#
# Copyright Michael Orlitzky
#
# http://michael.orlitzky.com/
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# http://www.fsf.org/licensing/licenses/gpl.html
#

require 'src/website'
require 'cgi'

class Bliptv < Website

  VALID_BLIPTV_URL_REGEX = /^(http:\/\/)?([[:alnum:]\-]+\.)?blip\.tv\/file\/(\d+)(.*)?$/
  VALID_BLIPTV_REDIR_URL_REGEX = /^(http:\/\/)?([[:alnum:]\-]+\.)?blip\.tv\/play\/[[:alnum:]_]+$/

  def self.owns_url?(url)
    return (url =~ VALID_BLIPTV_URL_REGEX ||
            url =~ VALID_BLIPTV_REDIR_URL_REGEX)
  end


  def get_video_url()
    page_data = ''

    if (@url =~ VALID_BLIPTV_URL_REGEX)
      page_data = self.get_page_data(@url)
    else
      # It's a redirect. Figure out the RSS page URL from the redirect.
      redir_url = get_redirect_url
      rss_page_url = parse_rss_url(redir_url)
      rss_data = get_page_data(rss_page_url)

      # The "real" page URL is embedded in the RSS feed. Once we get the
      # real URL, we can proceed as if we were given that URL in the first
      # place.
      real_page_url = parse_page_url(rss_data)
      page_data = self.get_page_data(real_page_url)
    end

    filepath = parse_video_url(page_data)

    return filepath
  end


  protected;

  def parse_page_url(data)
    # A simplified VALID_BLIPTV_URL_REGEX.
    page_url_regex = /http:\/\/blip\.tv\/file\/\d+/
    matches = page_url_regex.match(data)

    if matches.nil?
      raise StandardError.new("Couldn't parse the real page URL from the RSS page.")
    end

    return matches[0]
  end

  def parse_rss_url(url)
    rss_id_regex = /\/flash\/(\d+)/
    matches = rss_id_regex.match(url)

    if matches.nil? or (matches.length < 2)
      raise StandardError.new("Couldn't parse the video ID from the redirect URL: #{url}")
    end

    return "http://blip.tv/rss/flash/#{matches[1]}"
  end

  def get_redirect_url
    uri = URI.parse(@url)

    response = Net::HTTP.start(uri.host, uri.port) do |http|
      http.get(uri.request_uri, {})
    end

    return CGI::unescape(response['location'])
  end


  def parse_video_url(page_data)
    # First, try to find the MOV video. The source videos are usually
    # encoded with MOV.
    video_url_regex = /"Quicktime \(\.mov\)", "attribute" : "(.*?\.mov)/i
    matches = video_url_regex.match(page_data)

    if not matches.nil? and (matches.length > 1)
      return matches[1]
    end

    # I've seen some free software videos encoded as OGG/Vorbis, too.
    video_url_regex = /"Ogg Theora\/Vorbis \(\.og[gv]\)", "attribute" : "(.*?\.og[gv])/i
    matches = video_url_regex.match(page_data)

    if not matches.nil? and (matches.length > 1)
      return matches[1]
    end

    # If that didn't work, try the WMV format, which is occasionally
    # used for the source as well.
    video_url_regex = /"Windows Media \(\.wmv\)", "attribute" : "(.*?\.wmv)/i
    matches = video_url_regex.match(page_data)

    if not matches.nil? and (matches.length > 1)
      return matches[1]
    end


    # If neither of the source formats are present, just grab the
    # video URL from the Flash variable and be done with it.
    video_url_regex = /setPrimaryMediaUrl\("(.*?\.(flv|mov|wmv|mp4|og[gv]))/i
    matches = video_url_regex.match(page_data)

    if matches.nil? or (matches.length < 2)
      raise StandardError.new("Couldn't parse any of the video format URLs.")
    end

    return matches[1]
  end

end