哈夫曼树编码以及构造--数据结构实验三

基于哈夫曼树的数据压缩算法
 

描述

 

输入一串字符串,根据给定的字符串中字符出现的频率建立相应哈夫曼树,构造哈夫曼编码表,在此基础上可以对待压缩文件进行压缩(即编码),同时可以对压缩后的二进制编码文件进行解压(即译码)。



 

输入

多组数据,每组数据一行,为一个字符串(只考虑26个小写字母即可)。当输入字符串为“0”时,输入结束。

输出

每组数据输出2n+3行(n为输入串中字符类别的个数)。第一行为统计出来的字符出现频率(只输出存在的字符,格式为:字符:频度),每两组字符之间用一个空格分隔,字符按照ASCII码从小到大的顺序排列。第二行至第2n行为哈夫曼树的存储结构的终态(形如教材139页表5.2(b),一行当中的数据用空格分隔)。第2n+1行为每个字符的哈夫曼编码(只输出存在的字符,格式为:字符:编码),每两组字符之间用一个空格分隔,字符按照ASCII码从小到大的顺序排列。第2n+2行为编码后的字符串,第2n+3行为解码后的字符串(与输入的字符串相同)。

输入样例 1 

aaaaaaabbbbbccdddd
aabccc
0

输出样例 1

a:7 b:5 c:2 d:4
1 7 7 0 0
2 5 6 0 0
3 2 5 0 0
4 4 5 0 0
5 6 6 3 4
6 11 7 2 5
7 18 0 1 6
a:0 b:10 c:110 d:111
00000001010101010110110111111111111
aaaaaaabbbbbccdddd
a:2 b:1 c:3
1 2 4 0 0
2 1 4 0 0
3 3 5 0 0
4 3 5 2 1
5 6 0 3 4
a:11 b:10 c:0
111110000
aabccc

问题:一开始不明白最后状态是什么意思,然后就是不知道怎么得出a、b、c用0、1表示



这表示HuffmanCode是一个char**类型的代名词。

char*可以理解为指向一个字符串第一个字的指针。

char**可以理解为字符串数组,
char **a = new char* [10];
for (int i = 0; i < 10; i++) a[i] = new char [30];
这就创建了一个a,a[n]代表第n+1个字符串,a[n][m]表示第n+1个字符串的第m+1个字符。
#include<iostream>
#include<cstring>
#include<stdio.h>
#include<string>
#define MAX 100
using namespace std;

int coun[26]; //频率 
char saveletter[26];//存字母 
char temp[MAX];//暂存被译码串 

typedef struct htnode
{
    int weight;
    int lchild, rchild, parent;
    char data;
    int frequency;//出现频率 
}*huftree;

typedef char **hufcode;

void select(huftree &hf, int x, int &s1, int &s2)//在叶子结点里找最小的两个 
{
    int min = 999, cmin = 999;//最小值和次小值 
    int i = 1;
    while (i <= x)
    {
        if (hf[i].parent == 0)
        {
            if (hf[i].weight < min)//寻找权值最小 
            {
                min = hf[i].weight;
                s1 = i;
            }
            i++;
        }
        else
            i++;
    }
    int flag = s1;
    i = 1;
    while (i <= x)
    {
        if (hf[i].parent == 0)
        {
            if ((hf[i].weight > min && hf[i].weight < cmin) || (hf[i].weight == min && flag != i))//找次小值 
            {
                cmin = hf[i].weight;
                s2 = i;
            }
            i++;
        }
        else
            i++;
    }

}

void Create(huftree &hf, int n)//叶子为n的哈树有2n-1个结点 
{
    int m = 2 * n - 1, s1 = 0, s2 = 0;

    if (n <= 1) return;

    hf = new htnode[m + 1];//0号单元不用

    for (int i = 1; i <= m; i++)//都初始化为0 
    {
        hf[i].parent = 0;
        hf[i].rchild = 0;
        hf[i].lchild = 0;
        hf[i].data = saveletter[i - 1];//字母
    }

    for (int i = 1; i <= n; i++)
        hf[i].weight = coun[i - 1];//输入权值 

    for (int i = n + 1; i <= m; i++)//前n个为叶子,后面需要构建 
    {
        select(hf, i - 1, s1, s2);//选择最小的两个节点,返回序号 
        hf[s1].parent = i;
        hf[s2].parent = i;//结点双亲变为i
        hf[i].lchild = s1;
        hf[i].rchild = s2;//i的左右孩子
        hf[i].weight = hf[s1].weight + hf[s2].weight;
        //i权值更改 
    }

}


void Show(huftree &hf, int x)
{
    for (int i = 1; i <= 2 * x - 1; i++)
    {
        cout << i << " ";
        cout << hf[i].weight << " " << hf[i].parent << " " << hf[i].lchild << " " << hf[i].rchild << endl;
    }
}

void count(char str[], huftree &hf, int &n)//出现频率 ,字母个数 

{
    int num[26];
    char ch;
    int i = 0, j = 0;
    memset(num, 0, sizeof(num));
    while (str[i] != '\0')
    {
        j = str[i] - 97;
        num[j]++;
        i++;
    }
    j = 0;
    for (i = 0; i < 26; i++)
    {
        if (num[i] != 0)
        {
            saveletter[j] = char(i + 97);
            coun[j] = num[i];
            j++;
        }
    }
    n = j;
    for (int i = 0; i < n; i++)
    {
        if (i == n - 1)
            cout << saveletter[i] << ":" << coun[i];
        else
            cout << saveletter[i] << ":" << coun[i] << " ";
    }
    cout << endl;
}

void hfcode(huftree &hf, hufcode &hc, int n)
{
    char *cd;
    int start = 0, c, f;
    hc = new char*[n + 1];//编码表 
    cd = new char[n];//每个字符的编码一定小于n 
    cd[n - 1] = '\0';
    for (int i = 1; i <= n; i++)
    {
        start = n - 1;
        c = i;
        f = hf[i].parent;

        while (f != 0)//不是根节点
        {
            start--;

            if (hf[f].lchild == c)
                cd[start] = '0';
            else
                cd[start] = '1';

            c = f;//向上回溯 
            f = hf[f].parent;
        }
        hc[i] = new char[n - start];
        strcpy(hc[i], &cd[start]);//把临时空间的编码复制到编码表中 
    }
    delete cd;

    int i, j, z = 0;
    for (j = 1; j <= n; j++)//输出字母编码 
    {
        if (j == n)
            cout << saveletter[j - 1] << ":" << hc[j];
        else
            cout << saveletter[j - 1] << ":" << hc[j] << " ";
    }
    cout << endl;
}
void transtonum(huftree &hf, hufcode &hc, int n, char str[])
{
    for (int i = 0; str[i] != '\0'; i++)
        for (int j = 1; j <= n; j++)
            if (str[i] == saveletter[j - 1])
            {
                cout << hc[j];
                strcat(temp, hc[j]);
            }

    cout << endl;
}
void transtoletter(huftree &hf, hufcode &hc, int n)
{
    int i = 2 * n - 1;
    int j = 0;

    while (temp[j] != '\0')
    {
        if (temp[j] == '0')
            i = hf[i].lchild;   //左孩子
        else if (temp[j] == '1')
            i = hf[i].rchild;    //右孩子
        if (hf[i].lchild == 0)
        {
            cout << hf[i].data;
            i = 2 * n - 1;
        }
        j++;   //无论是否找到叶子节点都读取下一个编码串字符
    }
    cout << endl;
}

int main()
{

    while (1)
    {
        huftree hf;
        hufcode hc;
        int n;
        char str[MAX];
        scanf("%s", &str);
        if (str[0] == '0')
            break;
        count(str, hf, n);
        Create(hf, n);
        Show(hf, n);
        hfcode(hf, hc, n);
        transtonum(hf, hc, n, str);
        transtoletter(hf, hc, n);
        memset(coun, 0, sizeof(coun));
        memset(saveletter, '\0', sizeof(saveletter));
        memset(temp, '\0', sizeof(temp));
        delete hf;
        delete hc;
    }
    return 0;
}

 

posted @ 2019-11-18 20:17  Tomorrow1126  阅读(2899)  评论(0编辑  收藏  举报