#import job
type=command
command=sh import.sh
#import.sh
#!/bin/bash
dateStr=`date "+%Y%m%d"`
echo "---->${dateStr}"
log_path=/user_clicks/${dateStr}/
hive -e "use default; load data inpath '${log_path}' into table user_clicks partition(dt='${dateStr}');"
#analysis job
type=command
dependencies=import
command=sh analysis.shtype=command
dependencies=import
command=sh analysis.sh
# analysis.sh
#!/bin/bash
dateStr=`date "+%Y%m%d"`
hive -e "use default; insert into table user_info select count(distinct user_id) as active_num, dt from user_clicks where dt =$dateStr"
大数据作业
最新推荐文章于 2023-04-08 15:53:58 发布