大数据之Hive：with tmp1 as ()

1、with tmp1 as ()

语义：将（）内查询的结果，取表名tmp1；

2、好处

本质是子查询，这样写的好处是：

1. 写sql和看的sql的时候结构更清晰明了；
2. 如果你写是sql 有复用这些子查询结果的情况，只需要先算一次就可以了，因为这个结果会缓存在内存里面，如果采用传统的方式，用几次就算几次。

3、实例

with
tmp_login as
(
    select
        user_id,
        count(*) login_count
    from dwd_start_log
    where dt='2021-03-20'
    and user_id is not null
    group by user_id
),
tmp_cart as
(
    select
        user_id,
        count(*) cart_count
    from dwd_action_log
    where dt='2021-03-20'
    and user_id is not null
    and action_id='cart_add'
    group by user_id
),tmp_order as
(
    select
        user_id,
        count(*) order_count,
        sum(final_total_amount) order_amount
    from dwd_fact_order_info
    where dt='2021-03-20'
    group by user_id
) ,
tmp_payment as
(
    select
        user_id,
        count(*) payment_count,
        sum(payment_amount) payment_amount
    from dwd_fact_payment_info
    where dt='2021-03-20'
    group by user_id
),
tmp_order_detail as
(
    select
        user_id,
        collect_set(named_struct('sku_id',sku_id,'sku_num',sku_num,'order_count',order_count,'order_amount',order_amount)) order_stats
    from
    (
        select
            user_id,
            sku_id,
            sum(sku_num) sku_num,
            count(*) order_count,
            cast(sum(final_amount_d) as decimal(20,2)) order_amount
        from dwd_fact_order_detail
        where dt='2021-03-20'
        group by user_id,sku_id
    )tmp
    group by user_id
)

insert overwrite table dws_user_action_daycount partition(dt='2021-03-20')
select
    tmp_login.user_id,
    login_count,
    nvl(cart_count,0),
    nvl(order_count,0),
    nvl(order_amount,0.0),
    nvl(payment_count,0),
    nvl(payment_amount,0.0),
    order_stats
from tmp_login
left join tmp_cart on tmp_login.user_id=tmp_cart.user_id
left join tmp_order on tmp_login.user_id=tmp_order.user_id
left join tmp_payment on tmp_login.user_id=tmp_payment.user_id
left join tmp_order_detail on tmp_login.user_id=tmp_order_detail.user_id;