我目前正在尝试使用Laravel / Eloquent执行查询,但是我决定首先编写SQL查询,以便找出问题所在。
我想按日期分组并使用多个联接合并两个表的结果,但是当我像这样添加product_metrics
表时,结果会出错:
inner join `product_metrics` on `products`.`id` = `product_metrics`.`product_id`
select
date_format(`date` + INTERVAL 1 HOUR, '%Y-%m-%d') as day,
round(sum(`item_price` * `quantity_ordered`), 2) as sales,
sum(`quantity_ordered`) as units,
sum(
CASE
WHEN `is_refund` THEN `item_price` * `quantity_ordered`
ELSE 0
END
) as refunds,
sum(
CASE
WHEN `is_refund` THEN `quantity_ordered`
ELSE 0
END
) as refund_units,
sum(`promotion_discount`) as discounts,
sum(`sessions`) as sessions,
count(DISTINCT(customer_order_id)) as orders,
round(sum(`product_sales`), 2) as sales_sc,
round(
sum(`unit_session_percentage` * `product_sales`) / sum(`product_sales`),
2
) as cvr,
sum(`page_views`) as views
from
`products`
inner join `product_brands` on `products`.`brand_id` = `product_brands`.`id`
inner join `customer_order_items` on `products`.`id` = `customer_order_items`.`product_id`
inner join `customer_orders` on `customer_order_items`.`customer_order_id` = `customer_orders`.`id`
inner join `product_metrics` on `products`.`id` = `product_metrics`.`product_id`
inner join `customers` on `products`.`customer_id` = `customers`.`id`
inner join `marketplaces` on `products`.`marketplace_id` = `marketplaces`.`id`
where
(
`date` between "2019-05-01 00:00:00"
and "2019-05-31 00:00:00"
or `purchase_date` between "2019-05-01 00:00:00"
and "2019-05-31 00:00:00"
and `customers`.`id` = 18
)
and `products`.`deleted_at` is null
group by
date_format(`date` + INTERVAL 1 HOUR, '%Y-%m-%d')
order by
`day` asc
我想要/需要做的是:
使用where子句查询products
,并添加日期以获取给定时间段内与这些产品相关的所有product_metrics
和所有customer_orders
。
答案 0 :(得分:0)
你是说...
date between "2019-05-01 00:00:00" and "2019-05-31 00:00:00"
or ( purchase_date between "2019-05-01 00:00:00" and "2019-05-31 00:00:00"
and customers.id = 18)
?
此外,如果您为表加上别名并使用这些别名限定列,那么操作起来会容易得多。
答案 1 :(得分:0)
我的猜测是您缺少括号以确保OR优先处理。
A or B and C ==> A or ( B and C )
您可能想要
( A or B ) and C
所以
where
(
( `date` between "2019-05-01 00:00:00" and "2019-05-31 00:00:00"
or `purchase_date` between "2019-05-01 00:00:00" and "2019-05-31 00:00:00"
)
and `customers`.`id` = 18
)
答案 2 :(得分:0)
考虑到产品和指标之间的一对多关系,这是预期的。