先看看代码
#define _CRT_SECURE_NO_WARNINGS 1
#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);
Status WordCount(char *data, NumCount *paraCnt);
Status Show(NumCount *paraCnt);
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(int argc, char** argv) {
char data[MaxSize];
NumCount Cntarray;
ReadData(data);
WordCount(data, &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)
{
//打开文件读入数据
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;
}
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) // 没有记录,则新增
{
paraCnt->count[paraCnt->length].ch = data[i];
++paraCnt->count[paraCnt->length].cnt;
++paraCnt->length;
}
}
return OK;
}
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;
}
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) // 初始化
{
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;
}
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;
}
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;
return 1;
}
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; j <= 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;
}
Status Decode(HuffmanTree HT, int length)
{
char codetxt[100001];
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;
}
运行结果如下
哈夫曼树是一种特殊的二叉树,它的每个叶子节点都有一个权值,而且权值越大的叶子节点越靠近根节点。哈夫曼树的构造方法是:
- 首先将所有的叶子节点按照权值从小到大排序,然后每次取出最小的两个节点,合并成一个新的节点,新节点的权值为两个子节点的权值之和,然后将新节点插入到原来的序列中,重复这个过程直到只剩下一个节点,这个节点就是哈夫曼树的根节点。
- 其次,从根节点开始遍历这棵树,并且给每条边赋予一个bit值:左边为0,右边为1。最后我们可以得到每个叶子节点对应的编码:从根节点到叶子节点经过的所有边上的bit值组成。
以我自己的理解来说,就是一直取最小的合并生成树,如此反复能将原始数据变成权重最小的一颗二叉树。算法运行逻辑简单易懂,但对我来说进行代码实现还是有难度。