src/website.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 # Necessary in a lot of subclasses; plus, we need it
  20 # to parse the server name out of our URL.
  21 require 'uri'
  22
  23 # This class keeps track of all its subclasses
  24 # We use this to loop through every "website" in an
  25 # attempt to determine to which site a URL belongs.
  26 class Website
  27
  28   protected;
  29
  30   @url = nil
  31
  32
  33   def self.inherited(subclass)
  34     if superclass.respond_to? :inherited
  35       superclass.inherited(subclass)
  36     end
  37
  38     # Every time we're subclassed, add the new
  39     # subclass to our list of subclasses.
  40     @subclasses ||= []
  41     @subclasses << subclass
  42   end
  43
  44
  45   def server
  46     # Get the HTTP server portion of our URI
  47     uri = URI.parse(@url)
  48     return uri.host
  49   end
  50
  51
  52
  53   def get_page_data(url)
  54     # A naive implementation that just grabs the
  55     # data from a page.
  56     uri = URI.parse(url)
  57
  58     response = Net::HTTP.start(uri.host, uri.port) do |http|
  59       http.get(uri.request_uri)
  60     end
  61
  62     return response.body
  63   end
  64
  65
  66
  67   public;
  68
  69   def initialize(url)
  70     @url = url
  71   end
  72
  73
  74   def self.create(url)
  75     # Factory method returning an instance of
  76     # the appropriate subclass.
  77
  78     # Check the URL against each website's class.
  79     # The class will know whether or not the URL
  80     # "belongs" to its website.
  81     @subclasses.each do |w|
  82       if w.owns_url?(url)
  83         return w.new(url)
  84       end
  85     end
  86
  87     # If nothing matched, we don't return an instance
  88     # of anything.
  89     return nil
  90   end
  91
  92
  93   # Abstract definition. Each subclass of Website
  94   # should support it on its own.
  95   def self.owns_url?(url)
  96     raise NotImplementedError
  97   end
  98
  99
 100   # Same here. Abstract.
 101   def get_video_url()
 102     raise NotImplementedError
 103   end
 104
 105
 106   # The website class should be responsible for determining the
 107   # video's filename. By default, we can take the last component
 108   # of the video URL, but in some cases, subclasses will want
 109   # to override this behavior.
 110   def get_video_filename()
 111     # Use whatever comes after the final front slash.
 112     return get_video_url().split('/').pop()
 113   end
 114
 115 end