Dagens Nyheter (DN) PDF downloader for paying subscribers.

dnloader.rb

# Ruby script for DN subscribers to download the latest DN as a PDF – because it beats their iPad app.
# By Henrik Nyh 2015-12-08 under the MIT license.
#
# INSTRUCTIONS (for advanced users)
#
# Fetch the script dependencies:
#
#     (sudo) gem install mechanize
#
# Put your username and password in a ~/.dnloader file separarated by a ":", e.g.:
#
#     echo "[email protected]:myp4ssw0rd" > ~/.dnloader
#
# Optionally, customize where the files end up, here:

DEST_DIR = File.expand_path("~/Dropbox/DN")

# Optionally, customize how many issues to keep:

MAX_ISSUES = 14

# Now make sure the script works:
#
#     ruby dnloader.rb
#
# You can add a cron job to run this script every few minutes. (If the file is already downloaded, nothing happens.)
# E.g. to run it every 5 minutes:
#
#     */5 * * * * ruby /Users/foo/bin/dnloader.rb > /dev/null

require "mechanize"
require "date"

DATE = Date.today
PDF_INDEX_URL = "https://kund.dn.se/mitt-konto/dn-som-pdf/"
PDF_PATH = "/service/download/#{DATE.strftime("%Y%m%d")}/DN.pdf"
PDF_URL = "https://kund.dn.se#{PDF_PATH}"
PDF_DEST = "#{DEST_DIR}/DN_#{DATE}.pdf"
REMOVAL_GLOB = "#{DEST_DIR}/DN_*.pdf"

if File.exist?(PDF_DEST)
  puts "Already downloaded: #{PDF_DEST}"
  exit
end

FileUtils.mkdir_p(DEST_DIR)

CONFIG_FILE = File.expand_path("~/.dnloader")
abort("Missing config file! See docs.") unless File.exist?(CONFIG_FILE)

username, password = File.read(CONFIG_FILE).strip.split(":", 2)
abort("Missing username or password!") unless username && password


# Log in and download (we need a session to get the PDF)

agent = Mechanize.new
agent.pluggable_parser.pdf = Mechanize::Download

login_page = agent.get("https://auth.dn.se/login?appId=dagensnyheter.se&lc=sv&callback=http%3A%2F%2Fkund.dn.se%2Fservice%2Floginplus%3Fredirect%3D%2F")

login_result_page = login_page.form_with(id: "loginForm") do |form|
  form.field_with(name: "form.username").value = username
  form.field_with(name: "form.password").value = password
end.submit

if login_result_page.body.include?("Logga ut")

  # For some reason, they will sometimes make "today's" PDF available right after midnight but with yesterday's content.
  # But they don't seem to update links until there is a real issue, so we use that.
  page = agent.get(PDF_INDEX_URL)
  unless page.links_with(href: PDF_PATH).any?
    puts "Seems today's issue is not published yet."
    exit
  end

  agent.get(PDF_URL).save(PDF_DEST)
  puts "Downloaded: #{PDF_DEST}"
else
  # If you want to debug it:
  #p login_result_page
  #puts login_result_page.body

  abort "Error logging in!"
end


# Remove old files.

files_to_remove = Dir[REMOVAL_GLOB].sort.reverse.drop(MAX_ISSUES)

if files_to_remove.any?
  FileUtils.rm(files_to_remove)
  puts "Removed files: #{files_to_remove.inspect}"
end

This goes together well with the "synced folders" feature of GoodReader for the iPad.