src/websites/youtube.rb

   1 #
   2 # Copyright Michael Orlitzky
   3 #
   4 # http://michael.orlitzky.com/
   5 #
   6 # This program is free software: you can redistribute it and/or modify
   7 # it under the terms of the GNU General Public License as published by
   8 # the Free Software Foundation, either version 3 of the License, or
   9 # (at your option) any later version.
  10 #
  11 # This program is distributed in the hope that it will be useful,
  12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  14 # GNU General Public License for more details.
  15 #
  16 # http://www.fsf.org/licensing/licenses/gpl.html
  17 #
  18
  19 require 'src/website'
  20
  21 # Needed to download the page, which is in turn
  22 # needed because it contains the video URL.
  23 require 'net/http'
  24 require 'uri'
  25
  26
  27 class Youtube < Website
  28
  29   VALID_YOUTUBE_URL_REGEX = /^(http:\/\/)?([a-z0-9]+\.)?youtube\.com\/((watch\?v=)|(v\/))[a-z0-9_\-]+(\&.*)?\#?$/i
  30
  31   def self.owns_url?(url)
  32     return url =~ VALID_YOUTUBE_URL_REGEX
  33   end
  34
  35
  36   def get_video_url()
  37     video_id = self.parse_video_id()
  38
  39     # The video's URL (the "page data" URL) may be different from the
  40     # URL that was passed to the program. We support the /v/video_id
  41     # URL format, but that is *not* the main video page where we can
  42     # retrieve the "t" parameter. We can only get that from the
  43     # /watch?v=video_id form.
  44     page_data_url = "http://www.youtube.com/watch?v=#{video_id}"
  45     page_data = self.get_page_data(page_data_url)
  46
  47     # Magic.
  48     t_parameter = self.parse_t_parameter(page_data)
  49
  50     video_url = "http://www.youtube.com/get_video?video_id=#{video_id}&t=#{t_parameter}"
  51
  52     return video_url
  53   end
  54
  55
  56   def get_video_filename()
  57     return (self.parse_video_id() + '.flv')
  58   end
  59
  60
  61   protected;
  62
  63   # Get the video id from the URL. Should be relatively easy,
  64   # unless Youtube supports some URL formats of which I'm unaware.
  65   def parse_video_id()
  66     # Return nil if we get no matches below.
  67     video_id = nil
  68
  69     # Both URLs are fairly easy to parse if you handle
  70     # them one at a time. The only tricky situation is when
  71     # parameters like "&hl=en" are tacked on to the end.
  72     # We'll call /watch?v=video_id the "first form."
  73     first_form_video_id_regex = /v=([0-9a-z_\-]+)/i
  74     first_form_matches = first_form_video_id_regex.match(@url)
  75     return first_form_matches[1] if not (first_form_matches.nil? ||
  76                                          first_form_matches.length < 2)
  77
  78     # First form didn't work? Try the second.
  79     second_form_video_id_regex = /\/v\/([0-9a-z_\-]+)/i
  80     second_form_matches = second_form_video_id_regex.match(@url)
  81     video_id = second_form_matches[1] if not (second_form_matches.nil? ||
  82                                               second_form_matches.length < 2)
  83
  84     return video_id
  85   end
  86
  87
  88   # Parse out the "t" parameter from the video's page. I'm not sure
  89   # what "t" stands for, but it's located in some JSON, and is required
  90   # for the final video URL to work.
  91   def parse_t_parameter(page_data)
  92     t_parameter = nil
  93
  94     t_parameter_regex = /\"t\"\:[[:space:]]\"([^\"]+?)\"/
  95     matches = t_parameter_regex.match(page_data)
  96     t_parameter = matches[1] if not (matches.nil? || matches.length < 2)
  97
  98     return t_parameter
  99   end
 100
 101
 102   def get_page_data(url)
 103     uri = URI.parse(url)
 104
 105     response = Net::HTTP.start(uri.host, uri.port) do |http|
 106       http.get(uri.request_uri)
 107     end
 108
 109     return response.body
 110   end
 111
 112 end