谁能说下把正则式转换成NFA应该怎么建立数据结构，就是实现Thomson算法的数据结构

[weiyiabout]

如题
谢谢

[C1053710211]

节点就是case，每个case后的statement 就是一个状态，case之间的转化模拟有穷自动机
举个例子，这个是以前做过的题
给定一文本文件,存储内容如下：
ID,TYPE
a*bc,001
a*v,002
a?cd,003
a*bbc,004
其中，ID列为正则表达式。另外，给定一字符串s="abc",现在需要求出s所对应的TYPE,

#include <stdio.h>
#include <string>
#include <string.h>
#include <stdlib.h>
#include <iostream>
#include <fstream>
#define LEN 1<<16
using namespace std;
char s[LEN],c,*lexeme_beginning;
char *token_beginning = s,*forward = s ;
string token;
int start,state,step,symbol_lex;
char nextchar()
{
     return *forward++;          
}
void retract(int pos)
{
     forward--;     
}
string token_nexttoken()
{
     while(1)
     {
          switch(state)
          {
               case 0: 
                    c=nextchar();
                    if(c=='a') state = 8; 
                    else if(c=='b') state = 1;
                    else if(c=='c') state = 2;
                    else if(c=='v') state = 3;
                    else state =30;         
                    break;
               case 1: 
                    c=nextchar(); 
                    if(c=='c') state = 4;
                    else if(c=='b') state = 5;
                    else state = 30; 
                    break;
               case 2: 
                    c=nextchar();
                    if(c=='d') state = 7;
                    else state = 30;
                    break;
               case 3:
                    if(*forward=='\0') return "002";
                    else state = 30;
                    break;
               case 4:
                    if(*forward=='\0') return "001";
                    else state = 30;
                    break;
               case 5:
                    c=nextchar();
                    if(c=='c') state = 6;
                    else state = 30;
                    break;
               case 6:
                    if(*forward=='\0') return "004";
                    else state = 30;
                    break;
               case 7:
                    if(*forward=='\0') return "003";
                    else state = 30;
                    break;
               case 8: 
                    c=nextchar();
                    if(c=='a') state = 9;
                    else if(c=='b') state = 1;
                    else if(c=='c') state = 2;
                    else if(c=='v') state = 3; 
                    else state = 30;
                    break; 
               case 9:
                    c=nextchar();
                    if(c=='a') state = 9;
                    else if(c=='b') state = 1;
                    else if(c=='v') state = 3; 
                    else state = 30;
                    break;
               case 30:     
                    return "[false]";
          }        
     }     
}
void Init()
{
}
int main()
{
    int len;
    string tmp;
    while(scanf("%s",s)!=EOF)
    {
          state = 0;
          lexeme_beginning = s;
          forward = s;
          len = strlen(s);
          while(forward!=s+len)
          {            
                start = 0;
                state = 0;  
                token_beginning = lexeme_beginning;
                tmp = token_nexttoken();
                cout<<tmp.c_str()<<endl;
                if(state == 30 ) break;
          }
    }
    system("pause");
    return 0;    
}

[C1053710211]

似乎不用数据结构，switch-case语句就够了

[weiyiabout]

到NFA的各个结点，得用某个数据结构来表示吧？

[knowledge_Is_Life]

有点难度哦

[lyt31]

这种方法不可能构造正则表达式引擎。。
只能判断某个字符串是否匹配某个【预先给定】的正则表达式