#encoding:utf-8

require File.expand_path("../grab_base.rb", __FILE__)
module Grabepg
  # To change this template use File | Settings | File Templates.


  class GrabTvsou
    include Grabepg
    #首页
    attr_reader :home_page

    #代理列表
    attr_reader :proxy_list

    attr_reader :grabbase

    #频道存储
    attr_reader :channels

    #时间表存储
    attr_reader :schedules

    #俩个节目间的最小间隔时间
    attr_reader :default_min_interval


    ChannelTypeMap = {"yg_ys_li"=>"央视","yg_ws_li"=>"卫视","yg_hw_li"=>"海外","yg_df_li"=>"地方"}

    #type 从mobie还是网站接口抓取数据
    def initialize(grabtype,proxy_list)
      @home_page = get_url(grabtype)
      @proxy_list = proxy_list
      @grabbase = GrabBase.new
      @channels = {}
      @site="http://m.tvsou.com"
    end

    #获取从tvsou的什么网站上获取
    #type: mobile,webpage
    def get_url(type)
      return "http://m.tvsou.com/index.asp" if type.eql?("mobile")
    end

    def get_data_year_month_day(time)

      month = time.month.to_s
      if month.length<2
        month = "0"+month
      end
      return {time:"#{time.year}-#{time.month}-#{time.day}",date:"#{@grabbase.conversion_what_day(time.wday)}(#{month}-#{time.day})"}
    end

    #获取时间
    #start_time 时间起始点
    #use_time   天数
    def get_data(start_time,use_time)
      time = Time.now+start_time*24*60*60
      ret = []
      use_time.times.each do |i|
        _time = time + i*24*60*60
        ret << get_data_year_month_day(_time)
      end
      ret
    end


    #对首页进行处理获取部分频道的URL和嘻嘻
    def dispose_home_page

      get_channellist = lambda { |li,type|
        channellist = {}
        li.css('a').each do |a|
          channellist.merge!({a.content=>{url:a.get_attribute("href"),type:type}}) unless channellist.has_key?(a.content)
        end
        channellist
      }


      doc = @grabbase.get_doc_with_proxy(@proxy_list,@home_page)
      doc.css("li").each do |li|
        case ChannelTypeMap[li.get_attribute("class")]
          when "央视"
            @channels.merge!(get_channellist.call(li,"CCTV"))
          when "卫视"
            @channels.merge!(get_channellist.call(li,"WTV"))
          when "海外"

          when "地方"

        end
      end
      return @channels
    end


    #获取频道列表
    #url是获取频道列表的首页
    #地方需要调用此函数
    def dispose_channel_page(url,channel_type)

    end



    #获取频道时间表URL
    def dispose_href_schedule_data(href,start_time,use_time)
     hrefs=href.split("&programDT=")
     _hrefs=hrefs[1].split("&")
     ret = []
     get_data(start_time,use_time).each do |time|
       _hrefs[0]=time[:time]
       url = hrefs[0]+"&programDT=" + time[:time]
       1.upto(_hrefs.length-1).each do |i|
         url += "&"+_hrefs[i]
       end
       ret<<{url:url,time:time[:time],date:time[:date]}
     end
     ret
    end

   #根据URL解析时间表页面
   def dispose_schedule_page(url,start_time,use_time)
     url = @site +"/"+url
     urls = url.split("?")
     doc = @grabbase.get_doc_with_proxy(@proxy_list,url)
     _url = doc.css("div[class='week']")[0].css('a')[0].get_attribute("href")
     _url = urls[0]+_url
     urls = dispose_href_schedule_data(_url,start_time,use_time)
     ret = {}
     last_time = -5
     last_schedule = {}
     urls.each do |url|
       p "Grab url: #{url}"
       if url
         doc = @grabbase.get_doc_with_proxy(@proxy_list,url[:url])
         schedules = []
         doc.css('div[class="time"]')[0].css("li[class='gray']").each do |schedule|
           begin
             _dispose = schedule.content
             _dispose_show =schedule.css("span")[0].text
             time = _dispose.gsub(_dispose_show,"")
             _url = @site+"/" + schedule.css('a')[0].get_attribute("href") if schedule.css('a')[0]
             schedules << {time:time,schedule_name:_dispose_show.delete(" 剧情"),url:_url}
             now = time.gsub(":","").to_i
             if((now-last_time)<5)
               schedules.delete(last_schedule)
             end
             last_schedule = {time:time,schedule_name:_dispose_show.gsub(" 剧情",""),url:_url}
             last_time = now
           rescue => err
             p "Schedule: #{schedule}"
           end
         end
         ret.merge!({url[:date]=>schedules})
      end
     end
     return ret
   end





    #解析节目详情页面
    def dispose_show_info(url)
      doc = @grabbase.get_doc_with_proxy(@proxy_list,url)
      show_name = doc.css('div[class="tv_info_top"]')[0].content
      _doc=doc.css("div[class='tv_info']")
      img_url = _doc.css("img")[0].get_attribute("src").gsub(" ","")
      show_info = _doc.css("p")[0].content.gsub("[全文]","")
      {show_name:show_name,img_url:img_url,show_info:show_info}
    end


  end

end