#!/usr/bin/perl
##############################################################################
# By BumbleBeeWare.com 2006
# Simple RRS XML grabber and parser
# rss.cgi
# simple program to display rss from text based xml files on your web pages
##############################################################################

# the url of the xml file you want to access include the http://
$feedurl = "http://bumblebeeware.com/rssxml/test.xml";

# set max number of seconds to access feed
$timeout = 30;

###################################################
# MAIN PROGRAM
###################################################

use LWP::UserAgent;
	
$ua = new LWP::UserAgent;
 
# identify this user agent to the website being crawled
$ua->agent("http://bumblebeeware.com RSS for PERL");

# set time out for accessing page
$ua->timeout($timeout);

# get page from actual website
$req = HTTP::Request->new(GET => "$feedurl");
$res = $ua->request($req);

	if ($res->is_success) {
	 	
	 	$tempdata = $res->content;
	 	
	 	# strip the file and make a datafile
	    	&print_page;

	}
	

###################################################
# SUBROUTINES
###################################################

sub print_page {

print "Content-Type: text/html\n\n";
&parse_data;
exit;	
	
}


sub parse_data {

@lines = $tempdata;
@lines = split(/\n/,$tempdata);

# count pages accessed
foreach $line (@lines)  {
chomp $line;

# break down the file into seperate records using the <item> tag
if ($line =~ /\<item\>/i){
	$linecount++;
	$newrecord = 1;}
if ($line =~ /\<\/item\>/i){$newrecord = 0;}

if ($newrecord == 1){$LINE{$linecount} = "$LINE{$linecount} $line";}


}

foreach $linecount (sort keys %LINE) {

#####################################################################################
# you may need to modify these fields to match the xml feed you are accessing
# these seem to be the most common fields used although anything is possible with xml
# title link url description image
#####################################################################################
$title = "";
$link = "";
$description = "";
$url = "";
$image = "";

# get each field
if ($LINE{$linecount} =~ /<title>(.*)<\/title>/i){$title = $1;}
if ($LINE{$linecount} =~ /<link>(.*)<\/link>/i){$link = $1;}
if ($LINE{$linecount} =~ /<description>(.*)<\/description>/i){$description = $1;}
if ($LINE{$linecount} =~ /<url>(.*)<\/url>/i){$url = $1;}
if ($LINE{$linecount} =~ /<image>(.*)<\/image>/i){$image = $1;}

# remove any CDATA junk used to hold the xml together
if ($title =~ /\<\!\[CDATA\[(.*)\]\]/){$title = $1;}
if ($link =~ /\<\!\[CDATA\[(.*)\]\]/){$link = $1;}
if ($description =~ /\<\!\[CDATA\[(.*)\]\]/){$description = $1;}
if ($url =~ /\<\!\[CDATA\[(.*)\]\]/){$url = $1;}
if ($image =~ /\<\!\[CDATA\[(.*)\]\]/){$image = $1;}
	
# change this line to customize your actual html output
print "<p align=left><a href=\"$link\"><b><font face=\"arial\">$title</font></b></a><font face=\"arial\"> - $description</font></p>\n";

}


}