page.cpp

此源码功能是捕获网页中的链接并进行分析

/*Page handling
 */

#include <iostream>
#include <string>
#include <map>
#include <vector>
#include <iterator>
#include "Url.h"
#include "Page.h"
#include "StrFun.h"
#include "h/code.h"

CPage::CPage()
{
	m_nStatusCode = 0;
	m_nContentLength = 0;
	m_sLocation = "";
	m_bConnectionState = false;
	m_sContentEncoding = "";
	m_sContentType = "";
	m_sCharset = "";
	m_sTransferEncoding = "";

	m_sContentLinkInfo = "";
        m_sLinkInfo4SE = "";
        m_sLinkInfo4History = "";

        m_sContentNoTags = "";
	m_nRefLink4SENum = 0;
	m_nRefLink4HistoryNum = 0;
        m_eType = PLAIN_TEXT;

	for(int i=0; i< MAX_URL_REFERENCES; i++ ){
		m_RefLink4SE[i].link = NULL;
		m_RefLink4SE[i].anchor_text = NULL;
		m_RefLink4SE[i].strCharset = "";

		if(i < MAX_URL_REFERENCES/2){
			m_RefLink4History[i].link = NULL;
		}
	}

}

CPage::CPage( string strUrl, string strLocation, char* header, char* body, int nLenBody)
{
	//assert( header != NULL );
	//assert( body != NULL );
	//assert( nLenBody > 0 );

	// CPage();
	m_nStatusCode = 0;
	m_nContentLength = 0;
	m_sLocation = "";
	m_bConnectionState = false;
	m_sContentEncoding = "";
	m_sContentType = "";
	m_sCharset = "";
	m_sTransferEncoding = "";

	m_sContentLinkInfo = "";
        m_sLinkInfo4SE = "";
        m_sLinkInfo4History = "";

        m_sContentNoTags = "";
	m_nRefLink4SENum = 0;
	m_nRefLink4HistoryNum = 0;
        m_eType = PLAIN_TEXT;

	for(int i=0; i< MAX_URL_REFERENCES; i++ ){
		m_RefLink4SE[i].link = NULL;
		m_RefLink4SE[i].anchor_text = NULL;
		m_RefLink4SE[i].strCharset = "";

		if(i < MAX_URL_REFERENCES/2){
			m_RefLink4History[i].link = NULL;
		}
	}


	m_sUrl = strUrl;
	m_sLocation = strLocation;
	m_sHeader = header;
	m_nLenHeader = strlen(header);

	m_sContent.assign(body, nLenBody);
	m_nLenContent = nLenBody;

}

CPage::~CPage()
{
}

void CPage::ParseHeaderInfo(string strHeader)
{
	GetStatusCode(strHeader);
	GetContentLength(strHeader);
	GetLocation(strHeader);
	GetConnectionState(strHeader);

        GetCharset(strHeader);

        GetContentEncoding(strHeader);
        GetContentType(strHeader);
	GetTransferEncoding(strHeader);
}

void CPage::GetStatusCode(string headerBuf)
{
	CStrFun::Str2Lower( headerBuf, headerBuf.length() );

	char *charIndex = strstr(headerBuf.c_str(), "http/");
	if (charIndex == NULL)
	{
		m_nStatusCode = -1;
		return;
	}

	while(*charIndex != ' '){
		charIndex++;
	}
	charIndex++;
	
	int ret = sscanf(charIndex, "%i", &m_nStatusCode);
	if (ret != 1)  m_nStatusCode = -1;
}

void CPage::GetContentLength(string headerBuf)
{
	CStrFun::Str2Lower( headerBuf, headerBuf.length() );

	char *charIndex = strstr(headerBuf.c_str(), "content-length");
	if (charIndex == NULL) return;

	while(*charIndex != ' '){
		charIndex++;
	}
	charIndex++;
	
	int ret = sscanf(charIndex, "%i", &m_nContentLength);
	if (ret != 1)  m_nContentLength = -1;
}

void CPage::GetLocation(string headerBuf)
{
	string::size_type pre_idx,idx;
	const string delims("\r\n");

	CStrFun::Str2Lower( headerBuf, headerBuf.length() );

	idx = headerBuf.find("location:");
	if (idx != string::npos)
	{
		pre_idx = idx + sizeof("location: ") -1;
		idx = headerBuf.find_first_of(delims, pre_idx );
		if (idx != string::npos)
		{
			m_sLocation = headerBuf.substr(pre_idx, idx - pre_idx);
		}
	}
}

void CPage::GetCharset(string headerBuf)
{
	string::size_type pre_idx,idx;
	const string delims(" \",;>");

	CStrFun::Str2Lower(headerBuf, headerBuf.size());

	idx = headerBuf.find("charset=");
	if( idx != string::npos) {
		m_sCharset = headerBuf.substr(idx + sizeof("charset=") -1);
		idx = m_sCharset.find("\r\n");
		if (idx!=string::npos) m_sCharset = m_sCharset.substr(0,idx);
		idx = m_sCharset.find("\n");
		if (idx!=string::npos) m_sCharset = m_sCharset.substr(0,idx);

		return;
	}

	headerBuf = m_sContent;
	headerBuf = headerBuf.substr(0,2024) ;
	CStrFun::Str2Lower( headerBuf, headerBuf.length() );
	idx = headerBuf.find("charset=");
	if (idx != string::npos)
	{
		pre_idx = idx + sizeof("charset=") -1;
		idx = headerBuf.find_first_of(delims, pre_idx );
		if(idx != string::npos){
			m_sCharset = headerBuf.substr(pre_idx, idx - pre_idx);
		}
		return;
	}

	headerBuf = m_sContent;
	char *s = (char*)headerBuf.c_str();
	RemoveTags(s);
	headerBuf = s;

	long iCntGBK= 0, iCntBig5 = 0, iCntOther =0 , i=0;
	//cout << "Content: " << headerBuf << endl;

	//unsigned char *ch = &(unsigned char)headerBuf[i];
	unsigned char *ch = (unsigned char*) &headerBuf[i];
    while (i< headerBuf.size()) {
		if(is_InASCII(ch)){
			i++;
			iCntGBK++;
		}else if(IS_GBK(ch)){
			i=i+2;
			iCntGBK++;
		}else if (IS_BIG5(ch)){
			i = i+2;
			iCntBig5++;
		}else{
			i ++;
			iCntOther ++;
		}	

		//ch= &(unsigned char)headerBuf[i];
		ch= (unsigned char*) &headerBuf[i];
	}	

	//if (Max(iCntGBK,iCntBig5,iCntOther) == iCntGBK)  m_sCharset = "GBK";
	if (iCntGBK >= iCntBig5 && iCntGBK >= iCntOther )
		m_sCharset = "gbk";
	if (iCntBig5 >= iCntGBK && iCntBig5 >= iCntOther )
		m_sCharset = "big5";
	
}

//remove tag and content of scripts, css, java, embeddedobjects, comments, etc
void CPage::RemoveTags(char *s)
{
	int intag;
	char *p, *q;

	if (!s || !*s) return;

	for (p=q=s, intag=0; *q; q++) {
		switch (*q){
		case '<':
			intag = 1;
			*p++ = ' ';
			break;
		case '>':
			intag = 0;
			break;
		default:
			if (!intag) {
				*p++ = *q;
			}
			break;
		}
	}

	*p = '\0';
}

void CPage::GetContentEncoding(string headerBuf)
{
	string::size_type pre_idx,idx;
	const string delims("\r\n");

	CStrFun::Str2Lower( headerBuf, headerBuf.length() );

	idx = headerBuf.find("content-encoding:");
	if (idx != string::npos)
	{
		pre_idx = idx + sizeof("content-encoding: ") -1;
		idx = headerBuf.find_first_of(delims, pre_idx );
		if (idx != string::npos)
		{
			m_sContentEncoding = headerBuf.substr(pre_idx, idx - pre_idx);
		}
	}
}

void CPage::GetConnectionState(string headerBuf)
{
	string::size_type pre_idx,idx;
	const string delims(";\r\n");

	CStrFun::Str2Lower( headerBuf, headerBuf.length() );

	idx = headerBuf.find("connection:");
	if (idx != string::npos)
	{
		pre_idx = idx + sizeof("connection: ") -1;
		idx = headerBuf.find_first_of(delims, pre_idx );
		if (idx != string::npos)
		{
			string str = headerBuf.substr(pre_idx, idx - pre_idx);
			//cout << "Connection state: " << str << endl;
			//if (str == "close") m_bConnectionState = false;
			if (str == "keep-alive") m_bConnectionState = true;
		}
	}
}

void CPage::GetContentType(string headerBuf)
{
	string::size_type pre_idx,idx;
	const string delims(";\r\n");

	CStrFun::Str2Lower( headerBuf, headerBuf.size() );

	idx = headerBuf.find("content-type:");
	if (idx != string::npos)
	{
		pre_idx = idx + sizeof("content-type: ") -1;
		idx = headerBuf.find_first_of(delims, pre_idx );
		if (idx != string::npos)
		{
			m_sContentType = headerBuf.substr(pre_idx, idx - pre_idx);
		}
	}
}


void CPage::GetTransferEncoding(string headerBuf)
{
	string::size_type pre_idx,idx;
	const string delims(";\r\n");

	CStrFun::Str2Lower( headerBuf, headerBuf.size() );

	idx = headerBuf.find("transfer-encoding:");
	if ( idx != string::npos)
	{
		pre_idx = idx + sizeof("transfer-encoding: ") -1;
		idx = headerBuf.find_first_of(delims, pre_idx );
		if(idx != string::npos)
		{
			m_sTransferEncoding = headerBuf.substr(pre_idx, idx - pre_idx);
		}
	}
}

/*
 * Filter spam links
 * If it is, return ture; otherwise false
 */
bool CPage::IsFilterLink(string plink)
{
	if( plink.empty() ) return true;
	if( plink.size() > URL_LEN ) return true;

	string link = plink, tmp;
	string::size_type idx = 0;

	
	CStrFun::Str2Lower( link, link.length() );

	// find two times following symbols, return false
	tmp = link;
	idx = tmp.find("?");
	if( idx != string::npos ){
		tmp = tmp.substr(idx+1);
		idx = tmp.find("?");
		if( idx != string::npos ) return true;
	}

	tmp = link;
	idx = tmp.find("-");
	if( idx != string::npos ){
		tmp = tmp.substr(idx+1);
		idx = tmp.find("+");
		if( idx != string::npos ) return true;
	}

	tmp = link;
	idx = tmp.find("&");
	if( idx != string::npos ){
		tmp = tmp.substr(idx+1);
		idx = tmp.find("&");
		if( idx != string::npos ) return true;
	}

	tmp = link;
	idx = tmp.find("//");
	if( idx != string::npos ){
		tmp = tmp.substr(idx+1);
		idx = tmp.find("//");
		if( idx != string::npos ) return true;
	}

	tmp = link;
	idx = tmp.find("http");
	if( idx != string::npos ){
		tmp = tmp.substr(idx+1);
		idx = tmp.find("http");
		if( idx != string::npos ) return true;
	}

	tmp = link;
	idx = tmp.find("misc");
	if( idx != string::npos ){
		tmp = tmp.substr(idx+1);
		idx = tmp.find("misc");
		if( idx != string::npos ) return true;
	}

	tmp = link;
	idx = tmp.find("ipb");
	if( idx != string::npos ){
		tmp = tmp.substr(idx+1);
		idx = tmp.find("ipb");
		if( idx != string::npos ) return true;
	}

	const char *filter_str[]={
	"cgi-bin",	"htbin",	"linder",	"srs5",		"uin-cgi",  // robots.txt of http://www.expasy.org/
	"uhtbin",	"snapshot",	"=+",		"=-",		"script",
	"gate",		"search",	"clickfile",	"data/scop",	"names",
	"staff/",	"enter",	"user",		"mail",	"pst?",
	"find?",	"ccc?",		"fwd?",		"tcon?",	"&amp",
	"counter?",	"forum",	"cgisirsi",	"{",		"}",
	"proxy",	"login",	"00.pl?",	"sciserv.pl",	"sign.asp",
	"<",		">",		"review.asp?",	"result.asp?",	"keyword",
	"\"",		"'",		"php?s=",	"error",	"showdate",
	"niceprot.pl?",	"volue.asp?id",	".css",		".asp?month",	"prot.pl?",
	"msg.asp",	"register.asp", "database",	"reg.asp",	"qry?u",
	"p?msg",	"tj_all.asp?page", ".plot.",	"comment.php",	"nicezyme.pl?",
	"entr",		"compute-map?", "view-pdb?",	"list.cgi?",	"lists.cgi?",
	"details.pl?",	"aligner?",	"raw.pl?",	"interface.pl?","memcp.php?",
	"member.php?",	"post.php?",	"thread.php",	"bbs/",		"/bbs"
	};
	int filter_str_num = 75;

	
	for(int i=0; i<filter_str_num; i++){
		if( link.find(filter_str[i]) != string::npos)
		return true;
	}	

	return false;
}

/////////////////////////////
// just for ImgSE
// e.g: http://www.people.com.cn/GB/tupian/index.html
// 	http://news.xinhuanet.com/photo/
// 	http://photo.tom.com/
/////////////////////////////
// comment previous one and open this one

/*
bool CPage::IsFilterLink(string plink)
{
	if( plink.empty() ) return true;
	if( plink.size() > URL_LEN ) return true;

	return false;

	string link = plink, tmp;
	string::size_type idx = 0;

	
	CStrFun::Str2Lower( link, link.length() );

	const char *filter_str[]={
		"tupian", "photo", "ttjstk"
		};
	int filter_str_num = 3;

	CStrFun::Str2Lower( link, link.length() );

	for(int i=0; i<filter_str_num; i++){
		if( link.find(filter_str[i]) != string::npos)
		return false;
	}	

	return true;
}
*/


/*****************************************************************
** Function name: ParseHyperLinks
** Input argv:
**      --
** Output argv:
**      --
** Return:
        true: success
        false: fail
** Function Description:  Parse hyperlinks from the web page
** Version: 1.0
** Be careful:
*****************************************************************/
bool CPage::ParseHyperLinks()
{
	if( GetContentLinkInfo() == false ) return false;

	if( m_sContentLinkInfo.empty() ) return false;

	bool bFind4SE = false;
	//bool bFind4History = false;
	if( GetLinkInfo4SE() ){
		if( FindRefLink4SE() ) bFind4SE = true;
	} 

	//if( GetLinkInfo4History() ){
		//if( FindRefLink4History() ) bFind4History = true;
	//}

	//if( !bFind4SE && !bFind4History ){
	if( !bFind4SE ){
		 return false;
	}

	//return   GetHref(m_sContentLinkInfo.c_str(), "href", m_listLink4SE);

	return true;
}


/*****************************************************************
** Function name: GetContentLinkInfo
** Input argv:
**      --
** Output argv:
**      --
** Return:
        true: success
        false: fail
** Function Description:  Parse hyperlinks from the web page
** Version: 1.0
** Be careful:
*****************************************************************/
bool CPage::GetContentLinkInfo()
{
	if( m_sContent.empty() ) return false;
	
	m_sContentLinkInfo = m_sContent;

	string& s = m_sContentLinkInfo;

	// transform all separation into one space character
	//CStrFun::ReplaceStr(s, "\t", " ");
	//CStrFun::ReplaceStr(s, "\r", " ");
	//CStrFun::ReplaceStr(s, "\n", " ");
	const string delims(" \t\r\n");
	string::size_type idx=0, pre_idx;

	while( (idx = s.find_first_of(delims, idx)) != string::npos ){
		pre_idx = idx;
		s.replace(idx,1,1,' ');
		idx++;
		while( (idx = s.find_first_of(delims, idx)) != string::npos ){
			if( idx-pre_idx == 1 ){
				s.erase(idx, 1);
			} else {
				break;
			}
		}

		idx--;
	}

	// transform all "<br>" into one space character
	CStrFun::ReplaceStr(s, "<br>", " ");

	if( s.size() < 20 ) return false;

	// Keep only <img ...>, <area ...>,<script ...> and <a href ...> tags.
	string::size_type idxHref=0,idxArea=0,idxImg=0;
	string dest;

	do{
		if( s.empty() ) break;

		idxHref = CStrFun::FindCase(s, "href");
		idxArea = CStrFun::FindCase(s, "<area");
		idxImg = CStrFun::FindCase(s, "<img");

		pre_idx = idxHref > idxArea? idxArea: idxHref;
		pre_idx = idxImg > pre_idx? pre_idx: idxImg;
		if( pre_idx == string::npos) break;

		s = s.substr(pre_idx);
		idx = s.find_first_of('<',1);
		if( idx != string::npos ){
			dest = dest + s.substr(0,idx);
		}else{
			break;
		}

		s = s.substr(idx);
		idxHref=0; idxArea=0; idxImg=0;
	}while(1);

	s = dest;

	
	/* erase all '\' character
	 * too avoid the following situations:
	 *      document.write("<A href=\"/~webg/refpaper/index.html\">t2</A>");
	*/
	CStrFun::EraseStr(s, "\\");

	if( s.size() < 20 ) return false;

	return true;
}

/*****************************************************************
** Function name: GetLinkInfo4SE()
** Input argv:
**      --  
** Output argv:
**      --
** Return:
       true: success
       false: fail
** Function Description:  Get links for SE
** Version: 1.0
** Be careful:
*****************************************************************/
bool CPage::GetLinkInfo4SE()
{