update

task/daily_search_word_count.py

@@ -65,7 +65,7 @@ spark.sql("CREATE TEMPORARY FUNCTION arrayMerge AS 'com.gmei.hive.common.udf.UDF
 # huidu_device_id_df = spark.sql(huidu_device_id_sql)
 # huidu_device_id_df.createOrReplaceTempView("dev_view")
 task_list = []
-task_days = 80
+task_days = 3
 
 for t in range(0, task_days):
     day_num = 0 - t

task/daily_search_word_count_fix.py

+# -*- coding:UTF-8 -*-
+# @Time  : 2020/11/18 14:05
+# @File  : daily_search_word_count_fix.py
+# @email : litao@igengmei.com
+# @author : litao
+
+# -*- coding:UTF-8 -*-
+# @Time  : 2020/11/13 11:08
+# @File  : daily_search_word_count.py
+# @email : litao@igengmei.com
+# @author : litao
+
+# coding=utf-8
+import hashlib
+import time
+from pyspark import SparkConf
+from pyspark.sql import SparkSession, DataFrame
+
+import pymysql
+# from elasticsearch import Elasticsearch
+import datetime
+
+# from maintenance.func_send_email_with_file import send_file_email
+# import zipfile
+
+# es = Elasticsearch([
+#     {
+#         'host': '172.16.31.17',
+#         'port': 9200,
+#     }, {
+#         'host': '172.16.31.11',
+#         'port': 9200,
+#     }])
+
+
+startTime = time.time()
+sparkConf = SparkConf()
+sparkConf.set("spark.sql.crossJoin.enabled", True)
+sparkConf.set("spark.debug.maxToStringFields", "100")
+sparkConf.set("spark.tispark.plan.allow_index_double_read", False)
+sparkConf.set("spark.tispark.plan.allow_index_read", True)
+sparkConf.set("spark.hive.mapred.supports.subdirectories", True)
+sparkConf.set("spark.hadoop.mapreduce.input.fileinputformat.input.dir.recursive", True)
+sparkConf.set("spark.serializer", "org.apache.spark.serializer.KryoSerializer")
+sparkConf.set("mapreduce.output.fileoutputformat.compress", False)
+sparkConf.set("mapreduce.map.output.compress", False)
+sparkConf.set("prod.gold.jdbcuri",
+              "jdbc:mysql://172.16.30.136/doris_prod?user=doris&password=o5gbA27hXHHm&rewriteBatchedStatements=true")
+sparkConf.set("prod.mimas.jdbcuri",
+              "jdbc:mysql://172.16.30.138/mimas_prod?user=mimas&password=GJL3UJe1Ck9ggL6aKnZCq4cRvM&rewriteBatchedStatements=true")
+sparkConf.set("prod.gaia.jdbcuri",
+              "jdbc:mysql://172.16.30.143/zhengxing?user=work&password=BJQaT9VzDcuPBqkd&rewriteBatchedStatements=true")
+sparkConf.set("prod.tidb.jdbcuri",
+              "jdbc:mysql://172.16.40.158:4000/eagle?user=st_user&password=aqpuBLYzEV7tML5RPsN1pntUzFy&rewriteBatchedStatements=true")
+sparkConf.set("prod.jerry.jdbcuri",
+              "jdbc:mysql://172.16.40.158:4000/jerry_prod?user=st_user&password=aqpuBLYzEV7tML5RPsN1pntUzFy&rewriteBatchedStatements=true")
+sparkConf.set("prod.tispark.pd.addresses", "172.16.40.158:2379")
+sparkConf.set("prod.tispark.pd.addresses", "172.16.40.170:4000")
+sparkConf.set("prod.tidb.database", "jerry_prod")
+sparkConf.setAppName("test")
+
+spark = (SparkSession.builder.config(conf=sparkConf).config("spark.sql.extensions", "org.apache.spark.sql.TiExtensions")
+         .config("spark.tispark.pd.addresses", "172.16.40.170:2379").enableHiveSupport().getOrCreate())
+
+spark.sql("ADD JAR hdfs:///user/hive/share/lib/udf/brickhouse-0.7.1-SNAPSHOT.jar")
+spark.sql("ADD JAR hdfs:///user/hive/share/lib/udf/hive-udf-1.0-SNAPSHOT.jar")
+spark.sql("CREATE TEMPORARY FUNCTION json_map AS 'brickhouse.udf.json.JsonMapUDF'")
+spark.sql("CREATE TEMPORARY FUNCTION is_json AS 'com.gmei.hive.common.udf.UDFJsonFormatCheck'")
+spark.sql("CREATE TEMPORARY FUNCTION arrayMerge AS 'com.gmei.hive.common.udf.UDFArryMerge'")
+
+# print(huidu_device_id_sql)
+# huidu_device_id_df = spark.sql(huidu_device_id_sql)
+# huidu_device_id_df.createOrReplaceTempView("dev_view")
+task_list = []
+task_days = 80
+
+for t in range(0, task_days):
+    day_num = 0 - t
+    now = (datetime.datetime.now() + datetime.timedelta(days=day_num))
+    last_30_day_str = (now + datetime.timedelta(days=-30)).strftime("%Y%m%d")
+    today_str = now.strftime("%Y%m%d")
+    yesterday_str = (now + datetime.timedelta(days=-1)).strftime("%Y%m%d")
+    one_week_age_str = (now + datetime.timedelta(days=-7)).strftime("%Y%m%d")
+    sql_search_ctr = r"""
+    SELECT query,
+           partition_date,
+           all_search_uv as all_search_uv, --全部搜索uv
+           all_search_pv as all_search_pv --全部搜索pv
+    FROM (
+             --搜索pvuv
+             SELECT query
+                  , count(click.cl_id)          as all_search_pv
+                  , count(distinct click.cl_id) as all_search_uv,
+                  partition_date
+             FROM (
+                      SELECT cl_id,
+                             partition_date,
+                             action,
+                             params['page_name']  as page_name,
+                             params['input_type'] as input_type,
+                             app_version,
+                             params['query']      as query
+                      FROM online.bl_hdfs_maidian_updates
+                      WHERE partition_date >= '{start_date}'
+                        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,
+                             action,
+                             params['page_name']  as page_name,
+                             params['input_type'] as input_type,
+                             app_version,
+                             params['query']      as query
+                      FROM online.bl_hdfs_maidian_updates
+                      WHERE partition_date >= '{start_date}'
+                        AND partition_date < '{end_date}'
+                        AND action = 'do_search'
+                        and params['input_type'] = 'everyone_watch'
+                        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,
+                             action,
+                             'search_home'   as page_name,
+                             ''              as input_type,
+                             app_version,
+                             params['query'] as query
+                      FROM online.bl_hdfs_maidian_updates
+                      WHERE partition_date >= '{start_date}'
+                        AND partition_date < '{end_date}'
+                        AND action = 'on_click_card'
+                        AND params['page_name'] = 'search_home'
+
+                      union all
+                      SELECT cl_id,
+                             partition_date,
+                             action,
+                             'home'              as page_name,
+                             '首页-猜你喜欢'           as input_type,
+                             app_version,
+                             params['card_name'] as query
+                      FROM online.bl_hdfs_maidian_updates
+                      WHERE partition_date >= '{start_date}'
+                        AND partition_date < '{end_date}'
+                        AND action = 'on_click_card'
+                        AND params['in_page_pos'] = '猜你喜欢'
+                        --AND params['tab_name']='精选'
+                        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,
+                             action,
+                             params['page_name'] as page_name,
+                             '美购首页-大家都在搜'        as input_type,
+                             app_version,
+                             params['card_name'] as query
+                      FROM online.bl_hdfs_maidian_updates
+                      WHERE partition_date >= '{start_date}'
+                        AND partition_date < '{end_date}'
+                        AND action = 'on_click_card'
+                        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,
+                             action,
+                             params['page_name'] as page_name,
+                             '高亮词'               as input_type,
+                             app_version,
+                             params['card_name'] as query
+                      FROM online.bl_hdfs_maidian_updates
+                      WHERE partition_date >= '{start_date}'
+                        AND partition_date < '{end_date}'
+                        AND action = 'on_click_card'
+                        AND params['card_type'] = 'highlight_word'
+                  ) click
+                  LEFT JOIN
+        (
+            select distinct device_id
+            from ml.ml_d_ct_dv_devicespam_d  --去除机构刷单设备，即作弊设备（浏览和曝光事件去除）
+            WHERE partition_day='{start_date}'
+    
+            union all
+    
+            select distinct device_id
+            from dim.dim_device_user_staff   --去除内网用户
+        )spam_pv
+        on spam_pv.device_id=click.cl_id
+        LEFT JOIN
+        (
+            SELECT partition_date,device_id
+            FROM
+                (--找出user_id当天活跃的第一个设备id
+              SELECT user_id,partition_date,
+                      if(size(device_list) > 0, device_list [ 0 ], '') AS device_id
+                FROM online.ml_user_updates
+                WHERE partition_date='{start_date}'
+            )t1
+                    JOIN
+                (  --医生账号
+              SELECT distinct user_id
+              FROM online.tl_hdfs_doctor_view
+              WHERE partition_date = '{start_date}'
+    
+              --马甲账号/模特用户
+              UNION ALL
+              SELECT user_id
+              FROM ml.ml_c_ct_ui_user_dimen_d
+              WHERE partition_day = '{start_date}'
+              AND (is_puppet = 'true' or is_classifyuser = 'true')
+    
+              UNION ALL
+              --公司内网覆盖用户
+              select distinct user_id
+              from dim.dim_device_user_staff
+    
+              UNION ALL
+              --登陆过医生设备
+              SELECT distinct t1.user_id
+                FROM
+                (
+                    SELECT user_id, v.device_id as device_id
+                    FROM online.ml_user_history_detail
+                        LATERAL VIEW EXPLODE(device_history_list) v AS device_id
+                    WHERE partition_date = '{start_date}'
+                ) t1
+                JOIN
+                (
+                    SELECT device_id
+                    FROM online.ml_device_history_detail
+                    WHERE partition_date = '{start_date}'
+                    AND is_login_doctor = '1'
+                ) t2
+                    ON t1.device_id = t2.device_id
+            )t2
+                on t1.user_id=t2.user_id
+            group by partition_date,device_id
+        )dev
+        on click.partition_date=dev.partition_date and click.cl_id=dev.device_id
+    WHERE spam_pv.device_id IS NULL
+      and dev.device_id is null
+                  
+                  
+             GROUP BY query,click.partition_date
+         ) t3 order by all_search_uv desc
+
+    """.format(start_date=yesterday_str, end_date=today_str)
+
+    print(sql_search_ctr)
+    search_ctr_df = spark.sql(sql_search_ctr)
+    search_ctr_df.show(1)
+    sql_res = search_ctr_df.collect()
+    tag_names_list_week = []
+    for name in sql_res:
+        # print(name)
+        word = name.query
+        nums = name.all_search_pv
+        uv = name.all_search_uv
+        partition_date = str(now + datetime.timedelta(days=-1))
+        tag_names_list_week.append((word, nums, uv, partition_date))
+
+    db = pymysql.connect(host='172.16.40.158', port=4000, user='st_user', passwd='aqpuBLYzEV7tML5RPsN1pntUzFy',
+                         db='jerry_prod')
+    cursor = db.cursor()
+    insert_sql = "replace into daily_search_word_count_fix(word, nums, uv,pid,partition_day) VALUES(%s,%s,%s,%s,%s)"
+    insert_list = []
+    for count, item in enumerate(tag_names_list_week):
+        word, nums, uv, partition_date = item
+        try:
+            if len(word) >= 200:
+                continue
+            pid = hashlib.md5((partition_date + word).encode("utf8")).hexdigest()
+            insert_sql_tuple = (word, nums, uv, pid, partition_date)
+            insert_list.append(insert_sql_tuple)
+            # print(insert_sql_tuple)
+        except:
+            continue
+        if count % 100 == 0:
+            cursor.execute("set names 'UTF8'")
+            res = cursor.executemany(insert_sql, insert_list)
+            db.commit()
+            # print(res)
+            insert_list = []
+            # print(count)
+    res = cursor.executemany(insert_sql, insert_list)
+    db.commit()
+    db.close()
+    print(res)
+
+if __name__ == "__main__":
+    tag_names_list = []
+    tag_names_list_week = []
+    all_data_day = []
+    all_data_week = []

task/recommend_strategy_fix.py

@@ -79,8 +79,7 @@ spark.sql("CREATE TEMPORARY FUNCTION is_json AS 'com.gmei.hive.common.udf.UDFJso
 spark.sql("CREATE TEMPORARY FUNCTION arrayMerge AS 'com.gmei.hive.common.udf.UDFArryMerge'")
 
 
-task_days = 2
-for t in range(0, 60):
+for t in range(6, 7):
     day_num = 0 - t
     now = (datetime.datetime.now() + datetime.timedelta(days=day_num))
     last_30_day_str = (now + datetime.timedelta(days=-30)).strftime("%Y%m%d")
@@ -839,10 +838,11 @@ for t in range(0, 60):
         pid = hashlib.md5((day_id + device_os_type + active_type + card_content_type + recommend_type).encode("utf8")).hexdigest()
         try:
             card_click_diff = data_dict.get(pid) - card_click
-            print(card_click_diff)
-            card_click += card_click_diff
-            card_exposure += card_click_diff
-            click_ctr = round(card_click/card_exposure,5)
+            if card_click_diff:
+                print(card_click_diff)
+                card_click += card_click_diff
+                card_exposure += card_click_diff
+                click_ctr = round(card_click/card_exposure,5)
         except:
             print("add error")
         instert_sql = """replace into recommend_strategy_d_fix(

task/spark_test.py

@@ -4,12 +4,6 @@
 # @email : litao@igengmei.com
 # @author : litao
 
-# -*- coding:UTF-8 -*-
-# @Time  : 2020/9/4 17:07
-# @File  : search_meigou_ctr.py
-# @email : litao@igengmei.com
-# @author : litao
-
 import hashlib
 import json
 

task/每日搜索词.sql

 --搜索日报-英赫
-SELECT
-    partition_date as `日期`
-    ,device_os_type as `平台`,active_type as `活跃类型`,channel as `渠道类型`
-    ,dau as `DAU`
-    ,all_search_uv as `完成搜索uv`
-    ,all_search_pv as `完成搜索pv`
-    ,if(dau <> 0 ,concat(cast((all_search_uv/dau)*100 as decimal(18,2)),'%') , '-') as `完成搜索的用户比例`
-    ,if(all_search_uv <> 0 ,concat(cast((all_search_pv/all_search_uv) as decimal(18,2)),'') , '-') as `搜索用户人均使用次数`
 
-FROM
-(
-    SELECT
-        coalesce(t1.partition_date,t2.partition_date,t3.partition_date,t4.partition_date,t5.partition_date,t6.partition_date,t7.partition_date,t8.partition_date,t9.partition_date) as partition_date
-        ,coalesce(t1.active_type,t2.active_type,t3.active_type,t4.active_type,t5.active_type,t6.active_type,t7.active_type,t8.active_type,t9.active_type) as active_type
-        ,coalesce(t1.device_os_type,t2.device_os_type,t3.device_os_type,t4.device_os_type,t5.device_os_type,t6.device_os_type,t7.device_os_type,t8.device_os_type,t9.device_os_type) as device_os_type
-        ,coalesce(t1.channel,t2.channel,t3.channel,t4.channel,t5.channel,t6.channel,t7.channel,t8.channel,t9.channel) as channel
-        ,coalesce(t1.dau,0) as dau
-        ,coalesce(t3.all_search_uv,0) as all_search_uv                                           --全部搜索uv
-        ,coalesce(t3.all_search_pv,0) as all_search_pv                                           --全部搜索pv
+SELECT query
+     , coalesce(t3.all_search_uv, 0) as all_search_uv --全部搜索uv
+     , coalesce(t3.all_search_pv, 0) as all_search_pv --全部搜索pv
+     , t3.partition_date
+FROM (
+         --搜索pvuv
+         SELECT query
+              , click.partition_date        as partition_date
+              , count(click.cl_id)          as all_search_pv
+              , count(distinct click.cl_id) as all_search_uv
+         FROM (
+                  SELECT cl_id,
+                         partition_date,
+                         action,
+                         params['page_name']  as page_name,
+                         params['input_type'] as input_type,
+                         app_version,
+                         params['query']      as query
+                  FROM online.bl_hdfs_maidian_updates
+                  WHERE partition_date >= '${start_date}'
+                    AND partition_date < '${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,
+                         action,
+                         params['page_name']  as page_name,
+                         params['input_type'] as input_type,
+                         app_version,
+                         params['query']      as query
+                  FROM online.bl_hdfs_maidian_updates
+                  WHERE partition_date >= '${start_date}'
+                    AND partition_date < '${end_date}'
+                    AND action = 'do_search'
+                    and params['input_type'] = 'everyone_watch'
+                    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,
+                         action,
+                         'search_home'   as page_name,
+                         ''              as input_type,
+                         app_version,
+                         params['query'] as query
+                  FROM online.bl_hdfs_maidian_updates
+                  WHERE partition_date >= '${start_date}'
+                    AND partition_date < '${end_date}'
+                    AND action = 'on_click_card'
+                    AND params['page_name'] = 'search_home'
+
+                  union all
+                  SELECT cl_id,
+                         partition_date,
+                         action,
+                         'home'              as page_name,
+                         '首页-猜你喜欢'           as input_type,
+                         app_version,
+                         params['card_name'] as query
+                  FROM online.bl_hdfs_maidian_updates
+                  WHERE partition_date >= '${start_date}'
+                    AND partition_date < '${end_date}'
+                    AND action = 'on_click_card'
+                    AND params['in_page_pos'] = '猜你喜欢'
+                    --AND params['tab_name']='精选'
+                    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,
+                         action,
+                         params['page_name'] as page_name,
+                         '美购首页-大家都在搜'        as input_type,
+                         app_version,
+                         params['card_name'] as query
+                  FROM online.bl_hdfs_maidian_updates
+                  WHERE partition_date >= '${start_date}'
+                    AND partition_date < '${end_date}'
+                    AND action = 'on_click_card'
+                    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,
+                         action,
+                         params['page_name'] as page_name,
+                         '高亮词'               as input_type,
+                         app_version,
+                         params['card_name'] as query
+                  FROM online.bl_hdfs_maidian_updates
+                  WHERE partition_date >= '${start_date}'
+                    AND partition_date < '${end_date}'
+                    AND action = 'on_click_card'
+                    AND params['card_type'] = 'highlight_word'
+              ) click
+         GROUP BY click.partition_date, query
+     ) t3
 
 
-
-    FROM
-    (  --dau
-        SELECT mas.partition_date,count(DISTINCT mas.device_id) as dau
-        FROM
-        (
-            SELECT
-            partition_date,m.device_id
-            FROM online.ml_device_day_active_status m
-                LEFT JOIN
-                (SELECT code,is_ai_channel,partition_day
-                 FROM DIM.DIM_AI_CHANNEL_ZP_NEW
-                 WHERE partition_day>= '${start_date}' AND partition_day < '${end_date}'  ) tmp
-                ON  m.partition_date=tmp.partition_day AND first_channel_source_type=code
-            where partition_date >= '${start_date}'
-            AND partition_date < '${end_date}'
-            AND active_type in ('1','2','4')
-        ) mas
-        GROUP BY mas.partition_date
-    )t1
-    full JOIN
-    (
-        --搜索发起，搜索框点击,724版本埋点修复
-        SELECT
-            click.partition_date as partition_date
-            ,count(click.cl_id) as all_search_click_pv_724
-            ,count(distinct click.cl_id) as all_search_click_uv_724
-        FROM
-        (
-            SELECT cl_id,partition_date,action,params['page_name'] as page_name,'' as input_type,app_version
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'on_click_navbar_search'
-              AND (int(split(app_version,'\\.')[0]) = 7 AND int(split(app_version,'\\.')[1]) >= 24)
-
-            UNION all
-            SELECT cl_id,partition_date,action,params['page_name'] as page_name,params['input_type'] as input_type,app_version
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'do_search'
-              AND params['input_type'] = '详情页默认词'
-
-            UNION all
-            SELECT cl_id,partition_date,action,page_name,'旧首页-大家都在看' as input_type,app_version
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-            AND partition_date < '${end_date}'
-            AND action = 'do_search'
-            and params['input_type']='everyone_watch'
-            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,action,'home' as page_name,'新首页-猜你喜欢' as input_type,app_version
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'on_click_card'
-              AND params['in_page_pos']='猜你喜欢'
-              --AND params['tab_name']='精选'
-              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,action,params['page_name'] as page_name,'美购首页-大家都在搜' as input_type,app_version
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'on_click_card'
-              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,action,params['page_name'] as page_name,'高亮词' as input_type,app_version
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'on_click_card'
-              AND params['card_type'] ='highlight_word'
-        )click
-
-        GROUP BY click.partition_date,active_type,device_os_type,channel
-    )t2
-      on t2.partition_date = t1.partition_date
-        AND t2.active_type = t1.active_type
-        AND t2.device_os_type = t1.device_os_type
-        AND t2.channel = t1.channel
-    full JOIN
-    (
-        --搜索pvuv
-        SELECT
-            click.partition_date as partition_date
-            ,active_type
-            ,device_os_type
-            ,channel
-            ,count(click.cl_id) as all_search_pv
-            ,count(distinct click.cl_id) as all_search_uv
-            ,count(case when int(split(app_version,'\\.')[0]) = 7 AND int(split(app_version,'\\.')[1]) >= 26 AND input_type in ('联想','聚合模块') then click.cl_id end) as sug_search_pv_726
-            ,count(distinct case when int(split(app_version,'\\.')[0]) = 7 AND int(split(app_version,'\\.')[1]) >= 26 AND input_type in ('联想','聚合模块') then click.cl_id end) as sug_search_uv_726
-        FROM
-        (
-            SELECT cl_id,partition_date,action,params['page_name'] as page_name,params['input_type'] as input_type,app_version,params['query'] as query
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${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,action,params['page_name'] as page_name,params['input_type'] as input_type,app_version,params['query'] as query
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'do_search'
-              and params['input_type']='everyone_watch'
-              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,action,'search_home' as page_name,'' as input_type,app_version,params['query'] as query
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'on_click_card'
-              AND params['page_name']='search_home'
-
-            union all
-            SELECT cl_id,partition_date,action,'home' as page_name,'首页-猜你喜欢' as input_type,app_version,params['card_name'] as query
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'on_click_card'
-              AND params['in_page_pos']='猜你喜欢'
-              --AND params['tab_name']='精选'
-              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,action,params['page_name'] as page_name,'美购首页-大家都在搜' as input_type,app_version,params['card_name'] as query
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'on_click_card'
-              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,action,params['page_name'] as page_name,'高亮词' as input_type,app_version,params['card_name'] as query
-            FROM online.bl_hdfs_maidian_updates
-            WHERE partition_date >= '${start_date}'
-              AND partition_date < '${end_date}'
-              AND action = 'on_click_card'
-              AND params['card_type'] ='highlight_word'
-        )click
-
-        GROUP BY click.partition_date,active_type,device_os_type,channel
-    )t3
-      on t3.partition_date = t1.partition_date
-        AND t3.active_type = t1.active_type
-        AND t3.device_os_type = t1.device_os_type
-        AND t3.channel = t1.channel
-
-)t
-ORDER BY `日期`,`平台`,`活跃类型`,`渠道类型`
-