update

task/spark_test.py

@@ -98,7 +98,14 @@ for t in range(2, task_days):
                       AND page_name in ('search_result_more','search_result_welfare'))
                    or (action = 'goto_welfare_detail' AND params ['from'] = 'search_result_welfare_recommend')
                    or (action = 'on_click_card' AND params['card_content_type'] in ('service') AND page_name in ('search_result_more','search_result_welfare')))
-       """.format(partition_day="20210224", end_date="20210225", tomorrow_str=tomorrow_str)
+                  UNION ALL
+SELECT *  FROM online.bl_hdfs_maidian_updates
+                         WHERE partition_date >= '{partition_day}'
+                           AND partition_date < '{end_date}'
+                           AND ((action in ('on_click_topic_card','on_click_diary_card','search_result_click_infomation_item')
+                               AND page_name in ('search_result_more','search_result_diary','search_result_post'))
+                               or (action = 'on_click_card' AND params['card_content_type'] in ('answer','diary') AND page_name in ('search_result_more','search_result_diary','search_result_question_answer')))
+    """.format(partition_day="20210224", end_date="20210225", tomorrow_str=tomorrow_str)
 
     #   print(new_urser_device_id_sql)
     new_urser_device_id_df = spark.sql(new_urser_device_id_sql)
@@ -107,7 +114,7 @@ for t in range(2, task_days):
     sql_res = new_urser_device_id_df.collect()
     res_list = []
     for res in sql_res:
-        print(res)
+        # print(res)
         query = res.params.get("query","").replace("\r","").replace("\n","")
         card_name = res.params.get("card_name","").replace("\r","").replace("\n","")
         card_id = res.params.get("card_id","")