.gitignore

pages/*
episode_pages/*
videos/*
cookies.txt

railscasts_download.rb

#!/usr/bin/env ruby

require "nokogiri"
require "open-uri"
require "thread/pool"

$pool = Thread.pool(24)

def download(url, output_file)
  unless File.exists?(output_file)
    $pool.process {  system("wget -c #{url} --load-cookies=cookies.txt -O #{output_file}") }
  end
end

def download_pages(max = 48)
  (1..max).each do |i|
    download "http://railscasts.com/?page=#{i}", "pages/page#{i}.html"
  end
end

def extract_episode_links
  links = []
  Dir["pages/*.html"].each do |page_html|
    doc = Nokogiri::HTML(open(page_html))
    doc.css("div.episode h2 a").each do |link|
      links << "http://railscasts.com#{link['href']}".tap { |l| puts l }
    end
  end
  links
end

def extract_episode_name(episode_link)
  episode_link.match(/.*\/(.+)/)[1]
end

def download_episode_pages(episode_link)
  download episode_link, "episode_pages/#{extract_episode_name(episode_link)}.html"
end

def extract_mp4_links
  links = []
  Dir["episode_pages/*.html"].each do |page_html|
    # puts page_html
    doc = Nokogiri::HTML(open(page_html))
    doc.css("a").each do |link|
      links << link['href'].tap { |link| puts link } if link['href'] =~ /mp4$/
    end
  end
  links
end

trap(:INT) {
  puts "INT signal caught, now exiting"
  exit
}

# step 1: mkdir pages then ->
#download_pages(48)

# step 2: mkdir episode_pages then ->
#extract_episode_links.each { |link| download_episode_pages(link) }

# step 3:
extract_mp4_links.each { |link| download link, "videos/#{extract_episode_name(link)}" }

# puts extract_episode_name("http://railscasts.com/episodes/90-fragment-caching-revised")
$pool.shutdown