ニコニコ動画の検索APIで特定のタグの投稿の新しい順に検索し、情報を拾ってくるスクリプト その2

Nicovideo.rb

#!/usr/bin/env ruby
# -*- coding: utf-8 -*-

if RUBY_VERSION < '1.9'
  require 'rubygems'
end

require 'json'
require 'mechanize'
require 'cgi'
require 'rexml/document'

class NicoVideo
  
  def getVideoList
    nico_id = ''
    nico_pass = ''
    secure_uri = 'https://secure.nicovideo.jp/secure/login?site=niconico'
    agent = Mechanize.new
    agent.user_agent = 'NicoLiveAlertTwitterBot 0.2'
    agent.post(secure_uri,'mail' => nico_id ,'password' => nico_pass)
    cookie = agent.cookie_jar 
    search_uri = 'http://ext.nicovideo.jp/api/search/tag/'
    search_word = CGI.escape('アイドルマスター')
    51200.times do |i|
      i += 1
      html = agent.get("#{search_uri}#{search_word}?mode=watch&order=d&page=#{i}&sort=f", {'Cookie' => cookie} )
      str = html.body
      str.gsub!(/\\u([0-9a-f]{4})/) {
        [$1.hex].pack("U")
      }
      
      json = JSON.parser.new(str)
      hash = json.parse()
      lists = hash['list']
      next unless lists
      File::open("Nicovideo_List.csv","w"){|f|
        lists.map do |list|
          thumb = agent.get("http://ext.nicovideo.jp/api/getthumbinfo/#{list['id']}")
          next unless thumb
          source = thumb.body
          next unless source
          info = REXML::Document.new source
          next unless info
          f.puts "#{list['id']},#{list['title']},#{info.elements['nicovideo_thumb_response/thumb/user_id'].text},#{list['view_counter']},#{list['num_res']},#{list['mylist_counter']}"
          puts info.elements['nicovideo_thumb_response/thumb/title'].text
          break if list['id'] == "sm15177209"
          sleep 0.1
        end
      }
    end
  end
end

nicovideo = NicoVideo.new
nicovideo.getVideoList