int length);  // 将读入的文件编码，写到txt文件 
Status Decode(HuffmanTree HT,int length);  //读入编码文件，解码 

int main()
{
	char data[MaxSize];
	NumCount Cntarray;
	ReadData(data);
	WordCount(data, &Cntarray);
	//Show(&Cntarray);
	HuffmanTree tree;
	CreateHuffmanTree(tree, Cntarray.length, Cntarray);
	HuffmanCode code;
	CreateHuffmanCode(tree, code, Cntarray.length);
	Encode(data, code, Cntarray.length);
	Decode(tree, Cntarray.length);
	cout<<"Please view the generated TXT file to check the result"<<endl;
	return 0;
}
Status ReadData(char *source)
{
	ifstream infile;
	infile.open("in.txt");
	cout<<"Reading..."<<endl;
	cout<<"the input file is:"<<endl;
	infile.getline(source, MaxSize);
	cout<<source<<endl;
	infile.close();
	cout<<endl;
	return OK;
}//of ReadData

Status WordCount(char *data, NumCount *paraCnt)
{
	int flag;
	int len = strlen(data);
	for(int i = 0; i < len; i++)
	{
		flag = 0;
		for(int j = 0; j < paraCnt->length; j++)
		{
			if(paraCnt->count[j].ch == data[i])
			{
				paraCnt->count[j].cnt++;
				flag = 1;
				break;
			}
		}
		if(!flag)
		{
			paraCnt->count[paraCnt->length].ch = data[i];
			paraCnt->count[paraCnt->length].cnt++;
			paraCnt->length++;
		}
	}
	
	return OK;
}//of WordCount

Status Show(NumCount *paraCnt)
{
	cout<<"the length is "<<paraCnt->length<<endl;
	for(int i = 0; i < paraCnt->length; i++)
	{
		cout<<"The character "<<paraCnt->count[i].ch<<" appears "<<paraCnt->count[i].cnt<<endl;
	}
	cout<<endl;
	return OK;
}//of Show

Status CreateHuffmanTree(HuffmanTree &HT, int length, NumCount cntarray)
{
	if(length <= 1) return ERROR;
	int s1, s2;
	int m = length * 2 - 1;
	HT = new HTNode[m + 1];
	for(int i = 1; i <= m; i++)
	{
		HT[i].parent = 0;
		HT[i].lchild = 0;
		HT[i].rchild = 0;
	}
	
	for(int i = 1; i <= length; i++)
	{
		HT[i].data = cntarray.count[i - 1].ch;
		HT[i].weight = cntarray.count[i - 1].cnt;
	}
	
	for(int i = length + 1; i <= m; i++)
	{
		select(HT, i - 1, &s1, &s2);
		HT[s1].parent = i;
		HT[s2].parent = i;
		HT[i].lchild = s1;
		HT[i].rchild = s2;
		HT[i].weight = HT[s1].weight + HT[s2].weight;
	}
	return OK;
}//of CreateHuffmanTree

Status select(HuffmanTree HT, int top, int *s1, int *s2)
{
	int min = INT_MAX;
	for(int i = 1; i <= top; i++)
	{
		if(HT[i].weight < min && HT[i].parent == 0)
		{
			min = HT[i].weight;
			*s1 = i;
		}
	}
	
	min = INT_MAX;
	for(int i = 1; i <= top; i++)
	{
		if(HT[i].weight < min && i != *s1 && HT[i].parent == 0)
		{
			min = HT[i].weight;
			*s2 = i;
		}
	}
	return OK;
}//of select

Status CreateHuffmanCode(HuffmanTree HT, HuffmanCode &HC, int length)
{
	HC = new HCode[length + 1];
	char *cd = new char[length];
	cd[length - 1] = '\0';
	int c, f, start;
	for(int i = 1; i <= length; i++)
	{
		start = length - 1;
		c = i;
		f = HT[c].parent;
		while (f != 0)
		{
			start--;
			if(HT[f].lchild == c)
			{
				cd[start] = '0';
			}
			else
			{
				cd[start] = 1;
			}
			c = f;
			f = HT[c].parent;
		}
		HC[i].str = new char[length - start];
		HC[i].data = HT[i].data;
		strcpy(HC[i].str,&cd[start]); 
	}
	delete cd;
}//of CreateHuffmanCode

Status Encode(char *data, HuffmanCode HC, int length)
{
	ofstream outfile;
	outfile.open("code.txt");
	for(int i = 0; i < strlen(data); i++)
	{
		for(int j = 1; i <= length; j++)
		{
			if(data[i] == HC[j].data)
			{
			outfile<<HC[j].str;
			}
		}
	}
	outfile.close();
	cout<<"the code txt has been written"<<endl;
	cout<<endl;
	return OK;
}//of Encode

Status Decode(HuffmanTree HT,int length)
{
	char codetxt[MaxSize*length];
	ifstream infile;
	infile.open("code.txt");
	infile.getline(codetxt,MaxSize*length);
	infile.close();
	
	ofstream outfile;
   	outfile.open("out.txt");
	
	int root = 2*length-1;  
	for(int i = 0;i < strlen(codetxt);++i)
	{
		if(codetxt[i] == '0') root = HT[root].lchild;  
		else if(codetxt[i] == '1') root = HT[root].rchild; 
		if(HT[root].lchild == 0 && HT[root].rchild == 0)   
		{
			outfile<<HT[root].data;
			root = 2*length-1;
		}
	}
	outfile.close();
	cout<<"the output txt has been written"<<endl;
	cout<<endl;
	return OK;
}