关于Huffman编码和KMP的知识请看我的这两篇文章:Huffman编码、KMP。
#include<bits/stdc++.h>
using namespace std;
/*
Huffman解码:
为什么把这个单独提出来,是因为我想把它和KMP结合起来使用。
*/
class Node
{
public:
char c; //被编码的字符
string code; //字符的编码
int *next; //编码的next数组
};
int *getNext(string p)
{
int l = p.length(), k = -1, i = 0;
int *next = new int[l];
next[0] = -1;
while(i < l - 1)
{
while(k >= 0 && p[k] != p[i])
k = next[k];
i++;
k++;
if(p[i] == p[k])
next[i] = next[k];
else
next[i] = k;
}
return next;
}
int KMPMatching(int *next, string p, string r)
{
//在 p 中寻找 r
int i = 0, j = 0; //i指向p,j指向r
int pl = p.length(), rl = r.length();
if(pl < rl)
return -1;
while(i < pl && j < rl)
{
if(j == -1 || p[i] == r[j])
i++, j++;
else
j = next[j];
}
if(j >= rl) //说明r被扫描完了,即找到匹配的了
return (i - rl);
else
return -1;
}
int main()
{
/*
不想把Huffman树的构建再复制一遍,所以直接使用了一些以编码的字符
*/
Node *arr = new Node[6];
arr[0].c = 'c'; arr[0].code = "00"; arr[0].next = getNext(arr[0].code);
arr[1].c = 'b'; arr[1].code = "01"; arr[1].next = getNext(arr[1].code);
arr[2].c = 'd'; arr[2].code = "100"; arr[2].next = getNext(arr[2].code);
arr[3].c = 'f'; arr[3].code = "1010"; arr[3].next = getNext(arr[3].code);
arr[4].c = 'e'; arr[4].code = "1011"; arr[4].next = getNext(arr[4].code);
arr[5].c = 'a'; arr[5].code = "11"; arr[5].next = getNext(arr[5].code);
string str = "11010010010111010";
int i = 0, j, l = str.length();
while(i < l)
{
int flag = 0;
for(j = 0; j < 6; j++)
{
int tl = arr[j].code.length();
if(KMPMatching(arr[j].next, str.substr(i, tl), arr[j].code) == 0)
{
cout << arr[j].c;
i += tl;
flag = 1;
break;
}
}
if(flag == 0)
{
cout << "待解码的字符串有问题!";
break;
}
}
return 0;
}
/*
字符 权重 编码
c 4 00
b 5 01
d 3 100
f 1 1010
e 2 1011
a 6 11
*/