Hello guys! I'm trying to learn the Boost regex library for a more-convenient means of error-checking and the asio library to allow a fairly portable-means of manipulating data over TCP/IP.

My goal is to retrieve the file information from a website (basically http layer).

My first problem is determining the Regex for an IP. I've tried the following Regex's--

"^w{3}\\.{1}[.]+\\.{1}[.]+"
"^w{3}" // got desperate here, and it didn't work!

-- the first one I tried to match against 3 w's at the start of the line, a period, and then other characters, then another period and finally whatever other characters to follow. I'm aware that some websites don't require the www prefix, I just wanted to do something interesting for checking to experiment around. Apparently, I misunderstood the Regex syntax I've looked up on Wikipedia and Boost/regex. Any help with this would be much appreciated.

My second problem is the understanding of how to make my client-program retrieve a server-file from the internet much like a web-browser does. I can't stress enough that I'm mainly experimenting with this to see if it can be done, so hopefully someone that is network savvy and/or someone with experience with Boost asio can help me with the next part. Here is my code (and please bare with me I know it's ugly right now, but I have a format for debugging each function that is really helpful to me)--

#ifndef INTERNET_DATA_PARSER_H
#define INTERNET_DATA_PARSER_H

#include <string>
#include <map>
#include <ostream>
#include <iostream>

using std::string;
using std::map;
using std::cout;
using std::ostream;

/**
 * The purpose of this class is to do the following--
 *
 *	- Open a link to a specified URL and apply a pattern-match to retrieve all valid matches
 *	- of the pattern, then send the data to some output (cout, a stringstream, a file, etc).
 */
namespace Parser{

	typedef class InternetDataParser{

		public:

			InternetDataParser(const string& = "style=([^>.]+)>", const string& = "www.google.com", ostream& = std::cout);
			string						setInfo(const string&, const string&, ostream&);
			void						setPattern(const string&);
			string						setURL(const string&);
			void						setOutputStream(ostream&);
			string						write();
			string						getPattern();
			string						getURL();
			ostream&					getOstream();
			static string				getReason(int);

		private:

			string						pattern;
			string						url;
			ostream*					out;
			static map<int, string>		reasons;
			static bool					initNotCalled;
			static void					init();
			
	} IDP;

};

#endif
#include "InternetDataParser.h"
#include <string>
#include <map>
#include <ostream>
#include <iostream>
#include <boost/regex.hpp>
#include <boost/array.hpp>
#include <boost/asio.hpp>

using std::string;
using std::map;
using std::pair;
using std::copy;
using std::stringstream;
using Parser::IDP;
using boost::asio::ip::tcp;

bool IDP::initNotCalled = true;
map<int, string> IDP::reasons = map<int, string>();

/*
 * Attempts to assign all necessary information at once upon construction.
 *
 * This function also calls IDP::init() and initializes the static members
 */
IDP::InternetDataParser(const string& p, const string& u, ostream& o) : pattern(p), url(u), out(&o) {

	IDP::init();
}

/*
 * Attempts to set the Pattern, URL and output stream. Only fails if an invalid URL is passed.
 *
 *	-Input: The Pattern, URL and output stream
 *	-Returns a reason (reason for failure, or "Good" if the URL is a good URL)
 */
string IDP::setInfo(const string& p, const string& u, ostream& o){
	
		if(strcmp(setURL(u).c_str(), reasons[6].c_str()) == 0)
			return reasons[6];
		else{
			setPattern(p);
			setOutputStream(o);
		}

	return reasons[0];
}

/*
 * Sets the member variable pattern with the specified pattern
 */
void IDP::setPattern(const string& p){

	pattern = p;
}

/*
 * Attempts to set the URL to the value specified.
 *
 *	-Returns the reason for failure, or "Good" if it did not fail
 */
string IDP::setURL(const string& u){

	string lower_u;

	for(size_t i = 0; i < u.length(); i++)
		lower_u += tolower(u[i]);
	
	// Problem here also
	static const boost::regex url_format ("w{3}\\.{1}[.]+\\.{1}[.]+");

	(*out) << lower_u << std::endl;

	if( !boost::regex_match( lower_u, url_format ) ){
		(*out) << reasons[6] << std::endl;
		return reasons[6];
	}else url = lower_u;

	return reasons[0];
}

/*
 * Sets the ostream& out member with the ostream provided
 */
void IDP::setOutputStream(ostream& o){

	out = &o;
}

/*
 * Retrieves data from the URL and writes it to the output stream
 *
 *	-Returns the reason for failure, or "Good" if nothing failed.
 */
string IDP::write(){

	// MOST OF THIS LOGIC WAS BORROWED DIRECTLY FROM Boost asio!!!!
	// Please help me understand the network/application layer of TCP to http
	// for retrieving file data on the internet!
	try{

		boost::asio::io_service io_service;
		tcp::resolver resolver(io_service);
		tcp::resolver::query query( url.c_str() , "http");
		tcp::resolver::iterator endpoint_iterator = resolver.resolve(query);
		tcp::resolver::iterator end;
		tcp::socket socket(io_service);
		boost::system::error_code error = boost::asio::error::host_not_found;

		while (error && endpoint_iterator != end){

			socket.close();
			socket.connect(*endpoint_iterator++, error);
		}
		if (error){
			return reasons[7];
		}

		for (;;){
			boost::array<char, 128> buf;
			boost::system::error_code error;

			(*out) << socket.available();
			size_t len = socket.receive(boost::asio::buffer(buf));

			(*out).write(buf.data(), len);

			// When the server closes the connection, the ip::tcp::socket::read_some() function will exit with the 
			// boost::asio::error::eof error, which is how we know to exit the loop.

			if (error == boost::asio::error::eof){
				(*out) << reasons[8];
				break; // Connection closed cleanly by peer.
			}else if (error)
				throw boost::system::system_error(error); // Some other error.
			
		}

	}catch (std::exception& e){
		std::cerr << e.what() << std::endl;
	}

	
	return reasons[0];
}

/*
 * -Returns the regex pattern assigned to 'pattern' member variable
 */
string	IDP::getPattern(){

	return pattern;
}

/*
 * Simply returns the URL as a string
 *
 *	-Returns the URL as a string
 */
string	IDP::getURL(){

	return url;
}

/*
 * Simply returns the output stream currently used by this IDP
 *
 *	-Returns the ostream& used by this object.
 */
ostream& IDP::getOstream(){

	return *out;
}

/*
 * A reason is nothing more than a stored string in the reasons map.
 * The reason represents what happened after a function call.
 *
 *	-Input: The reason code.
 *	-Returns the string representation of the reason code.
 */
string IDP::getReason(int r){

	IDP::init();
	
	return ( (reasons.find(r) != reasons.end() ) ? reasons[r] : "Reason not found." );
}

/*
 * Initializes the static map 'reasons'
 */
void IDP::init(){

	if(initNotCalled){

		initNotCalled = false;

		reasons.insert( pair<int, string>(0, "Good") );
		reasons.insert( pair<int, string>(1, "No matches found") );
		reasons.insert( pair<int, string>(2, "Unable to open URL") );
		reasons.insert( pair<int, string>(3, "Unable to open File") );
		reasons.insert( pair<int, string>(4, "Unable to write to File") );
		reasons.insert( pair<int, string>(6, "Invalid URL") );
		reasons.insert( pair<int, string>(7, "No Host Found") );
		reasons.insert( pair<int, string>(8, "Connection closed.") );
	}
}
#include <iostream>
#include "InternetDataParser.h"

using namespace std;
using namespace Parser;

int main(){

	IDP idp;
	idp.setURL("www.yahoo.com");// fails even though it should match regex?
	idp.write();

	cin.get();
	return 0;
}

-- After running the application, I see one number (the number of bytes that can be read before a block) which is 0, and then "End of File". I don't see any repeats of numbers or and characters being printed to the console screen which means after the function call of receive, it immediately caught an error and the IDP::write function returns. What I'd like to know is how do I not use TCP to wait for a specific server response other than the data of the file itself? How do I stream the bytes of the file into the buffer?

Any help fixing my code would be much appreciated. Thanks!

EDIT:

I found this link from boost from the advanced examples section--

http://www.boost.org/doc/libs/1_44_0/doc/html/boost_asio/example/http/client/async_client.cpp

-- now I just need to figure out how to apply regex and I'm good to go!

Too much code for me to look at right now... Going to bed soon... So I will answer only your first question.

Your regular expression is incorrect. the {#} is not supposed to be used the way you do. If you want to match 3 of w's, just spell it out...

i.e. /^www\.[^\.]*\.[^\.]*\.\w+{2,4}/i

That's just an example which is not a perfect solution. It just simply matches a string starting with 'www' followed by anything but a '.' followed by anything but another '.' and end with any character (not number or special character... please look at what \w is for short-handed regex).

i.e. www.abd345_ok.doreme_22_.blah would pass
i.e. www.abc123.not321_.blahh will not pass
i.e. www.abc123.not321_.b will not pass

PS: It is not easy to match all possible URL strings. It is the same with e-mail address. There are so many possible formats nowadays... I would just check for /^www\.\w+/i and that's all. It is very general (a string starts with www. and has at least one character follow it). My assumption is that if a string start with www., then it should be a valid hyper link.

commented: Thank you very much. +0

Okay, I have a question.

What exactly does the i at the end represent for each of your patterns? Just the literal i, or does it have some other meaning?

I don't see it mentioned on Boost regex, nor have I encountered it in Lua.

I don't expect you to answer anytime soon... you're asleep ;p

The 'i' indicates the match to 'ignore case'. You may need to read about regexp somewhere else because it is standard.

Sorry, still don't have much time to answer your 2nd question. Hope someone else will handle this. Need to get going and won't be back until late here.

commented: You've helped more than enough. Thank you! +2
Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.