update

new_user_analysis/new_user_word_count.py

@@ -94,19 +94,13 @@ for t in range(0, task_days):
 
     sql_search_ctr = r"""
 
-  select t0.device_id as device_id,click.query from 
-    (
-      (select device_id from online.ml_device_day_active_status where partition_date = '{today_str}' and active_type in (1,2))t2
-    LEFT JOIN
-    (
-    select distinct device_id
-    from ML.ML_D_CT_DV_DEVICECLEAN_DIMEN_D
-    where PARTITION_DAY = '{today_str}'
-    AND is_abnormal_device = 'true'
-    )dev
-    on t2.device_id=dev.device_id
-WHERE dev.device_id is null and t2.device_id is not null) t0 left join
-                     ( SELECT cl_id,
+
+             --搜索pvuv
+             SELECT query
+                  partition_date,
+                  cl_id
+             FROM (
+                      SELECT cl_id,
                              partition_date,
                              action,
                              params['page_name']  as page_name,
@@ -118,7 +112,7 @@ WHERE dev.device_id is null and t2.device_id is not null) t0 left join
                         AND partition_date < '{end_date}'
                         AND ((action = 'do_search' AND params['input_type'] <> 'everyone_watch') or
                              action = 'search_result_click_search')
-
+    
                       UNION all
                       SELECT cl_id,
                              partition_date,
@@ -135,7 +129,7 @@ WHERE dev.device_id is null and t2.device_id is not null) t0 left join
                         and params['tab'] = '精选'
                         and page_name = 'home'
                         AND params['query'] not in ('AI测颜值', 'AI测肤质') --这两个词不跳转搜索结果页
-
+    
                       union all
                       SELECT cl_id,
                              partition_date,
@@ -149,7 +143,7 @@ WHERE dev.device_id is null and t2.device_id is not null) t0 left join
                         AND partition_date < '{end_date}'
                         AND action = 'on_click_card'
                         AND params['page_name'] = 'search_home'
-
+    
                       union all
                       SELECT cl_id,
                              partition_date,
@@ -167,7 +161,7 @@ WHERE dev.device_id is null and t2.device_id is not null) t0 left join
                         AND params['card_type'] = 'search_word'
                         AND params['card_name'] not in ('AI测颜值', 'AI测肤质') --这两个词不跳转搜索结果页
                         --AND page_name='home' android的page_name为空
-
+    
                       union all
                       SELECT cl_id,
                              partition_date,
@@ -183,7 +177,7 @@ WHERE dev.device_id is null and t2.device_id is not null) t0 left join
                         AND params['page_name'] = 'welfare_home'
                         AND params['card_type'] = 'search_word'
                         AND params['in_page_pos'] = '大家都在搜'
-
+    
                       union all
                       SELECT cl_id,
                              partition_date,
@@ -198,8 +192,6 @@ WHERE dev.device_id is null and t2.device_id is not null) t0 left join
                         AND action = 'on_click_card'
                         AND params['card_type'] = 'highlight_word'
                   ) click
-        
-        on  click.cl_id=t0.device_id
 
     """.format(start_date=yesterday_str, end_date=today_str,today_str=yesterday_str)
 
@@ -211,9 +203,9 @@ WHERE dev.device_id is null and t2.device_id is not null) t0 left join
     for name in sql_res:
         # print(name)
         word = name.query
-        doc_id = name.device_id
+        cl_id = name.cl_id
         if "发型" in word:
-            print(word,doc_id)
+            print(word,cl_id)
         # partition_date = str(now + datetime.timedelta(days=-1))
     #     tag_names_list_week.append((word, nums, uv, partition_date))
     #