#!/usr/bin/perl
# 
# plosbiology2rss by Pieter Edelman
# Generated by script4rss (http://script4rss.sf.net)
# 
# This script converts http://www.plosbiology.org/plosonline/?request=get-issue to an RSS feed
# 
# ==============================================================================
# PLoS Biology is an open-access scientific biology magazine
# This script attempts to convert the page linking to the current issue to an RSS feed
# ==============================================================================
# 
# Copyright(c) 2004 Pieter Edelman
# Released under the terms of the GNU General Public License (GPL) Version 2.
# See http://www.gnu.org/ for details.


sub fixHTML {
  # Close open tags before new ones are opened
  # Rewrite everything within < and > with a fixed number of dashes
  local($clean_string, $line, $in_tag);
  
  $in_tag = 0;
  
  $clean_string = "";
  split/([<>])/, $_[0];
  while (scalar(@_)) {
    $line = shift(@_);
    if ($line =~ /^<$/) {
      if ($in_tag) {
        $clean_string .= ">";
      }
      $in_tag = 1;
      $clean_string = $clean_string.$line;
      $line = shift(@_);
      $line =~ s/\s*=\s*/=/g;
      $line =~ s/\s+/ /g;
    } elsif ($line =~ /^>$/) {
      $in_tag = 0;
    }
    $clean_string = $clean_string.$line;
  };
  $clean_string;
}

sub cleanup {
  # Convert HTML entries into valid HTML
  #$_[0] =~ s/&(?!\S*;)/&amp;/g;
  $_[0] =~ s/&/&amp;/g;
  $_[0] =~ s/</&lt;/g;
  $_[0] =~ s/>/&gt;/g;
  $_[0];
}

# Read lines from stdin
@lines = <>;

# Print the header
print "<?xml version=\"1.0\" encoding=\"ISO-8859-1\"?>\n";
print "<!DOCTYPE rss PUBLIC \"-//Netscape Communications//DTD RSS 0.91//EN\"\n";
print "\"http://my.netscape.com/publish/formats/rss-0.91.dtd\">\n";
print "<rss version=\"0.91\">\n";
print "  <channel>\n";
print "    <title>PLoS Biology</title>\n";
print "    <link>http://www.plosbiology.org/plosonline/?request=get-issue</link>\n";
print "    <description>Public Library of Sciences: Biology</description>\n";
print "    <ttl>10080</ttl>\n";

# Keep reading lines until they are all used
while (scalar(@lines) > 0) {
  $line = &fixHTML(shift(@lines));

  # Find the catagory name
  if ($line =~ m/<div class=\"sectionTitle.*?<\/div>/) {
    # Find the catagory name
    @match = (split/<div class=\"sectionTitle.*?\">(.*?)<\/div>/, $line);
    $cat_name = '&lt;i&gt;'.&cleanup($match[1]).'&lt;/i&gt;';
  }

  # Find the stories
  if ($line =~ m/<a href=\".*?\" class=\"smallNav\"><img src=\"\/images\/icons\/.*?\" .*?><\/a><b>.*<\/b><br>/i) {
    # Create a line up until /<\/div>/ is found
    while ($line !~ m/<\/div>/) {
      chomp($line);
      $line .= &fixHTML(shift(@lines));
    }
    # Find the different parts
    @match = (split/<a href=\"(.*?)\" class=\"smallNav\"><img src=\"\/images\/icons\/(.*?)" alt=".*?><\/a><b>(.*?)<\/b><br>(.*?)<br>(.*)<br><div class=\"smallNav\">/i, $line);
    # Print out the item
    if (scalar(@match) > 1) {
      print "    <item>\n";
      print '      <title>'.$cat_name.': '.&cleanup($match[3]).'</title>'."\n";
      print '      <link>'.&cleanup($match[1]).'</link>'."\n";
      print '      <description>&lt;img src="http://www.plosbiology.org/images/icons/'.&cleanup($match[2]).'" /&gt;By '.&cleanup($match[4]).'&lt;br /&gt;'.&cleanup($match[5]).'</description>'."\n";
      print "    <\/item>\n";
    }
  }
}

# Print the footer
print "  <\/channel>\n";
print "<\/rss>\n";