Script for downloading all WWDC videos from a JSON feed

download_wwdc.rb

#!/usr/bin/env ruby

# 06/2021: https://devimages-cdn.apple.com/wwdc-services/pb9e2d31/0CF104D6-F0B3-4AE8-B9F0-6F2D9F54A502/contents.json
# 11/2020: https://devimages-cdn.apple.com/wwdc-services/n233a99f/5D23F1E9-9551-4768-ACF3-E3920F9C572D/contents.json
#  5/2019: https://devimages-cdn.apple.com/wwdc-services/j06970e2/296E57DA-8CE8-4526-9A3E-F0D0E8BD6543/contents.json  
#  8/2017: https://devimages-cdn.apple.com/wwdc-services/h8a19f8f/049CCC2F-0D8A-4F7D-BAB9-2D8F5BAA7030/contents.json
#  5/2016: http://devimages.apple.com.edgekey.net/wwdc-services/g7tk3guq/xhgbpyutb6wvn2xcrbcz/videos.json

require 'fileutils'
require 'json'

if ARGV.count != 2
  puts "Usage: #{$PROGRAM_NAME} <feed.json> <target_dir>"
  exit 1
end

feed_file, target_dir = ARGV

if !File.exist?(feed_file)
  puts "File #{feed_file} doesn't exist"
  exit 1
end

if !File.exist?(target_dir)
  FileUtils.mkdir_p(target_dir)
elsif !File.directory?(target_dir)
  puts "#{target_dir} is not a directory"
  exit 1
end

def download(session, url, dir)
  filename = url.split('/').last
  id = session['id'].split('-').last.to_i

  if filename.gsub(/[hs]d/i, '').gsub(/wwdc/i, '').gsub(/\.\w+$/, '') !~ /[a-z]/
    title = session['title'].downcase
      .gsub(/\-/, '_')
      .gsub(/@/, '_at_')
      .gsub(/[^\w\s]+/, '')
      .gsub(/\s+/, '_')
    filename = filename.sub(/#{session['eventId']}[_\-]/, '')
      .sub("#{id}_", "#{id}_#{title}_")
      .sub("#{id}-", "#{id}-#{title}-")
      .sub("#{id}.", "#{id}-#{title}.")
  end

  path = File.join(dir, filename)

  return if File.exist?(path)

  tries = 0
  max = 20

  while tries < max
    begin
      puts ">> (#{tries}) #{path}"
      puts
      system("curl -C - -o \"#{path}\" #{url}", exception: true)
      break
    rescue Exception => e
      tries += 1
    end
  end

  if tries == max
    File.delete(path)
    exit 1
  end
end

feed = JSON.parse(File.read(feed_file))

if feed.is_a?(Hash) && feed['contents']
  sessions = feed['contents']
    .select { |x| x['media'] && x['media']['downloadHD'] }
    .sort_by { |x| w = x['id'].split('-'); [w[0...-1].join, -w.last.to_i] }
    .reverse

  sessions.each do |session|
    event = session['eventId'].gsub(/wwdc/, '')
    subdir = File.join(target_dir, event)
    FileUtils.mkdir_p(subdir)

    download(session, session['media']['downloadHD'], subdir)
    download(session, session['media']['slides'], subdir) if session['media']['slides']
  end
elsif feed.is_a?(Hash) && feed['sessions']
  feed['sessions'].select { |x| x['date'] == 2012 }.each do |session|
    subdir = File.join(target_dir, '2012')
    FileUtils.mkdir_p(subdir)

    title_slug = session['title'].downcase
      .gsub(/([a-z])[\-\']([a-z])/, "\\1\\2").gsub(/: /, '__').gsub(/ \+ /, '__').gsub(/\W+/, '_')

    filename = "session_#{session['id']}__#{title_slug}.mov"
    path = File.join(subdir, filename)

    unless File.exist?(path)
      puts "https://download.developer.apple.com/videos/wwdc_2012__hd/#{filename}"
    end
  end
elsif feed.is_a?(Array)
  feed.each do |session|
    subdir = File.join(target_dir, session['year'])
    FileUtils.mkdir_p(subdir)

    ['HD', 'PDF'].each do |type|
      link = session['links'].detect { |l| l[0] == type }
      next unless link

      url = link[1]
      filename = url.split('/').last
      path = File.join(subdir, filename)

      unless File.exist?(path)
        puts "#{type.rjust(3)}: #{url}"
      end
    end
    puts
  end
end