http://paperlined.org/rss/feeds/kottke.gen.pl

#! /usr/bin/env perl
BEGIN{$^W=1}  use strict;

use Data::Dumper;
use File::Basename;
use LWP::Simple;
use HTML::Entities;


#############################################
# Load the main page, parse posts
#############################################
	# Idiots don't implement etag or last-modified, so we have to eat up 70k every time
my $main_page = get('http://www.kottke.org/index.html')
	or die "Unable to fetch kottke main page.\n\t";

print "Fetched bytes ", length($main_page), " from kottke\n";

my @stories;
while ($main_page =~ m#<div class="post">\s*<h2><a href="(.*?)">(.*?)</a></h2>\s*(.*?)<div class="meta">#gis) {
	my $story = {};

	my ($link, $title, $body) = ($1, $2, $3);

	$body =~ s#src="/#src="http://www.kottke.org/#gi;

	$story->{link} = $link;
	$story->{title} = HTML::Entities::encode_entities($title);
	$story->{description} = HTML::Entities::encode_entities($body);

	push(@stories, $story);
}


#############################################
# Output RSS file
#############################################
my $zero_dir = dirname $0;
open RSS, ">$zero_dir/kottke.rss"	or die "Unable to write to kottke.rss: $!";
print RSS <<"EOF";
<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF
  xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
  xmlns="http://purl.org/rss/1.0/"
  xmlns:dc="http://purl.org/dc/elements/1.1/"
>
        <channel rdf:about="http://paperlined.org/rss/feeds/kottke.rss">
                <title>Newcum's Kottke Feed</title>
                <link>http://www.kottke.org/index.html</link>
                <description>David Newcum's RSS Feed of kottke.org.  Contact rss_feeds\@paperlined.org for change requests.</description>
                <language>en-us</language>
        </channel>
EOF

foreach my $story (@stories)
{
	print RSS <<"EOF";
        <item rdf:about="$story->{link}">
                <title>$story->{title}</title>
                <link>$story->{link}</link>
                <description>$story->{description}</description>
        </item>
EOF
}
print RSS "</rdf:RDF>\n";