Script to download all Destroy All Software screencasts, account needed

das_download.rb

#! /usr/bin/env ruby
# usage: 
# $ das_download.rb email password [download_directory]

require 'mechanize'
# gem 'mechanize-progressbar'

email    = ARGV[0] or raise('Please provide the email address for your account')
password = ARGV[1] or raise('Please provide the password for your account')
path     = ARGV[2] || './' 

download = lambda do |url, file|
  agent = Mechanize.new
  agent.get 'https://www.destroyallsoftware.com/screencasts/users/sign_in'
  form  = agent.page.forms.first
  form['user[email]']    = email
  form['user[password]'] = password
  form.submit
  agent.pluggable_parser.default = Mechanize::Download
  agent.get(url).save(file)
end

agent = Mechanize.new
agent.get 'https://www.destroyallsoftware.com/screencasts/catalog'
screencasts = agent.page.search('li.screencast')

while screencast = screencasts.pop
  title = screencast.search('a').first.text
  url   = screencast.search('.download_link > a:first-child').first['href']
  index = screencasts.size
  file  = "#{path}/#{"%03d" % (index + 1)}-#{title.gsub(/\.|:|,/, '').gsub(/\/|\s/, '-').downcase}.mov"
  puts "Downloading #{title} - #{index} to go"
  next puts 'Already Downloaded' if File.exist? file
  download[url, file]
end

Allright! thanks! Excerpts from Wael M. Nasreddine's message of Mon Feb 27 17:54:21 -0600 2012:

…

Hey @maca, really useful Gist, thank you I've updated it a little bit, basically, what you're doing is downloading the screencasts in a reverse order, so the first screencast is actually number 54 and the last screencast (the newest) is screencast 1, which is just wrong, because the next time a screencast is released, your script will download all of them all over again because the numbering has changed (first or oldest one becomes 55 by then) Check the gist here https://gist.github.com/1927927/25831a6c53734b90d627175147f332ce22089fdb --- Reply to this email directly or view it on GitHub: https://gist.github.com/1798070

No Problem, Thank you :)

nice one

I would actually not prefix the index to the file name at all. I tried running this script again and the index to screencast mapping changed.

Actually, this is what worked for me:

...
agent = Mechanize.new
agent.get 'https://www.destroyallsoftware.com/screencasts/catalog'
screencasts = agent.page.search('li.screencast')
num = screencasts.size
index = 0

while screencast = screencasts.pop
title = screencast.search('a').first.text
url = screencast.search('.download_link > a:first-child').first['href']
index += 1
file = "#{path}/#{"%04d" % (index)}-#{title.gsub(/.|:|,/, '').gsub(//|\s/, '-').downcase}.mov"
puts "Downloading #{title} - #{num - index} to go"
next puts 'Already Downloaded' if File.exist? file
download[url, file]
end

made some changes to suit my prefs and updated to match the new layout of DAS: https://gist.github.com/jasondew/5583811

i made a realllllllllllly hacky version of this. however, it works and downloads in parallel!

#! /usr/bin/env ruby
# usage:
# $ das_download.rb email password [download_directory]

require 'mechanize'
# gem 'mechanize-progressbar'

email    = ARGV[0] or raise('Please provide the email address for your account')
password = ARGV[1] or raise('Please provide the password for your account')
agent = Mechanize.new

agent.user_agent = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/54.0.2840.100 Safari/537.36'
agent.get 'https://www.destroyallsoftware.com/screencasts/users/sign_in'
form = agent.page.forms.first
form['user[email]']    = email
form['user[password]'] = password
form.submit
agent.pluggable_parser.default = Mechanize::Download

agent.get 'https://www.destroyallsoftware.com/screencasts/catalog'
episodes = agent.page.search('.episode > a').map { |a| a['href'] }

episodes_meta = episodes.map do |screencast, index|
  page = agent.get 'https://www.destroyallsoftware.com' + screencast
  # take a bath after this
  url = page.search('video + script').first.children.first.to_s.match(/\s+source\.src\s=\s"(.+?)"/)[1]
  title = URI(url).path[1..-1]
  { url: url, title: title }
end

puts "About to download #{episodes_meta.length} episodes... hold tight."

episodes_meta.each_slice(14) do |slice|
  fork do
    slice.each do |meta|
      system('wget', '--quiet', '-O', meta[:title], meta[:url])
      puts "Finished #{meta[:title]}"
    end
  end
end

Process.waitall

@ianks Thanks a ton for this script! Note that you have to have wget on your system for this to work. Mine failed silently and this took a while to debug.

I wrote a small script to grab all the compendium articles, with their images and css so you can read and review offline: https://gist.github.com/AlessandroMinali/fbb9532d5db1f568481bca1f9c2cb9f5

I've forked https://gist.github.com/jasondew/5583811 and updated it to work with the latest DAS which is free this week: https://gist.github.com/jaredculp/f26f83d214cf926472dddd4269bd2538

On season three and haven't hit any problems. Thanks for this!

Hacked together another fork https://gist.github.com/itsgoingd/4e6f9b663a825143ebd6997806931e73

• based on the most up to date version https://gist.github.com/elochim-siemasz/64edda44c2fcfc28d6b7eed9f78cd62c
• re-added login support from https://gist.github.com/jasondew/5583811
• tweaked for a slight html change on the site (title images no longer have alt text)

Updated @itsgoingd's version in Apr 2021 https://gist.github.com/lukeholder/64f6a3bbc97c050ddfa605fa4d5dff8c

• Added the episode number to the filename and meta data
• Cleaned up the title format (squish!)
• Download as.mp4 extension
• Removed tags ("Watched", "Free") from episode names.