上传文件到数据库中
--上传文件逻辑
--#1 建表
hive -e"
drop table da_ppd_oba.tmp_liuzhen_bbs_huifu_70_20201201;
create table da_ppd_oba.tmp_liuzhen_bbs_huifu_70_20201201
(
openid string
)
comment '临时表'
row format delimited fields terminated by'\t'
stored as textfile;
"
--上传数据
#2 上传数据
hive -e"
load data local inpath '回复数大于70用户帐号.txt' into table da_ppd_oba.tmp_liuzhen_bbs_huifu_70_20201201;
"
随机抽取部分数据
order by rand() 函数
--第6层-发帖用户筛选
drop table da_ppd_oba.tmp_liuzhen_bbs_V6_20201201;
create table da_ppd_oba.tmp_liuzhen_bbs_V6_20201201
as
select
a.imei as imei
,num
from
(
select
imei
,count(distinct openid) as num
from
da_ppd_oba.tmp_liuzhen_bbs_post_imei_20201201
group by
imei
)a
where
num = 1
order by rand()
limit 40000