对于给定的权值进行Huffman编码

Huffman的特点是每次取出优先队列，也即是堆中最小的两个节点链接成一个新的节点，将新的节点放回到堆中。

Huffman树每一个叶子都是一个字符。非叶子节点只是我们生成的节点，不是我们需要编码的字符。

#include<iostream>
#include<stdlib.h>
#include<string.h>
#include<queue>
#define LEN 20
using namespace std;
typedef char ElementType;
typedef int Location;


struct cmp{
        template<typename T, typename U>
        bool operator()(T const& left, U const &right) {
            if (left.first > right.first) return true;
            return false;
        }
};






struct LetterNode{														//Letter为Huffman树节点 
	int weight;
	ElementType Data;
	char*  huffman_code;
	int Left;
	int Right;
}; 
typedef LetterNode *Letter;

int LetterWeight[] = {186,64,13,22,32,103,21,15,47,57,1,5,32,20,57,63,15,1,48,51,80,23,8,18,1,16,1};
ElementType LetterData[] = {' ','a','b','c','d','e','f','g','h','i','j','k','l','m','n','o','p','q','r','s','t','u','v','w','x','y','z'};



//1.建造Huffman树 
Location BuildLetterHuffman(int LetterWeight[], ElementType LetterData[],int LetterLength,LetterNode LetterHuTree[])
{
	Letter L;
	int Le,Ri;
	int LeftNode,RightNode;
	int Count ;
	Count = LetterLength;
	for(int i = 0; i < LetterLength; i++)
	{
		L = (Letter)malloc(sizeof(LetterNode));
		L->Data = LetterData[i];
		L->weight = LetterWeight[i];
		L->Left = -1;											//已经得到了字母表 
		L->Right = -1;
		L->huffman_code = (char*)malloc(10 * sizeof(char));
		LetterHuTree[i] = *L;
	}
	
	//将权值数组放入堆中
	priority_queue<pair<int, int>, vector<pair<int, int> >, cmp>  MinHeap;
	for(int i = 0; i <LetterLength ; i++)
	{
		pair<int, int> LEW(LetterWeight[i],i);
		MinHeap.push(LEW);
	} 
	
    //注意此处最小堆中应没有节点足够组成新节点了
	for(int i = 0;(MinHeap.size() != 0) && (MinHeap.size() != 1); i++)						//在此处循环条件要注意！！！！因为有可能已经有子节点的树会再次链接！！！！！ 
	{
		//1.选取出数组中最小的两个元素
		LeftNode = MinHeap.top().second;
		MinHeap.pop();
		RightNode = MinHeap.top().second;
		MinHeap.pop();
		
		
		//2.构建新的Huffman节点 
		 L = (Letter)malloc(sizeof(LetterNode));
		 L->weight = LetterHuTree[LeftNode].weight + LetterHuTree[RightNode].weight;
		 L->Left = LeftNode;
		 L->Data = 'M'; 
		 L->huffman_code = (char*)malloc(10 * sizeof(char));
		 L->Right = RightNode;
		 LetterHuTree[Count] = *L;
		 
		 
		 //3.将新生Huffman节点放入堆中 
		 pair<int, int> LEW(LetterHuTree[Count].weight,Count);					
		 MinHeap.push(LEW);
		 Count++;
	}
	Count--;
	return Count;									//此时L为构建的最后节点，即为Huffman树的根节点 
}


//2.对已经构建好的霍夫曼树的节点进行编码 
void HuffmanCode(Letter L,Location location,string code)					
{	
	string NowCode = code;
	strcpy(L[location].huffman_code , NowCode.c_str());
//	cout <<L[location].Data << "的Huffman编码是" <<L[location].huffman_code <<endl;
	if(L[location].Left != -1)
	{
		code = code + "0";
		HuffmanCode(L,L[location].Left,code);
	}
	if(L[location].Right != -1)
	{
		NowCode = NowCode + "1";
		HuffmanCode(L,L[location].Right,NowCode);
	}	
} 


//3.打印出编辑完成的Huffman编码 
void PrintHuCode(Letter L,int LetterLength)							//打印Huffman数组元素中的编码 
{
	cout << "该权值下的Huffman编码为：\n";
	for(int i = 0; i < LetterLength; i++)
	{
		cout <<L[i].Data<< "的编码为"<<L[i].huffman_code <<endl;
//		printf("%c的编码为%s\n",L[i].Data,L[i].huffman_code);
	}
}

int main()
{
//	int a = 0;
//	for(int i = 0 ; i< 27;i++)
// 	{
// 		a +=  LetterWeight[i];
// 	} 
// 	cout <<"a的值为：\n";
// 	cout << a;
	 
	LetterNode LetterHuTree[100];
	Location HuLocation = BuildLetterHuffman(LetterWeight, LetterData,27,LetterHuTree);
	HuffmanCode(LetterHuTree,HuLocation," ");
	PrintHuCode(LetterHuTree,27);
	
	return 0;
}