数据仓库 DWS层之用户行为宽表

为什么需要用户行为宽表？把每个用户单日的行为聚合起来组成一张多列宽表，以便之后关联用户维度信息后，进行不同角度的统计分析。

数据来源：DWD层相关的业务数据表

创建用户行为宽表：

这张宽表整合了下单、支付和评论3种行为。

drop table if exists dws_user_action;
create external table dws_user_action 
(   
    user_id          string      comment '用户 id',
    order_count     bigint      comment '下单次数 ',
    order_amount    decimal(16,2)  comment '下单金额 ',
    payment_count   bigint      comment '支付次数',
    payment_amount  decimal(16,2) comment '支付金额 ',
    comment_count   bigint      comment '评论次数'
) COMMENT '每日用户行为宽表'
PARTITIONED BY (`dt` string)
stored as parquet
location '/warehouse/gmall/dws/dws_user_action/'
tblproperties ("parquet.compression"="snappy");

数据导入脚本：

with as基本语法为如下，作用是定义一个临时表，可以在后续的语句中多次使用，提高sql可读性。注意多个临时表之间用逗号，而最后一个临时表和查询语句之间没有符号。

WITH t1 AS (
        SELECT *
        FROM carinfo
    ), 
    t2 AS (
        SELECT *
        FROM car_blacklist
    )
SELECT *
FROM t1, t2

#!/bin/bash

# 定义变量方便修改
APP=gmall
hive=/opt/module/hive/bin/hive

# 如果是输入的日期按照取输入日期；如果没输入日期取当前时间的前一天
if [ -n "$1" ] ;then
    do_date=$1
else 
    do_date=`date -d "-1 day" +%F`  
fi 

sql="

with  
tmp_order as
(
    select 
        user_id, 
        sum(oi.total_amount) order_amount, 
        count(*)  order_count
    from "$APP".dwd_order_info  oi
    where date_format(oi.create_time,'yyyy-MM-dd')='$do_date'
    group by user_id
)  ,
tmp_payment as
(
    select 
        user_id, 
        sum(pi.total_amount) payment_amount, 
        count(*) payment_count 
    from "$APP".dwd_payment_info pi 
    where date_format(pi.payment_time,'yyyy-MM-dd')='$do_date'
    group by user_id
),
tmp_comment as
(  
    select  
        user_id, 
        count(*) comment_count
    from "$APP".dwd_comment_log c
    where date_format(c.dt,'yyyy-MM-dd')='$do_date'
    group by user_id 
)

Insert overwrite table "$APP".dws_user_action partition(dt='$do_date')
select 
    user_actions.user_id, 
    sum(user_actions.order_count), 
    sum(user_actions.order_amount),
    sum(user_actions.payment_count), 
    sum(user_actions.payment_amount),
    sum(user_actions.comment_count) 
from
(
    select
        user_id,
        order_count,
        order_amount,
        0 payment_count,
        0 payment_amount,
        0 comment_count
    from tmp_order

    union all
    select
        user_id,
        0,
        0,
        payment_count,
        payment_amount,
        0
    from tmp_payment

    union all
    select
        user_id,
        0,
        0,
        0,
        0,
        comment_count 
    from tmp_comment
 ) user_actions
group by user_id;
"

$hive -e "$sql"