Skip to content

M
meta_base_code
Commit 883cd047 authored by litaolemo's avatar litaolemo
Browse files
Options

update

parent fe0390a9
Hide whitespace changes
Inline Side-by-side
Showing with 338 additions and 338 deletions
task/search_strategy_d.py
View file @ 883cd047
...@@ -380,344 +380,344 @@ for t in range(1, task_days): ...@@ -380,344 +380,344 @@ for t in range(1, task_days):
for res in sql_res: for res in sql_res:
print(res) print(res)
device_df.createOrReplaceTempView("data_table") device_df.createOrReplaceTempView("data_table")
#
sql = r""" # sql = r"""
SELECT t3.partition_date as partition_date # SELECT t3.partition_date as partition_date
,t3.device_os_type as device_os_type # ,t3.device_os_type as device_os_type
,t3.active_type as active_type # ,t3.active_type as active_type
,t3.channel as channel # ,t3.channel as channel
,NVL(t3.search_pv,0) as pv # ,NVL(t3.search_pv,0) as pv
,NVL(t3.search_uv,0) as uv # ,NVL(t3.search_uv,0) as uv
,if(NVL(t3.search_uv,0) <> 0 ,concat(cast((NVL(t4.hexin_card_click_pv,0)/NVL(t3.search_uv,0)) as decimal(18,2)),'') , '-') as search_core_pv # ,if(NVL(t3.search_uv,0) <> 0 ,concat(cast((NVL(t4.hexin_card_click_pv,0)/NVL(t3.search_uv,0)) as decimal(18,2)),'') , '-') as search_core_pv
,if(NVL(t3.search_uv,0) <> 0 ,concat(cast((NVL(t4.neirong_card_click_pv,0)/NVL(t3.search_uv,0)) as decimal(18,2)),'') , '-') as search_pv # ,if(NVL(t3.search_uv,0) <> 0 ,concat(cast((NVL(t4.neirong_card_click_pv,0)/NVL(t3.search_uv,0)) as decimal(18,2)),'') , '-') as search_pv
FROM # FROM
(--昨天总搜索量 # (--昨天总搜索量
SELECT partition_date,active_type,device_os_type,channel,search_pv,search_uv # SELECT partition_date,active_type,device_os_type,channel,search_pv,search_uv
FROM # FROM
( # (
SELECT t1.partition_date,active_type,device_os_type,channel # SELECT t1.partition_date,active_type,device_os_type,channel
,count(t1.cl_id) as search_pv # ,count(t1.cl_id) as search_pv
,count(distinct t1.cl_id) as search_uv # ,count(distinct t1.cl_id) as search_uv
FROM # FROM
( # (
SELECT partition_date # SELECT partition_date
,params['query'] as query # ,params['query'] as query
,cl_id # ,cl_id
FROM online.bl_hdfs_maidian_updates # FROM online.bl_hdfs_maidian_updates
WHERE partition_date >= {yesterday_str} # WHERE partition_date >= {yesterday_str}
AND partition_date < {today_str} # AND partition_date < {today_str}
AND action in ('do_search','search_result_click_search') # AND action in ('do_search','search_result_click_search')
#
UNION ALL # UNION ALL
SELECT partition_date,params['query'] as query,cl_id # SELECT partition_date,params['query'] as query,cl_id
FROM online.bl_hdfs_maidian_updates # FROM online.bl_hdfs_maidian_updates
WHERE partition_date >= {yesterday_str} # WHERE partition_date >= {yesterday_str}
AND partition_date < {today_str} # AND partition_date < {today_str}
AND action = 'on_click_card' # AND action = 'on_click_card'
AND params['page_name']='search_home' # AND params['page_name']='search_home'
#
UNION ALL # UNION ALL
SELECT partition_date # SELECT partition_date
,params['card_name'] as query # ,params['card_name'] as query
,cl_id # ,cl_id
FROM online.bl_hdfs_maidian_updates # FROM online.bl_hdfs_maidian_updates
WHERE partition_date >= {yesterday_str} # WHERE partition_date >= {yesterday_str}
AND partition_date < {today_str} # AND partition_date < {today_str}
AND action = 'on_click_card' # AND action = 'on_click_card'
AND params['in_page_pos']='猜你喜欢' # AND params['in_page_pos']='猜你喜欢'
AND params['tab_name']='精选' # AND params['tab_name']='精选'
AND params['card_type']='search_word' # AND params['card_type']='search_word'
--AND page_name='home' android的page_name为空 # --AND page_name='home' android的page_name为空
#
UNION ALL # UNION ALL
SELECT partition_date # SELECT partition_date
,params['card_name'] as query # ,params['card_name'] as query
,cl_id # ,cl_id
FROM online.bl_hdfs_maidian_updates # FROM online.bl_hdfs_maidian_updates
WHERE partition_date >= {yesterday_str} # WHERE partition_date >= {yesterday_str}
AND partition_date < {today_str} # AND partition_date < {today_str}
AND action = 'on_click_card' # AND action = 'on_click_card'
AND page_name='welfare_home' # AND page_name='welfare_home'
AND params['card_type'] ='search_word' # AND params['card_type'] ='search_word'
AND params['in_page_pos']='大家都在搜' # AND params['in_page_pos']='大家都在搜'
#
UNION ALL # UNION ALL
SELECT partition_date # SELECT partition_date
,params['card_name'] as query # ,params['card_name'] as query
,cl_id # ,cl_id
FROM online.bl_hdfs_maidian_updates # FROM online.bl_hdfs_maidian_updates
WHERE partition_date >= {yesterday_str} # WHERE partition_date >= {yesterday_str}
AND partition_date < {today_str} # AND partition_date < {today_str}
AND int(split(app_version,'\\.')[1]) >= 27 # AND int(split(app_version,'\\.')[1]) >= 27
AND action='on_click_card' # AND action='on_click_card'
AND params['card_type']='highlight_word' # AND params['card_type']='highlight_word'
)t1 # )t1
JOIN # JOIN
( # (
SELECT partition_date,device_id,t2.active_type,t2.channel,t2.device_os_type # SELECT partition_date,device_id,t2.active_type,t2.channel,t2.device_os_type
FROM # FROM
( # (
SELECT # SELECT
partition_date,m.device_id # partition_date,m.device_id
,array(device_os_type ,'合计') as device_os_type # ,array(device_os_type ,'合计') as device_os_type
,array(case WHEN active_type = '4' THEN '老活' # ,array(case WHEN active_type = '4' THEN '老活'
WHEN active_type in ('1','2') then '新增' END ,'合计') as active_type # WHEN active_type in ('1','2') then '新增' END ,'合计') as active_type
,array(CASE WHEN is_ai_channel = 'true' THEN 'AI' ELSE '其他' END , '合计') as channel # ,array(CASE WHEN is_ai_channel = 'true' THEN 'AI' ELSE '其他' END , '合计') as channel
FROM online.ml_device_day_active_status m # FROM online.ml_device_day_active_status m
LEFT JOIN # LEFT JOIN
(SELECT code,is_ai_channel,partition_day # (SELECT code,is_ai_channel,partition_day
FROM DIM.DIM_AI_CHANNEL_ZP_NEW # FROM DIM.DIM_AI_CHANNEL_ZP_NEW
WHERE partition_day>= {yesterday_str} # WHERE partition_day>= {yesterday_str}
AND partition_day < {today_str}) tmp # AND partition_day < {today_str}) tmp
ON m.partition_date=tmp.partition_day AND first_channel_source_type=code # ON m.partition_date=tmp.partition_day AND first_channel_source_type=code
WHERE partition_date >= {yesterday_str} # WHERE partition_date >= {yesterday_str}
AND partition_date < {today_str} # AND partition_date < {today_str}
AND active_type in ('1','2','4') # AND active_type in ('1','2','4')
AND first_channel_source_type not in ('yqxiu1','yqxiu2','yqxiu3','yqxiu4','yqxiu5','mxyc1','mxyc2','mxyc3' # AND first_channel_source_type not in ('yqxiu1','yqxiu2','yqxiu3','yqxiu4','yqxiu5','mxyc1','mxyc2','mxyc3'
,'wanpu','jinshan','jx','maimai','zhuoyi','huatian','suopingjingling','mocha','mizhe','meika','lamabang' # ,'wanpu','jinshan','jx','maimai','zhuoyi','huatian','suopingjingling','mocha','mizhe','meika','lamabang'
,'js-az1','js-az2','js-az3','js-az4','js-az5','jfq-az1','jfq-az2','jfq-az3','jfq-az4','jfq-az5','toufang1' # ,'js-az1','js-az2','js-az3','js-az4','js-az5','jfq-az1','jfq-az2','jfq-az3','jfq-az4','jfq-az5','toufang1'
,'toufang2','toufang3','toufang4','toufang5','toufang6','TF-toufang1','TF-toufang2','TF-toufang3','TF-toufang4' # ,'toufang2','toufang3','toufang4','toufang5','toufang6','TF-toufang1','TF-toufang2','TF-toufang3','TF-toufang4'
,'TF-toufang5','tf-toufang1','tf-toufang2','tf-toufang3','tf-toufang4','tf-toufang5','benzhan','promotion_aso100' # ,'TF-toufang5','tf-toufang1','tf-toufang2','tf-toufang3','tf-toufang4','tf-toufang5','benzhan','promotion_aso100'
,'promotion_qianka','promotion_xiaoyu','promotion_dianru','promotion_malioaso','promotion_malioaso-shequ' # ,'promotion_qianka','promotion_xiaoyu','promotion_dianru','promotion_malioaso','promotion_malioaso-shequ'
,'promotion_shike','promotion_julang_jl03','promotion_zuimei') # ,'promotion_shike','promotion_julang_jl03','promotion_zuimei')
AND first_channel_source_type not like 'promotion\_jf\_%' # AND first_channel_source_type not like 'promotion\_jf\_%'
) mas # ) mas
LATERAL VIEW explode(mas.channel) t2 AS channel # LATERAL VIEW explode(mas.channel) t2 AS channel
LATERAL VIEW explode(mas.device_os_type) t2 AS device_os_type # LATERAL VIEW explode(mas.device_os_type) t2 AS device_os_type
LATERAL VIEW explode(mas.active_type) t2 AS active_type # LATERAL VIEW explode(mas.active_type) t2 AS active_type
)t2 # )t2
on t1.cl_id=t2.device_id AND t1.partition_date = t2.partition_date # on t1.cl_id=t2.device_id AND t1.partition_date = t2.partition_date
#
LEFT JOIN # LEFT JOIN
( # (
SELECT DISTINCT device_id # 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） # 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={yesterday_str} # WHERE partition_day={yesterday_str}
#
UNION ALL # UNION ALL
SELECT DISTINCT device_id # SELECT DISTINCT device_id
FROM dim.dim_device_user_staff --去除内网用户 # FROM dim.dim_device_user_staff --去除内网用户
)spam_pv # )spam_pv
on spam_pv.device_id=t1.cl_id # on spam_pv.device_id=t1.cl_id
LEFT JOIN # LEFT JOIN
( # (
SELECT partition_date,device_id # SELECT partition_date,device_id
FROM # FROM
(--找出user_id当天活跃的第一个设备id # (--找出user_id当天活跃的第一个设备id
SELECT user_id,partition_date, # SELECT user_id,partition_date,
if(size(device_list) > 0, device_list [ 0 ], '') AS device_id # if(size(device_list) > 0, device_list [ 0 ], '') AS device_id
FROM online.ml_user_updates # FROM online.ml_user_updates
WHERE partition_date>='${{yesterday_str}}' AND partition_date<'${{today_str}}' # WHERE partition_date>='${{yesterday_str}}' AND partition_date<'${{today_str}}'
)t1 # )t1
JOIN # JOIN
( --医生账号 # ( --医生账号
SELECT distinct user_id # SELECT distinct user_id
FROM online.tl_hdfs_doctor_view # FROM online.tl_hdfs_doctor_view
WHERE partition_date = {yesterday_str} # WHERE partition_date = {yesterday_str}
#
--马甲账号/模特用户 # --马甲账号/模特用户
UNION ALL # UNION ALL
SELECT user_id # SELECT user_id
FROM ml.ml_c_ct_ui_user_dimen_d # FROM ml.ml_c_ct_ui_user_dimen_d
WHERE partition_day = {yesterday_str} # WHERE partition_day = {yesterday_str}
AND (is_puppet = 'true' or is_classifyuser = 'true') # AND (is_puppet = 'true' or is_classifyuser = 'true')
#
UNION ALL # UNION ALL
--公司内网覆盖用户 # --公司内网覆盖用户
select distinct user_id # select distinct user_id
from dim.dim_device_user_staff # from dim.dim_device_user_staff
#
UNION ALL # UNION ALL
--登陆过医生设备 # --登陆过医生设备
SELECT distinct t1.user_id # SELECT distinct t1.user_id
FROM # FROM
( # (
SELECT user_id, v.device_id as device_id # SELECT user_id, v.device_id as device_id
FROM online.ml_user_history_detail # FROM online.ml_user_history_detail
LATERAL VIEW EXPLODE(device_history_list) v AS device_id # LATERAL VIEW EXPLODE(device_history_list) v AS device_id
WHERE partition_date = {yesterday_str} # WHERE partition_date = {yesterday_str}
)t1 # )t1
JOIN # JOIN
( # (
SELECT device_id # SELECT device_id
FROM online.ml_device_history_detail # FROM online.ml_device_history_detail
WHERE partition_date = {yesterday_str} # WHERE partition_date = {yesterday_str}
AND is_login_doctor = '1' # AND is_login_doctor = '1'
)t2 # )t2
ON t1.device_id = t2.device_id # ON t1.device_id = t2.device_id
)t2 # )t2
on t1.user_id=t2.user_id # on t1.user_id=t2.user_id
group by partition_date,device_id # group by partition_date,device_id
)dev # )dev
on t1.partition_date=dev.partition_date and t1.cl_id=dev.device_id # on t1.partition_date=dev.partition_date and t1.cl_id=dev.device_id
WHERE (spam_pv.device_id IS NULL or spam_pv.device_id ='') # WHERE (spam_pv.device_id IS NULL or spam_pv.device_id ='')
and (dev.device_id is null or dev.device_id ='') # and (dev.device_id is null or dev.device_id ='')
GROUP BY t1.partition_date,t2.active_type,device_os_type,channel # GROUP BY t1.partition_date,t2.active_type,device_os_type,channel
)t # )t
)t3 # )t3
#
LEFT JOIN # LEFT JOIN
(--昨天搜索结果页卡片点击pv # (--昨天搜索结果页卡片点击pv
SELECT t1.partition_date,active_type,device_os_type,channel # SELECT t1.partition_date,active_type,device_os_type,channel
,sum(hexin) as hexin_card_click_pv # ,sum(hexin) as hexin_card_click_pv
,sum(neirong) as neirong_card_click_pv # ,sum(neirong) as neirong_card_click_pv
FROM # FROM
( # (
SELECT NVL(t2.partition_date,t3.partition_date) as partition_date # SELECT NVL(t2.partition_date,t3.partition_date) as partition_date
,NVL(t2.cl_id,t3.cl_id) as cl_id # ,NVL(t2.cl_id,t3.cl_id) as cl_id
,NVL(t2.query,t3.query) as query # ,NVL(t2.query,t3.query) as query
,NVL(t2.pv,0) as hexin # ,NVL(t2.pv,0) as hexin
,NVL(t3.pv,0) as neirong # ,NVL(t3.pv,0) as neirong
FROM # FROM
(--核心卡片点击 # (--核心卡片点击
SELECT partition_date # SELECT partition_date
,params['query'] as query # ,params['query'] as query
,cl_id # ,cl_id
,count(1) as pv # ,count(1) as pv
FROM online.bl_hdfs_maidian_updates # FROM online.bl_hdfs_maidian_updates
WHERE partition_date >= {yesterday_str} # WHERE partition_date >= {yesterday_str}
AND partition_date < {today_str} # AND partition_date < {today_str}
AND ((action in ('search_result_click_recommend_item','search_result_welfare_click_item','search_result_hospital_click_item','search_result_doctor_click_item','on_click_doctor_card', 'on_click_hospital_card') # AND ((action in ('search_result_click_recommend_item','search_result_welfare_click_item','search_result_hospital_click_item','search_result_doctor_click_item','on_click_doctor_card', 'on_click_hospital_card')
AND page_name in ('search_result_more','search_result_welfare','search_result_hospital','search_result_doctor')) # AND page_name in ('search_result_more','search_result_welfare','search_result_hospital','search_result_doctor'))
or (action = 'goto_welfare_detail' AND params [ 'from' ] = 'search_result_welfare_recommend') # or (action = 'goto_welfare_detail' AND params [ 'from' ] = 'search_result_welfare_recommend')
or (action = 'on_click_card' AND params['card_content_type'] in ('service','hospital','doctor') AND page_name in ('search_result_more','search_result_welfare','search_result_hospital','search_result_doctor')) # or (action = 'on_click_card' AND params['card_content_type'] in ('service','hospital','doctor') AND page_name in ('search_result_more','search_result_welfare','search_result_hospital','search_result_doctor'))
or (action = 'on_click_button' AND params['button_name'] = 'check_plan' AND page_name = 'search_result_more')) # or (action = 'on_click_button' AND params['button_name'] = 'check_plan' AND page_name = 'search_result_more'))
GROUP BY partition_date # GROUP BY partition_date
,params['query'] # ,params['query']
,cl_id # ,cl_id
)t2 # )t2
FULL JOIN # FULL JOIN
(--内容卡片点击 # (--内容卡片点击
SELECT partition_date # SELECT partition_date
,params['query'] as query # ,params['query'] as query
,cl_id # ,cl_id
,count(1) as pv # ,count(1) as pv
FROM online.bl_hdfs_maidian_updates # FROM online.bl_hdfs_maidian_updates
WHERE partition_date >= {yesterday_str} # WHERE partition_date >= {yesterday_str}
AND partition_date < {today_str} # AND partition_date < {today_str}
AND ((action in ('on_click_topic_card','on_click_diary_card','search_result_click_infomation_item') # 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')) # 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'))) # 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')))
GROUP BY partition_date # GROUP BY partition_date
,params['query'] # ,params['query']
,cl_id # ,cl_id
)t3 # )t3
on t3.partition_date=t2.partition_date # on t3.partition_date=t2.partition_date
AND t3.query=t2.query # AND t3.query=t2.query
AND t3.cl_id=t2.cl_id # AND t3.cl_id=t2.cl_id
)t1 # )t1
JOIN # JOIN
( # (
SELECT partition_date,device_id,t2.active_type,t2.channel,t2.device_os_type # SELECT partition_date,device_id,t2.active_type,t2.channel,t2.device_os_type
FROM # FROM
( # (
SELECT # SELECT
partition_date,m.device_id # partition_date,m.device_id
,array(device_os_type ,'合计') as device_os_type # ,array(device_os_type ,'合计') as device_os_type
,array(case WHEN active_type = '4' THEN '老活' # ,array(case WHEN active_type = '4' THEN '老活'
WHEN active_type in ('1','2') then '新增' END ,'合计') as active_type # WHEN active_type in ('1','2') then '新增' END ,'合计') as active_type
,array(CASE WHEN is_ai_channel = 'true' THEN 'AI' ELSE '其他' END , '合计') as channel # ,array(CASE WHEN is_ai_channel = 'true' THEN 'AI' ELSE '其他' END , '合计') as channel
FROM online.ml_device_day_active_status m # FROM online.ml_device_day_active_status m
LEFT JOIN # LEFT JOIN
(SELECT code,is_ai_channel,partition_day # (SELECT code,is_ai_channel,partition_day
FROM DIM.DIM_AI_CHANNEL_ZP_NEW # FROM DIM.DIM_AI_CHANNEL_ZP_NEW
WHERE partition_day>= {yesterday_str} # WHERE partition_day>= {yesterday_str}
AND partition_day < {today_str}) tmp # AND partition_day < {today_str}) tmp
ON m.partition_date=tmp.partition_day AND first_channel_source_type=code # ON m.partition_date=tmp.partition_day AND first_channel_source_type=code
WHERE partition_date >= {yesterday_str} # WHERE partition_date >= {yesterday_str}
AND partition_date < {today_str} # AND partition_date < {today_str}
AND active_type in ('1','2','4') # AND active_type in ('1','2','4')
AND first_channel_source_type not in ('yqxiu1','yqxiu2','yqxiu3','yqxiu4','yqxiu5','mxyc1','mxyc2','mxyc3' # AND first_channel_source_type not in ('yqxiu1','yqxiu2','yqxiu3','yqxiu4','yqxiu5','mxyc1','mxyc2','mxyc3'
,'wanpu','jinshan','jx','maimai','zhuoyi','huatian','suopingjingling','mocha','mizhe','meika','lamabang' # ,'wanpu','jinshan','jx','maimai','zhuoyi','huatian','suopingjingling','mocha','mizhe','meika','lamabang'
,'js-az1','js-az2','js-az3','js-az4','js-az5','jfq-az1','jfq-az2','jfq-az3','jfq-az4','jfq-az5','toufang1' # ,'js-az1','js-az2','js-az3','js-az4','js-az5','jfq-az1','jfq-az2','jfq-az3','jfq-az4','jfq-az5','toufang1'
,'toufang2','toufang3','toufang4','toufang5','toufang6','TF-toufang1','TF-toufang2','TF-toufang3','TF-toufang4' # ,'toufang2','toufang3','toufang4','toufang5','toufang6','TF-toufang1','TF-toufang2','TF-toufang3','TF-toufang4'
,'TF-toufang5','tf-toufang1','tf-toufang2','tf-toufang3','tf-toufang4','tf-toufang5','benzhan','promotion_aso100' # ,'TF-toufang5','tf-toufang1','tf-toufang2','tf-toufang3','tf-toufang4','tf-toufang5','benzhan','promotion_aso100'
,'promotion_qianka','promotion_xiaoyu','promotion_dianru','promotion_malioaso','promotion_malioaso-shequ' # ,'promotion_qianka','promotion_xiaoyu','promotion_dianru','promotion_malioaso','promotion_malioaso-shequ'
,'promotion_shike','promotion_julang_jl03','promotion_zuimei') # ,'promotion_shike','promotion_julang_jl03','promotion_zuimei')
AND first_channel_source_type not like 'promotion\_jf\_%' # AND first_channel_source_type not like 'promotion\_jf\_%'
) mas # ) mas
LATERAL VIEW explode(mas.channel) t2 AS channel # LATERAL VIEW explode(mas.channel) t2 AS channel
LATERAL VIEW explode(mas.device_os_type) t2 AS device_os_type # LATERAL VIEW explode(mas.device_os_type) t2 AS device_os_type
LATERAL VIEW explode(mas.active_type) t2 AS active_type # LATERAL VIEW explode(mas.active_type) t2 AS active_type
)dev # )dev
on t1.cl_id=dev.device_id and t1.partition_date = dev.partition_date # on t1.cl_id=dev.device_id and t1.partition_date = dev.partition_date
LEFT JOIN # LEFT JOIN
( # (
SELECT DISTINCT device_id # 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） # 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={yesterday_str} # WHERE partition_day={yesterday_str}
#
UNION ALL # UNION ALL
SELECT DISTINCT device_id # SELECT DISTINCT device_id
FROM dim.dim_device_user_staff --去除内网用户 # FROM dim.dim_device_user_staff --去除内网用户
)spam_pv # )spam_pv
on spam_pv.device_id=t1.cl_id # on spam_pv.device_id=t1.cl_id
LEFT JOIN # LEFT JOIN
( # (
SELECT partition_date,device_id # SELECT partition_date,device_id
FROM # FROM
(--找出user_id当天活跃的第一个设备id # (--找出user_id当天活跃的第一个设备id
SELECT user_id,partition_date, # SELECT user_id,partition_date,
if(size(device_list) > 0, device_list [ 0 ], '') AS device_id # if(size(device_list) > 0, device_list [ 0 ], '') AS device_id
FROM online.ml_user_updates # FROM online.ml_user_updates
WHERE partition_date>='${{yesterday_str}}' AND partition_date<'${{today_str}}' # WHERE partition_date>='${{yesterday_str}}' AND partition_date<'${{today_str}}'
)t1 # )t1
JOIN # JOIN
( --医生账号 # ( --医生账号
SELECT distinct user_id # SELECT distinct user_id
FROM online.tl_hdfs_doctor_view # FROM online.tl_hdfs_doctor_view
WHERE partition_date = {yesterday_str} # WHERE partition_date = {yesterday_str}
#
--马甲账号/模特用户 # --马甲账号/模特用户
UNION ALL # UNION ALL
SELECT user_id # SELECT user_id
FROM ml.ml_c_ct_ui_user_dimen_d # FROM ml.ml_c_ct_ui_user_dimen_d
WHERE partition_day = {yesterday_str} # WHERE partition_day = {yesterday_str}
AND (is_puppet = 'true' or is_classifyuser = 'true') # AND (is_puppet = 'true' or is_classifyuser = 'true')
#
UNION ALL # UNION ALL
--公司内网覆盖用户 # --公司内网覆盖用户
select distinct user_id # select distinct user_id
from dim.dim_device_user_staff # from dim.dim_device_user_staff
#
UNION ALL # UNION ALL
--登陆过医生设备 # --登陆过医生设备
SELECT distinct t1.user_id # SELECT distinct t1.user_id
FROM # FROM
( # (
SELECT user_id, v.device_id as device_id # SELECT user_id, v.device_id as device_id
FROM online.ml_user_history_detail # FROM online.ml_user_history_detail
LATERAL VIEW EXPLODE(device_history_list) v AS device_id # LATERAL VIEW EXPLODE(device_history_list) v AS device_id
WHERE partition_date = {yesterday_str} # WHERE partition_date = {yesterday_str}
)t1 # )t1
JOIN # JOIN
( # (
SELECT device_id # SELECT device_id
FROM online.ml_device_history_detail # FROM online.ml_device_history_detail
WHERE partition_date = {yesterday_str} # WHERE partition_date = {yesterday_str}
AND is_login_doctor = '1' # AND is_login_doctor = '1'
)t2 # )t2
ON t1.device_id = t2.device_id # ON t1.device_id = t2.device_id
)t2 # )t2
on t1.user_id=t2.user_id # on t1.user_id=t2.user_id
group by t1.partition_date,device_id # group by t1.partition_date,device_id
)dev1 # )dev1
on t1.partition_date=dev1.partition_date and t1.cl_id=dev1.device_id # on t1.partition_date=dev1.partition_date and t1.cl_id=dev1.device_id
WHERE (spam_pv.device_id IS NULL or spam_pv.device_id ='') # WHERE (spam_pv.device_id IS NULL or spam_pv.device_id ='')
and (dev1.device_id is null or dev1.device_id ='') # and (dev1.device_id is null or dev1.device_id ='')
GROUP BY t1.partition_date,active_type,device_os_type,channel # GROUP BY t1.partition_date,active_type,device_os_type,channel
)t4 # )t4
on t3.partition_date=t4.partition_date and t3.active_type=t4.active_type and t3.device_os_type = t4.device_os_type AND t3.channel = t4.channel # on t3.partition_date=t4.partition_date and t3.active_type=t4.active_type and t3.device_os_type = t4.device_os_type AND t3.channel = t4.channel
#
#
#
""".format(today_str=today_str, yesterday_str=yesterday_str, ) # """.format(today_str=today_str, yesterday_str=yesterday_str, )
device_df = spark.sql(sql) # device_df = spark.sql(sql)
device_df.show(1, False) # device_df.show(1, False)
sql_res = device_df.collect() # sql_res = device_df.collect()
for res in sql_res: # for res in sql_res:
print(res) # print(res)
device_df.createOrReplaceTempView("data_table") # device_df.createOrReplaceTempView("data_table")
# #
# collects_sql = """ # collects_sql = """
# SELECT device_type,active_type,channel_type,ROUND(if(NVL(sum(uv),0) <> 0 ,NVL(sum(search_core_pv),0)/NVL(sum(uv),0) ,0),5) as core_pv_division_uv, # SELECT device_type,active_type,channel_type,ROUND(if(NVL(sum(uv),0) <> 0 ,NVL(sum(search_core_pv),0)/NVL(sum(uv),0) ,0),5) as core_pv_division_uv,
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment