Snippets.PRO - AncientDragon

C++: StopForumSpam

17.05.2013 07:33 AncientDragon C++ 2 Comments

This is a complete console program compiled with Microsoft Visual Studio 2012 that demonstrates how to access the api from (SFS). The program sends a list of ip address to SFS and gets a reply that indicates which ones have been reported as spammers. The list of IP addresses is obtained from a tab deliminated file which was generated from a vBulletin forum web site. You can easily modify the program to get the list of ip address from some other stream.\r\n ... Reveal Code

#include "Socket.h"
#include 
#include  // required for SHGetSpecialFolderPath()
#include 
#include 
#include 
#include 
#include 
#pragma comment(lib, "Shell32.lib")

using namespace std;

// used by  to do case-insensitive comparison of keys
struct classcomp {
  bool operator() (const string& lhs, const string& rhs) const
  {
          return stricmp(lhs.c_str(),rhs.c_str()) < 0;
  }
};

//  required for SHGetSpecialFolderPath()
char buf[MAX_PATH] = {0};

// maximum number of ip addresses that can be sent 
// to www.stopforumspam.com at one time.
const int MaxCount = 15;


// extract ip address from a string in the form "1.1.1.1"
string finditem(string line, string s, int pos1)
{
        int pos2;
        // find end of text " by location the '>' character
        for(; line[pos1] != '>'; pos1++)
                ;
        // now find ""
        pos2= line.find(s);
        // cut out what's inbetween  and 
        line = line.substr(pos1+1,pos2-pos1-1);
        return line;
}

// write final results to data file.
void save(ofstream& out, map& gooditems)
{
        map::iterator it = gooditems.begin();
        for(; it != gooditems.end(); it++)
        {
                out << it->first << '\t' << it->second << '\n';
        }
}

// optional parameter to main() is the input filename, which may NOT contain
// path.  Files are expected to be located in "Libraries/Documents" folder.
int main(int argc, char* argv[]) {

        char* filename = NULL;
        map items;
        map gooditems;
        string line;
        if( argc == 2)
                filename = argv[1];
        else
                filename = "registrations.txt";
        // get full path to libraries/Documents folder
        BOOL b = SHGetSpecialFolderPath(0,buf,CSIDL_MYDOCUMENTS,0);
        // open both input and output files
        string fname = buf;
        fname += "//";
        fname += filename;
        ifstream in(fname.c_str());
        if( !in.is_open() )
        {
                cout << "Can't open file " << filename << '\n';
                return 1;
        }
        fname = buf;
        fname += "\\spam.txt";
        ofstream out(fname.c_str());
        if( !out.is_open() )
        {
                cout << "Can not open output file\n";
                return 1;
        }
        string uname,email,ip,s1,s2,s3;
        // read all the inputfile into vector.  The input file consists of six tab deliminated columns
        // user name, email address, ip address, and three more columns that we don't use.  The data
        // is derived from vBulletin's admin control panel Search Users Waiting Moderation.  Just highlight
        // the list, copy to clipboard, past into Notepad, then write to a file in Libraries/Documents folder.
        while( getline(in,uname,'\t') ) //in >> uname >> email >> ip >> s1 >> s2 >> s3 )
        {
                getline(in,email,'\t');
                getline(in,ip,'\t');
                getline(in,line);
                items[ip] = uname;
        }
        in.close();
        map::iterator it = items.begin();
        size_t pos;
        // Send the IP addresses to wwsw.stopforumspam.com who will return the number of times
        // the ip has been reported for spamming.  Ips can be batch processed, a maximum of 15
        // IPs batch processed at a time.  The socket is closed after each call, so we have to
        // reopen the socket every time we make a request.
        //
        // Requests are in HTM format, such as
        // GET /api?ip[] = 111.111&ip[]=222.222.222.222&ip{}=333.333.333.333, etc for up to 15 addresses
        //
        while( it != items.end() )
        {
                string cmd = "GET /api?";
                int counter = 0;
                do
                {
                        cmd += "ip[]=" + it->first + "&";
                        it++;7
                        counter++;
                } while( it != items.end() && counter < MaxCount);
                if(cmd[cmd.length()-1] == '&')
                        cmd = cmd.substr(0,cmd.length()-1);

                SocketClient s("www.stopforumspam.com", 80);
                s.SendLine(cmd.c_str());
                // stopforumspam.com is going to send back a whole bunch of worthless stuff
                // all we want is ipaddress and number times
                // If the  value is 0 that means the ip has NOT been reported for
                // spamming, and it will be saved in the output file that this program creates.
                while (1) {
                        string l = s.ReceiveLine();
                        if (l.empty()) break;
                        //out << l << '\n';
                        if( (pos = l.find("")) != string::npos)
                        {
                                ip = finditem(l,"",pos);
                                uname = items[ip];
                        }
                        if( l.find("0") != std::string::npos)
                        {
                                cout << uname << '\t' << ip << '\n';
                                // we want the gooditems list to be sorted by username instead
                                // of ip, so uname is used as the key.  We already know the ip 
                                // is unuque in the file, so by definition the uname will also be
                                // unique.  
                                gooditems[uname] = ip;
                        }
                }
    }
        save(out, gooditems);

}