文本抽取函数 比正则表达式好用

#include<iostream>
#include<fstream>
#include<cstring>
#include<algorithm>
#include<dirent.h>
#include<stdio.h>
#include<errno.h>
#include<malloc.h>
using namespace std;
							
#define MAXN 500
#define _match(a,b) ((a)==(b))
#define N_NAME 200
#define N_FIX 500
#define buffersize 10000000


struct Attribute			//define attribute class
{
	char name[N_NAME];
	char prefix[N_FIX];
	char start[N_FIX];
	char suffix[N_FIX];
}att;


struct att_text
{
	int att_id;
	int s,e;
}ex_text[100];
/******************************************************************************/
/*字符串模式匹配*/
int pat_match(int ls,char str[],int lp,char pat[]){
	int fail[MAXN]={-1},i=0,j;


	for (j=1;j<lp;j++)
	{
		for (i=fail[j-1];i>=0&&!_match(pat[i+1],pat[j]);i=fail[i]);
		fail[j]=(_match(pat[i+1],pat[j])?i+1:-1);
	}
	for (i=j=0;i<ls&&j<lp;i++)
	{
		if (_match(str[i],pat[j]))
			j++;
		else if (j)
			j=fail[j-1]+1,i--;
	}
	return j==lp?(i-lp):-1;
}
/********************************************************************************/
/*多重自动机，属性抽取*/
int multi_ex(char org_text[])
{
	int text_len = strlen(org_text);
	int i=0,ls,x,lps,lpe,lps2;
	int ex_num=0;//the number of attributes which have been founded
	ls=text_len;
	lps=strlen(att.prefix);
	lps2=strlen(att.start);
	lpe=strlen(att.suffix);
    while(1)
	{
		x=pat_match(ls,org_text+(text_len-ls),lps,att.prefix);
		if(x==-1)
			break;
		ls=ls-x-lps;
		x=pat_match(ls,org_text+(text_len-ls),lps2,att.start);
		if(x==-1)
			continue;
		else
			ex_text[ex_num].s=text_len-ls+x+lps2;
		ls=ls-x-lps2;
		x=pat_match(ls,org_text+(text_len-ls),lpe,att.suffix);
		if(x==-1)
			continue;
		else
			ex_text[ex_num].e=text_len-ls+x;
		ex_text[ex_num].att_id=i;
		ex_num++;
		ls=ls-x-1;
	}
	return ex_num;
}
/*********************************************************************************/
/*
参数	:
	org_text:待抽取的文本内容
	name	:抽取的属性名
	prefix	:抽取串开始的位置
	start	:抽取的开始位置
	suffix	:抽取的结束位置
返回值:
	result	:抽取的结果，每条记录以'\n'分割
*/
string extract(char org_text[], char name[], char prefix[] ,char start[], char suffix[])
{
	//获得待提取的属性信息
	strcpy(att.name,name);
	strcpy(att.prefix,prefix);
	strcpy(att.start,start);
	strcpy(att.suffix,suffix);
	
	int text_len = strlen(org_text);
	int result_num = multi_ex(org_text);
	
	string result;
	//输出提取结果
	int i=0,j;
	for(i=0;i<result_num;i++)
	{
		for(j=ex_text[i].s;j<ex_text[i].e;j++)
		{
			//fprintf(stdout,"%c",org_text[j]);
			result+=org_text[j];
		}
		//fprintf(stdout,"\n");
		result+='\n';
	}
    return result;
}
/*
int main()
{
	char name[100] = "title";
	char prefix[100] = "<title>">";
	char start[100] = "";
	char suffix[100] = "</title>";
	char org_text[buffersize];
   	 memset(org_text,0,sizeof(org_text));
	FILE *fp;
	fp=freopen("1.html","r",stdin);
   	if(fp==NULL)
		perror("failed fopen");
    	fread(org_text,sizeof(char),buffersize,fp);
	fclose(fp);
	string result;
	result = extract(org_text,name, prefix, start, suffix);
	cout<<result;
	return 0;
}


*/