htmlparse.cpp

html文件编码转换与解析

#include "filetobuf.h"
#include "htmlparse.h"

//Get the Url from <a href = "...">
bool GetAnchorURL(string &tag){
	int i = 0;
	while(!((tag[i] == 'h') && (tag[i+1] =='r') && (tag[i+2] == 'e') && (tag[i+3] == 'f'))){
		i ++;
	}
	i  = i + 4;
	while(tag[i++] != '='){
	}
	while(tag[i] == ' '){
		i++;
	}
	string url;
	while(tag[i] != '\0')
	{
		url += tag[i++];
	}
	string realurl;
	string tempurl = RemoveTarget(url);
	i = 1;
	if(tempurl[0] == 34)
	{
		while(tempurl[i] != 34){
			realurl += tempurl[i];
			i++;
		}
	}
	else if(tempurl[0] == 39)
	{
		while(tempurl[i] != 39){
			realurl += tempurl[i];
			i++;
		}
	}
//	int j = 0;
	else{
		realurl = tempurl;
	}
//	cout<<"url:"<<realurl<<endl;
	fanchor<<realurl<<endl;
	return true;
}

//initialization without argument
HtmlParser::HtmlParser() {
}

//initialization with argument
HtmlParser::HtmlParser(string input) {
	URL = input;
	SetBaseURL(URL);
	HtmlParser();
}

//finalization
HtmlParser::~HtmlParser() {
}

//Turning the revoluteURL to the AbusoluteURL
string HtmlParser::GetAbusoluteURL(string input) {
	string AbusoluteURL;
	AbusoluteURL = baseURL + input;
	return AbusoluteURL;
}

//set This parsing file's Base URL
bool HtmlParser::SetBaseURL(string URL) {
	baseURL = URL;
	return true;
}

//just get the URL for testing
string HtmlParser::GetURL() {
	return URL;
}

//just get the Base URL for testing
string HtmlParser::GetBaseURL() {
	return baseURL;
}

//Parse the html file
bool HtmlParser::Parse() {
	//	cout << "HtmlParser Parse: ParsePosition  " << parseposition << endl;
	tagtype type;
	while (parseposition != 1024) {
		if (buf[bufparseflag][parseposition] == '<') {
			type = ParseTag();
			switch (type) {
			case title:
				if(!ParseTitle()){
					return false;
				}
				break;
			case anchor:
				GetAnchorURL(tag);
				if(!ParseA()){
					return false;
				}
				break;
			case script:
				if(!ParseScript()){
					return false;
				}
				break;
			case style:
				if(!ParseStyle()){
					return false;
				}
				break;
			case fileend:
				return false;
			default:
				break;
			}
		} 
		else{
			ParseText();
//				cout<<buf[bufparseflag][parseposition]<<endl;
//				cout<< parseposition<<endl;
			parseposition ++;
		}
	}
	return true;
}

//Parse the Tag to clearfy the tagtype
tagtype HtmlParser::ParseTag() {
	//	cout << "HtmlParser ParseTag:" << endl;
	tag.clear();
	int i = parseposition;
	if(!IsBufEnd(i)){
		return fileend;
	}
//	tagstart = i;
	//check the tag end
	while (buf[bufparseflag][i] != '>') {
		tag += buf[bufparseflag][i];
		if(!IsBufEnd(i)){
			return fileend;
		}
	}
	IsBufEnd(i);
	parseposition = i;
	//case tag title
	MakeLower(tag);
	if ((tag[0] == 't') && (tag[1] == 'i') && (tag[2] == 't') && (tag[3] == 'l') && (tag[4] == 'e')) {
		return title;
	}
	//case tag anchor
	else if ((tag[0] == 'a') && (tag[1] == ' ')) {
		return anchor;
	}
	else if ((tag[0] == '/') && (tag[1] == 't') && (tag[2] == 'i') && (tag[3] == 't') && (tag[4] == 'l') && (tag[5] == 'e')) {
		return titleend;
	}
	else if ((tag[0] == '/') && (tag[1] == 'a')) {
		return anchorend;
	} 
	else if ((tag[0] == 's') && (tag[1] == 'c') && (tag[2] == 'r') && (tag[3] == 'i') && (tag[4] == 'p') && (tag[5] == 't')) {
		return script;
	}
	else if ((tag[0] == '!') && (tag[1] == '-') && tag[2] == '-') {
		int j = tag.size();
//		cout<<j<<endl;
//		cout<<tag[j-1]<<endl;
		if((tag[j-1] != '-') && (tag[j-2] != '-')){
			i = parseposition;
			if(!IsBufEnd(i)){
				return fileend;
			}
			wstring com;
			j = 0;
			while (!((buf[bufparseflag][i] == '>') && (com[j-1] == '-') && (com[j-2] == '-'))) {
				com[j++] = buf[bufparseflag][i];
				if(!IsBufEnd(i)){
					return fileend;
				}
			}
			parseposition = i + 1;
		}
		return comment;
	}
	else if ((tag[0] == 's') && (tag[1] == 't') && (tag[2] == 'y') && (tag[3] == 'l') && (tag[4] == 'e') ) {
		return style;
	}
	else if ((tag[0] == '/') && (tag[1] == 's') && (tag[2] == 't') && (tag[3] == 'y') && (tag[4] == 'l') && (tag[5] == 'e')) {
		return styleend;
	}
	else return other;
}

//Parse the <title> tag
bool HtmlParser::ParseTitle() {
	//	cout << "HtmlParser::ParseTitle()" << endl;
	//	int i = parseposition;
	int i;
	//IsBufEnd(i);
	string title;
	int end = 1;
	while (end == 1) {
		i = parseposition;
		while (buf[bufparseflag][i] != '<') {
			title += buf[bufparseflag][i];
			if(!IsBufEnd(i)){
				return false;
			};
		}
		parseposition = i;
		if (ParseTag() == titleend) {
			end = 0;
		}
	}
//	cout <<"title:"<< title << endl;
	ftext << title <<endl;
	return true;
}

//Parse the <a> tag
bool HtmlParser::ParseA() {
	//	cout << "HtmlParser::ParseA" << endl;
	//	int i = parseposition;
	int i;
	string anchor;
	int end = 1;
	while (end == 1) {
		i = parseposition;
		while (buf[bufparseflag][i] != '<') {
			anchor += buf[bufparseflag][i];
			if(!IsBufEnd(i)){
				return false;
			}
		}
		parseposition = i;
		tagtype type = ParseTag();
		if (type == anchorend) {
			end = 0;
		}
	}
//	cout<<"anchor:"<< anchor <<endl;
	fanchor << anchor << endl;
	return true;
}

//Parse the Script
bool HtmlParser::ParseScript(){
//	cout<< "ParseScript!"<<endl;
	int end = 1;
	char s[9];
	int sflag = 0;
	int j;
	int i = parseposition;
	while(end == 1)
	{
		s[sflag] = buf[bufparseflag][i];
		if((s[sflag] >= 65) && (s[sflag] <= 90)){
					s[sflag] += 32;
		}
		sflag = (sflag + 1) % 9;
		j = sflag;
		if((s[j] == '<') && (s[(j+1) %9] == '/') && (s[(j+2) %9] == 's') && (s[(j+3) %9] == 'c')
				&& (s[(j+4) %9] == 'r') && (s[(j+5) %9] == 'i') && (s[(j+6) %9] == 'p') 
				&& (s[(j+7) %9] == 't') && (s[(j+8) %9] == '>')){
			end = 0;
		}
		if(!IsBufEnd(i)){
			return false;
		}
	}
	parseposition = i;
//	cout<< anchor <<endl;
	return true;
}

//Parse the CSS style
bool HtmlParser::ParseStyle(){
//	cout<< "ParseScript!"<<endl;
	int i;
//	string anchor;
	int end = 1;
	while (end == 1) {
		i = parseposition;
		while (buf[bufparseflag][i] != '<') {
//			anchor += buf[bufparseflag][i];
			if(!IsBufEnd(i)){
				return false;
			}
		}
		parseposition = i;
		tagtype type = ParseTag();
		if (type == styleend) {
			end = 0;
		}
	}
//	cout<< anchor <<endl;
	return true;
}

//Parse the Text
bool HtmlParser::ParseText() {
	if ((buf[bufparseflag][parseposition] != '\n')
			&& (buf[bufparseflag][parseposition] != '\t') 
			&&	(buf[bufparseflag][parseposition] != '\v')
			&&	(buf[bufparseflag][parseposition] != '\r')) {
		text += buf[bufparseflag][parseposition];
	}
	return true;
}

int main() {
	char filename[100];
	cout<< "Please enter the file to be parsed:"<<endl;
	cin >> filename;
	initread(filename);
	cout<< "Please enter the file of saving the anchor" <<endl;
	char anchorfilename[100];
	cin >> anchorfilename;
	fanchor.open(anchorfilename, ios::out|ios::trunc);
	char textfilename[100];
	cout<< "Please enter the file of saving the text" <<endl;
	cin >> textfilename;
	ftext.open(textfilename, ios::out|ios::trunc);
	string URL;
	HtmlParser* hp = new HtmlParser(URL);
	do {
		endflag = readbuf();
		parseposition = 0;
		bufparseflag = (bufparseflag + 1) % 5;
		hp->Parse();
	} while (!endflag);
//	cout<<"text:"<< text << endl;
	ftext << text;
	/*	string URL;
	 cout<< hp->GetURL() << endl;
	 cout<< hp->GetBaseURL() << endl;
	 hp->ParseA(); */
	ReadEnd();
	return 0;
}