download_railscasts.rb

#!/usr/local/bin/ruby

# A script to download the latest episodes from railscasts.com
#
# author: modsaid <mahmoud@modsaid.com>
# mechanize support: Hossam Hammady <github@hammady.net>
#   gem install mechanize

require 'rubygems'
require 'mechanize'

RAILS_CASTS_FEED='http://feeds.feedburner.com/railscasts'
DOWNLOAD_DIRECTORY='.'

downloaded_episodes = Dir.glob("#{DOWNLOAD_DIRECTORY}/*.mp4").map {|episode| File.basename(episode, ".mp4")}
puts "Found #{downloaded_episodes.length} episode(s) cached"

puts "Checking railscasts feeds for new episodes..."

agent = Mechanize.new
agent.user_agent_alias = 'Linux Mozilla'
agent.history.max_size = 0 # so that memory usage doesn't get bloated by unnecessary pages

page = agent.get(RAILS_CASTS_FEED)
rss = Nokogiri::XML.parse(page.body, Mechanize::Util::uri_escape(RAILS_CASTS_FEED))
count=0

agent.pluggable_parser.default = Mechanize::Download

(rss/'//item').each do |rss_item|
  episode = rss_item.at('./link').text.split('/').last
  print "Found episode: #{episode}: "
  unless downloaded_episodes.include?(episode)
    title = rss_item.at('./title').text
    size = rss_item.at('./enclosure/@length').text.to_i/1024/1024
    puts "downloading #{size}MB..."
    link = rss_item.at('./enclosure/@url')
    agent.download(link, "#{DOWNLOAD_DIRECTORY}/#{episode}.mp4.tmp")
    File.rename("#{DOWNLOAD_DIRECTORY}/#{episode}.mp4.tmp", "#{DOWNLOAD_DIRECTORY}/#{episode}.mp4") # protect against broken downloads
    count += 1
  else
    puts "cached"
  end
end

puts "Done. #{count} new episodes were downloaded"