lib/grabepg/grab_tvsou.rb in grab_epg-0.2.1 vs lib/grabepg/grab_tvsou.rb in grab_epg-0.2.3

@@ -76,10 +76,11 @@
         channellist
       }
 
 
       doc = @grabbase.get_doc_with_proxy(@proxy_list,@home_page)
+      begin
       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"
@@ -88,10 +89,19 @@
 
           when "地方"
 
         end
       end
+      @error_num=0
+      rescue
+        unless @error_num
+          @error_num = 0
+        end
+        @error_num+=1
+        raise err.to_s  if @error_num==5
+        dispose_home_page
+      end
       return @channels
     end
 
 
     #获取频道列表
@@ -121,11 +131,21 @@
 
    #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("?")
+     begin
      doc = @grabbase.get_doc_with_proxy(@proxy_list,url)
+     @error_num = 0
+     rescue => err
+       unless @error_num
+         @error_num = 0
+       end
+       @error_num+=1
+       raise err.to_s  if @error_num==5
+       dispose_schedule_page(url,start_time,use_time)
+     end
      _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
@@ -133,11 +153,13 @@
      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|
+         div = doc.css('div[class="time"]')[0]
+         if div
+         div.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]
@@ -151,10 +173,13 @@
            rescue => err
              p "Schedule: #{schedule}"
            end
          end
          ret.merge!({url[:date]=>schedules})
+         else
+           p "Error In this url: #{url} couldn't get doc.css('div[class=time]')[0]"
+         end
       end
      end
      return ret
    end
 
@@ -163,14 +188,24 @@
 
 
     #解析节目详情页面
     def dispose_show_info(url)
       doc = @grabbase.get_doc_with_proxy(@proxy_list,url)
+      begin
       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]","")
+      @error_num = 0
       {show_name:show_name,img_url:img_url,show_info:show_info}
+      rescue => err
+        unless @error_num
+          @error_num = 0
+        end
+        @error_num+=1
+        raise err.to_s  if @error_num==5
+        dispose_show_info(url)
+      end
     end
 
 
   end
 
\ No newline at end of file