update

task/recommend_strategy_d.py

@@ -19,8 +19,8 @@ import sys
 import time
 from pyspark import SparkConf
 from pyspark.sql import SparkSession, DataFrame
-from pyspark.sql.functions import lit
-import pytispark.pytispark as pti
+# from pyspark.sql.functions import lit
+# import pytispark.pytispark as pti
 
 db = pymysql.connect(host='172.16.40.158', port=4000, user='st_user', passwd='aqpuBLYzEV7tML5RPsN1pntUzFy',
                      db='jerry_prod')
@@ -87,7 +87,9 @@ one_week_age_str = (now + datetime.timedelta(days=-7)).strftime("%Y%m%d")
 grey_select_sql = """SELECT *,
     NVL(ROUND((navbar_search+highlight_word+self_welfare_card+recommend_welfare_card+recommend_content_card*0.2+transfer_card+video_consultation)/card_exposure,5),0) as recommend_ctr,
     NVL(ROUND(card_click/card_exposure,5),0) as click_ctr,
-    NVL(ROUND((navbar_search+highlight_word+self_welfare_card+recommend_welfare_card+recommend_content_card*0.2+transfer_card+video_consultation)/card_click,5),0) as second_jump_ctr FROM pm.tl_pm_recommend_strategy_d_v2 where day_id >= {day_id}""".format(day_id=one_week_age_str)
+    NVL(ROUND((navbar_search+highlight_word+self_welfare_card+recommend_welfare_card+recommend_content_card*0.2+transfer_card+video_consultation)/card_click,5),0) as second_jump_ctr ,
+    NVL(ROUND((page_pv_20/page_pv,5),0) as page_pv_20_percent 
+    FROM pm.tl_pm_recommend_strategy_d_v2"""
 device_df = spark.sql(grey_select_sql)
 
 device_df.show(1, False)
@@ -106,6 +108,9 @@ for res in sql_res:
     navbar_search = res.navbar_search
     highlight_word = res.highlight_word
     self_welfare_card = res.self_welfare_card
+    page_pv_20 = res.page_pv_20
+    page_pv_20_percent = res.page_pv_20_percent
+
 
     recommend_welfare_card = res.recommend_welfare_card
     recommend_content_card = res.recommend_content_card
@@ -124,14 +129,14 @@ for res in sql_res:
     instert_sql = """replace into recommend_strategy_d(
     day_id,device_os_type,active_type,card_content_type,recommend_type,card_click,card_exposure,navbar_search,
     highlight_word,self_welfare_card,recommend_welfare_card,recommend_content_card,recommend_special_card,transfer_card,video_consultation,
-    partition_day,pid,recommend_ctr,second_jump_ctr,click_ctr
+    partition_day,pid,recommend_ctr,second_jump_ctr,click_ctr,page_pv_20_percent
     ) VALUES('{day_id}','{device_os_type}','{active_type}','{card_content_type}','{recommend_type}',{card_click},{card_exposure},
     {avg_page_stay},{navbar_search},{self_welfare_card},{recommend_welfare_card},{recommend_content_card},{recommend_special_card},
-    {transfer_card},{video_consultation},'{partition_day}','{pid}',{recommend_ctr},{second_jump_ctr},{click_ctr});""".format(
+    {transfer_card},{video_consultation},'{partition_day}','{pid}',{recommend_ctr},{second_jump_ctr},{click_ctr},{page_pv_20_percent});""".format(
         day_id=day_id, device_os_type=device_os_type, active_type=active_type, card_content_type=card_content_type,
         card_click=card_click, recommend_type=recommend_type, card_exposure=card_exposure, avg_page_stay=avg_page_stay,
         navbar_search=navbar_search, self_welfare_card=self_welfare_card, recommend_welfare_card=recommend_welfare_card,
-        recommend_content_card=recommend_content_card, recommend_special_card=recommend_special_card,
+        recommend_content_card=recommend_content_card, recommend_special_card=recommend_special_card,page_pv_20_percent=page_pv_20_percent,
         transfer_card=transfer_card,
         video_consultation=video_consultation, partition_day=partition_day, pid=pid, recommend_ctr=recommend_ctr,
         second_jump_ctr=second_jump_ctr, click_ctr=click_ctr