#!/usr/local/bin/perl -w
#This little script searches out bad grammer and email the offenders.

use LWP::UserAgent;
use HTTP::Request;
use HTTP::Request::Common qw(POST);
use HTTP::Response;
use URI::URL;
use URI::Escape;

$continue = 1;

print ("Didja remember to get a fresh timestamp?\n\n");
my $ua = new LWP::UserAgent; # We'll need this to check response headers
print "Grabbing URLs!\n";
& grabAltaVistaOffenders;
& delDupes;
while ($continue == 1)
{
print "Found $#newOffenders on this page!\n";
    for ($j=0;$j <= $#newOffenders;$j++) #As long as there are more frames, let's do this.
    {
	print "Checking $newOffenders[$j]\n";
	$badInstance = "null";
	my $ua = new LWP::UserAgent; # We'll need this to check response headers
	my $request = new HTTP::Request('GET', $newOffenders[$j]);
	my $response = $ua->request($request);
	if ($response->is_success)
	{
		print "Checking for Gets!\n";
	    if ($response->content =~ /(.{1,57}get's.{1,57})/im)
	    {
		$badInstance = $1;	
		$badInstance =~ s/<[^>]*>//g;
		
		print "Checking for mailto!\n";
		if ($response->content =~ /mailto\s*:\s*(\w+\@[^\"\'\s\>\?]*)/i)
		{
		    $theirEmail = $1;
		    if ($badInstance ne "null")
		    {
			$ua = new LWP::UserAgent;
			print ("Mailto: $theirEmail\n");
			print ("Found at $newOffenders[$j]\n");
			print("Is this bad? - $badInstance\n");	
			print chr(7);
			$answer = <STDIN>;
			if ($answer =~ /y/i)
			{
			        open (LOOSES, "gets.txt") or die "Could not open Loosers!";
                                @looses = <LOOSES>;
				close LOOSES;
                        	foreach $guy (@looses)
                        	{
                                if ($guy =~ /:::/)
                                {
                                        $guy =~ s/:::/"$badInstance"/;
                                }
                        	}
				$looses = join ("", @looses);
			    my $req = POST 'http://www.yawmail.com/cgi-bin/yawmail.cgi',
			    [ sessionid => 'grammarbot:0:1:0:6:180720203:16:4:1',
			      to => $theirEmail,
			      subject => 'Typo on '.$newOffenders[$j],
			      cc => '',
			      bcc => 'grammarbot@yawmail.com',
			      files => '',
			      msg => $looses,
			      send => 'Send',
			      ];
			    $req->authorization_basic('grammarbot', 'sowell');
			    $content = $ua->request($req)->as_string;
			    push (@mailed, $newOffenders[$j]);
			}
		    }
		    else
		    {
			print "something went wrong!";
		    }
		}
		else
		{
		    push (@noMailto, $newOffenders[$j]);
		}
	    }
	}
    }
    print "you have scoured 2 pages of Google responses. Would you like to do two more?";
print chr(7);
print chr(7);
    $answer = <STDIN>;
    if ($answer =~ /y/i)
    {
	$continue = 1;
	& grabGoogleOffenders;
	& delDupes;
    }
    else
    {
	$continue = 0;
    }
}
& printReport;

sub printReport
{
    open (REPORT, ">>report.txt");
    open (ALREADYBOTHERED, ">>GetsalreadyBothered.txt");
    foreach $user (@mailed)
    {
	print REPORT "Mailed to $user\n";
	print ALREADYBOTHERED "$user\n";
    }
    foreach $user (@noMailto)
    {
	print REPORT "Couldn\'t find Mailto at: $user\n";
    }
}


sub grabGoogleOffenders
{
    @urlList = ""; # Initialize the list
    @tempOffenders = "";
    @variousOffenders = "";
    @newOffenders = "";
    open (POSITION, "googlePosition.txt");
    while (<POSITION>)
    {
        $position = $_;
        #if ($position > 200)
        #{
        #    die "That's it! You did all twenty!"
        #}
    }
    close POSITION;
    for ($p = $position;$p < $position+20;$p+=10) #Let's go grab the Google pages
    {
        push (@urlList, "http://www.google.com/search?q=get%27s&start=".$p."&sa=N");
    }
    open (POSITION, ">googlePosition.txt");# Let's write position to a file so we don't "loose" our place.
        print POSITION $p;
    close POSITION;
    for ($k=0;$k<=$#urlList;$k++) #Now let's grab the URLs listed on those pages.
    {
        my $request = new HTTP::Request('GET', $urlList[$k]);
        my $response = $ua->request($request);
        if ($response->is_success)
        {
            $html = $response->content;
            @tempOffenders = ($html =~ /\/white\.gif\s*height=\d+\s*width=\d+\s*alt=\"\"><\/A>\s*<a\s*href=\s*"*([^\s\"\=]+(?:\.htm|\.html|\.asp))/gi);
        push (@variousOffenders, @tempOffenders);
    }
}
        @someOffenders = sort @variousOffenders;
        $prev = 'nonesuch';
        @variousOffenders = grep($_ ne $prev && ($prev = $_), @someOffenders); #Take out dupes from perl FAQ #4
}

sub grabLycosOffenders
{
    @urlList = ""; # Initialize the list
    @tempOffenders = "";
    @variousOffenders = "";
    @newOffenders = "";
    open (POSITION, "Lycosposition.txt");
    while (<POSITION>)
    {   
        $position = $_;
    }
    close POSITION;
    for ($p = $position;$p < $position+20;$p+=10) #Let's go grab the AltaVista pages
    {
        push (@urlList, "http://www.lycos.com/cgi-bin/pursuit?query=get%27s&cat=lycos&mtemp=pages&adv=1&first=".$p."&maxhits=10&ps=0&pm=1&u0=N&u1=Y&u2=Y&u3=Y&u4=cats&u5=1");
    }
    open (POSITION, ">Lycosposition.txt");# Let's write position to a file so we don't "loose" our place.
        print POSITION $p;
    close POSITION; 
    for ($k=0;$k<=$#urlList;$k++) #Now let's grab the URLs listed on those pages.
    {   
        my $request = new HTTP::Request('GET', $urlList[$k]);
        my $response = $ua->request($request);
        if ($response->is_success)
        {
            $html = $response->content;
            @tempOffenders = ($html =~ /target=(http%3A[^\s\"\=]+(?:\.htm\w*|\.asp))/gi);
        push (@variousOffenders, @tempOffenders);
    }   
}       
        @someOffenders = sort @variousOffenders;
        $prev = 'nonesuch';
        @variousOffenders = grep($_ ne $prev && ($prev = $_), @someOffenders); #Take out dupes from perl FAQ #4
}
sub grabAltaVistaOffenders
{
    @urlList = ""; # Initialize the list
    @tempOffenders = "";
    @variousOffenders = "";
    @newOffenders = "";
    open (POSITION, "position.txt");
    while (<POSITION>)
    {
	$position = $_;
	#$position += 10;
	if ($position > 200)
	{
	    die "That's it! You did all twenty!"
	}
    }
    close POSITION;
    for ($p = $position;$p < $position+20;$p+=10) #Let's go grab the AltaVista pages
    {
	push (@urlList, "http://www.altavista.com/cgi-bin/query?pg=q&kl=XX&q=get%27s+you&stq=".$p."&c9k");
    }
    open (POSITION, ">position.txt");# Let's write position to a file so we don't "loose" our place.
	print POSITION $p;
    close POSITION;
    for ($k=0;$k<=$#urlList;$k++) #Now let's grab the URLs listed on those pages.
    {
	my $request = new HTTP::Request('GET', $urlList[$k]);
	my $response = $ua->request($request);
	if ($response->is_success)
	{
	    $html = $response->content;
	    @tempOffenders = ($html =~ /\d\.\s*<\/b><a\s*href=\s*"*([^\s\"\=]+(?:\.htm\w*|\.asp))/gi);
        push (@variousOffenders, @tempOffenders);
    }
}
        @someOffenders = sort @variousOffenders;
        $prev = 'nonesuch';
        @variousOffenders = grep($_ ne $prev && ($prev = $_), @someOffenders); #Take out dupes from perl FAQ #4
}

sub delDupes
{
    open (ALREADYBOTHERED, "GetsalreadyBothered.txt");
    while(<ALREADYBOTHERED>)
    {
	chomp $_;
	push (@alreadyBothered, $_);
    }
#From Perl Cookbook:
    my %seen;
    my @aonly;
    @seen{@alreadyBothered} = ();
    
    foreach $item(@variousOffenders){
        push (@newOffenders, $item) unless exists $seen{$item};
    }
}