push codes

pm/daily_reply_content/report/daily_reply_content.sql

@@ -4,14 +4,15 @@ select a.*
 FROM
 (
     --有评论过日记帖的设备,排除疑似广告
-    SELECT diary_id as content_id,'日记贴' as type,user_id,create_time,content
+    SELECT t1.id as content_id,'日记贴' as type,user_id,create_time,content
     FROM
     (
         SELECT id,problem_id,user_id,reply_date as create_time,content
         FROM online.tl_hdfs_topicreply_view
         WHERE partition_date = regexp_replace(DATE_SUB(current_date,1) ,'-','')
         AND is_spam = 'false'  --排除疑似广告
-        and regexp_replace(substr(reply_date,1,10),'-','') >= regexp_replace(DATE_SUB(current_date,1) ,'-','')
+        and is_online='true'
+        and regexp_replace(substr(reply_date,1,10),'-','') = regexp_replace(DATE_SUB(current_date,1) ,'-','')
         group by id,problem_id,user_id,reply_date,content
     )t1
     JOIN
@@ -22,57 +23,60 @@ FROM
         group by id,diary_id
     )t2
     on t2.id=t1.problem_id
-    group by diary_id,user_id,create_time,content
+    group by t1.id,user_id,create_time,content
 
     UNION ALL
 
     --有评论过回答的设备,排除疑似广告
 
-    SELECT answer_id as content_id,'回答' as type,user_id,create_time,content
+    SELECT id as content_id,'回答' as type,user_id,create_time,content
     FROM online.tl_hdfs_answer_reply_view
     WHERE partition_date =regexp_replace(DATE_SUB(current_date,1) ,'-','')
     and (is_fake is NULL or is_fake = 'false')
     AND answer_id is not NULL
-    and regexp_replace(substr(create_time,1,10),'-','') >= regexp_replace(DATE_SUB(current_date,1) ,'-','')
-    group by answer_id,user_id,create_time,content
+    and is_online='true'
+    and is_spam = 'false'  --排除疑似广告
+    and regexp_replace(substr(create_time,1,10),'-','') = regexp_replace(DATE_SUB(current_date,1) ,'-','')
+    group by id,user_id,create_time,content
 
     UNION ALL
 
     --有评论过用户帖的设备
-    SELECT tractate_id as content_id,'帖子' as type,user_id,create_time,content
+    SELECT id as content_id,'帖子' as type,user_id,create_time,content
     FROM  online.tl_hdfs_api_tractate_reply_view
     WHERE partition_date =regexp_replace(DATE_SUB(current_date,1) ,'-','')
-    and regexp_replace(substr(create_time,1,10),'-','') >= regexp_replace(DATE_SUB(current_date,1) ,'-','')
-    group by tractate_id,user_id,create_time,content
+    and regexp_replace(substr(create_time,1,10),'-','') = regexp_replace(DATE_SUB(current_date,1) ,'-','')
+    and is_online='true'
+    group by id,user_id,create_time,content
 
 )a
-JOIN --限制用户是在app进行的回复
-(
-    SELECT a.partition_date,user_id
-    FROM
-    (
-        SELECT concat_ws('-',substr(partition_date,1,4),substr(partition_date,5,2),substr(partition_date,7,2)) as partition_date,user_id,device_id,action
-        FROM online.bl_hdfs_operation_updates
-        WHERE partition_date>=regexp_replace(DATE_SUB(current_date,1) ,'-','')
-        AND partition_date<=regexp_replace(DATE_SUB(current_date,1) ,'-','')
-    )a
-    JOIN
-    (
-        SELECT code
-        FROM  dim.dim_community_action_type
-        WHERE communityuserbehavior_type_name = '回帖'
-    )type
-    ON a.action = code
-    GROUP BY a.partition_date,user_id
-)b
-   ON a.user_id = b.user_id
-   AND substr(a.create_time,1,10)= b.partition_date
+--JOIN --限制用户是在app进行的回复
+--(
+--    SELECT a.partition_date,user_id
+--    FROM
+--    (
+--        SELECT concat_ws('-',substr(partition_date,1,4),substr(partition_date,5,2),substr(partition_date,7,2)) as partition_date,user_id,device_id,action
+--        FROM online.bl_hdfs_operation_updates
+--        WHERE partition_date>=regexp_replace(DATE_SUB(current_date,1) ,'-','')
+--        AND partition_date<=regexp_replace(DATE_SUB(current_date,1) ,'-','')
+--    )a
+--    JOIN
+--    (
+--        SELECT code
+--        FROM  dim.dim_community_action_type
+--        WHERE communityuserbehavior_type_name = '回帖'
+--    )type
+--    ON a.action = code
+--    GROUP BY a.partition_date,user_id
+--)b
+--   ON a.user_id = b.user_id
+--   AND substr(a.create_time,1,10)= b.partition_date
 left join
 (
     select distinct user_id
     from ml.ML_D_CT_UI_USERCLEAN_DIMEN_D
     where PARTITION_DAY =regexp_replace(DATE_SUB(current_date,1) ,'-','')
-    and is_abnormal_user = 'true'
+    and (is_classify_user = 'true' or is_puppet ='true')
 )c
 on a.user_id=c.user_id
 where c.user_id is null