Script to download all screencasts from ElixirSips using the RSS feed. Should work for other podcast feeds too.

get_all.rb

#! /usr/bin/env ruby

require "nokogiri"
require "date"
require "fileutils"
require "pry"

class GetAll
  def self.run
    check_upass
    fetch_all
  end
  
  UPASS = ENV["UPASS"]

  def self.check_upass
    raise "Define UPASS=email@address.net:p4ssw0rd in the environment" unless UPASS.to_s.size > 0
  end

  def self.fetch_all
    xml = Nokogiri.XML(File.open("feed.xml"))

    xml.xpath("/rss/channel/item").each do |node|
      new(node).fetch
    end
  end

  def initialize(node)
    @node = node
  end
  attr_reader :node

  def fetch
    FileUtils.mkdir_p dir

    Dir.chdir dir do
      File.write("episode.html", description)

      if enclosure
        fetch_per_enclosure
      else
        puts "No enclosure for #{dir}, looking at text content…"
        fetch_from_links
      end
    end
  end

  def fetch_per_enclosure
    url = enclosure["url"]
    size = enclosure["length"].to_i
    name = File.basename(url)
    if File.exists?(name) && File.size(name) == size
      # puts "#{dir} has mp4, skipping"
    else
      puts "Fetching #{name} to #{dir}:"
      system "curl", "-u", UPASS, "-kOL", url
    end
  end

  def fetch_from_links
    html = Nokogiri.HTML(description)
    first, *more = html.xpath("//ul/li/a").select {|n| n.text[/\.mp4$/] }
    if first.nil?
      puts "Nothing in content either; skipping"
      return
    elsif more.any?
      puts "*** Found #{1 + more.size} mp4s; don't know what to do!"
      return
    else
      name = first.text
      if File.exists?(name) and false
        puts "#{dir} has mp4 (complete or incomplete), skipping"
      else
        file_id = first["href"][/file_id=(\d+)/, 1]
        url = "https://elixirsips.dpdcart.com/subscriber/download?file_id=#{file_id}"
        puts "Fetching #{name} from #{url} to #{dir}:"
        puts ">>> No, not actually since it never seems to work"
        # system "curl", "-u", UPASS, "-kLo", name, url
      end
    end
  end

  def date
    DateTime.parse(node.at("pubDate").text)
  end

  def clean_name
    node.at("title").text.downcase.gsub(/[^-a-z0-9.]+/, "_")
  end

  def dir
    [date.strftime("%Y-%m-%d"), clean_name].join("-")
  end

  def description
    node.at("description").text
  end

  def enclosure
    node.at("enclosure")
  end
end

GetAll.run if $0 == __FILE__