htmlparse.h

html文件编码转换与解析

/* This is My HtmlParser's first protype.
 * author: Feng Zeyue
 */
#include <iostream>
#include <string>
#include <stdlib.h>

enum tagtype {
	other = 0,
	title = 1,
	anchor = 2,
	titleend = 3,
	anchorend = 4,
	script = 5,
	fileend = 6,
	style = 7,
	styleend = 8,
	comment = 9
};

int parseposition = 0;
int endflag;
int bufparseflag = 4;
string text;
string tag;

using namespace std;

//class definition
class HtmlParser {
public:
	HtmlParser();
	HtmlParser(string input);
	~HtmlParser();
	string GetAbusoluteURL(string input);
	bool SetBaseURL(string URL);
	bool Parse();
	tagtype ParseTag();
	bool ParseTitle();
	bool ParseA();
	bool ParseText();
	bool ParseScript();
	bool ParseStyle();
	string GetURL();
	string GetBaseURL();
private:
	string baseURL;
	string URL;
};

//change the upper string to lower
bool MakeLower(string &s)
{
	int i = 0;
	while(s[i] != '\0')
	{
		if((s[i] >= 65) && (s[i] <= 90)){
			s[i] += 32;
		}
		i++;
	}
	return true;
}

//check whether arrive at the end of the buffer
bool IsBufEnd(int &i) {
	if (i == 1023) {
		if(endflag == 1)
		{
			return false;
		}
		else{
			bufparseflag = (bufparseflag + 1) % 5;
			endflag = readbuf();
		}
		//		cout<<buf[flag];
	}
	i = (i + 1) % 1024;
	return true;
}

//Remove the target attribute from <a href..>

string RemoveTarget(string &url){
	int i = 0;
	string urlresult(url);
	while(url[i] != '\0')
	{
		if((url[i] == 't') && (url[i +1] =='a') && (url[i + 2] == 'r') && (url[i + 3] == 'g') && (url[i + 4] == 'e') && (url[i + 5] == 't')){
			int j = i + 6;
			while(url[j] == ' ')
			{
				j ++;
			}
			if(url[j] == '=')
			{
				urlresult.clear();
				j = 0;
				while(j < i)
				{
					urlresult += url[j];
					j++;
				}
				return urlresult;
			}
		}
		i++;
	}
	return urlresult;
}