#!/usr/bin/perl
# Binbot.pl: Reads a list of rss feeds from a file and searches for ke
+ywords, writing matching <title> and
# <link> elements to a new rss file.
# I use it for tracking USENET binaries, but it could be used for any 
+pupose.
#Files: 
#    urls.txt - A text file consisting of URLs to rss feeds, one per l
+ine. 
#        e.g. http://www.whatever.com/some-rss.rss
#    keywords.txt - A text file consisting of keywords to search for, 
+one per line.
# 9/24/06 - switched to URI::Fetch (from LWP::Simple) to allow reading
+ gzip compressed feeds.

use strict;
use warnings;

use XML::RSS::Parser::Lite;
use XML::RSS::SimpleGen;
use URI::Fetch;


#variables
my $url_file = "urls.txt";
my $rss_file = "binbot.rss";
my $keyword_file = "keywords.txt";
my @keywordlist;
my @rsslist;

#load keywords
open (KEYWORDFILE, $keyword_file) || die "No Keyword file ($keyword_fi
+le) found, aborting.\n";
while (<KEYWORDFILE>){
    chomp;
    push (@keywordlist, $_);
}
close KEYWORDFILE;

#load list of rss files
open (URLFILE,"$url_file") || die "No URL file ($url_file) found, abor
+ting.\n";
while (<URLFILE>){
    chomp;
    push (@rsslist, $_);
}
close URLFILE;

#create the output rss image
rss_new('htp://www.whatever_you_want.com',"BinBot Results");

#loop through the rss files searching for keywords
foreach my $rss (@rsslist) {

    print "Reading $rss\n";

    #Fetch the file. If successful, search for keywords
    my $res = URI::Fetch->fetch($rss) or die URI::Fetch->errstr;
    if ($res->is_success) {
              my $xml = $res->content;
          my $rp = new XML::RSS::Parser::Lite;
        $rp->parse($xml);
    
        # Search for keywords in rss file
        for (my $i = 0; $i < $rp->count(); $i++) {
            my $item = $rp->get($i);
            
            foreach my $kw (@keywordlist) {
                my $title = $item->get('title');
                my $url = $item->get('url');
                my $desc = $item->get('description');
                    
                # If the keyword exists in the title, add the entry to
+ the rss image.
                if ( $title =~ /$kw/i ) { 
                    print "Matched keyword $kw\n";
                    rss_item($url,$title,$desc) 
                }
            }
        }
    
    } else {
              print $res->status_line, "\n";
      }

}

# Save the rss image to a file.
rss_save("$rss_file");