Question

我正在使用以下带有3个左联接的查询：

SELECT `Date`,`Year`,`Month`, `Day`,`Hour`, `Order Line Total` from 

(SELECT order_updated_at, DATE(order_updated_at) as `Date`,  YEAR(order_updated_at) as `Year`, MONTHNAME(order_updated_at) as `Month`, DAYNAME(order_updated_at) as `Day`, order_time_hour_num as `Hour`, orderline_product_id, order_staff_member_id, (orderline_product_total * orderline_product_quantity) AS `Order Line Total` FROM order-lines-table) ORDERS

LEFT JOIN
(SELECT staff_id, first_name, last_name, concat(first_name, ' ', last_name) AS `Staff Member`
FROM staff-table) STAFF
ON ORDERS.order_staff_member_id = STAFF.staff_id

LEFT JOIN
(SELECT * from  
(SELECT category_id, product_id, product_name
FROM products-table group by product_id) PRODUCTS

LEFT JOIN     
(SELECT categories_id, categories_name
FROM categories-table group by categories_id) CATEGORIES
ON PRODUCTS.category_id = CATEGORIES.categories_id) PRODUCTS_AND_CATGORIES

ON ORDERS.orderline_product_id = PRODUCTS_AND_CATGORIES.product_id

这需要一些时间（2-3秒），这并不理想。

每个查询花费的时间不到100毫秒，但是在一起（大概是由于连接的原因）要花很多时间才能完成。

有人知道有其他方法可以使用不同的联接方法来优化此操作吗？或者这是唯一的选择吗？

Answer 1

在您已经使用联接的情况下，我认为在这种情况下您不需要子查询。如果我认为您的表结构和数据正确，则无需分组。尚未尝试过，但是以下内容应能提供与原始查询相同的结果，但速度要快得多（且可读）：

SELECT 
    DATE(order_updated_at) as `Date`,
    YEAR(order_updated_at) as `Year`,
    MONTHNAME(order_updated_at) as `Month`,
    DAYNAME(order_updated_at) as `Day`,
    order_time_hour_num as `Hour`, 
    (orderline_product_total * orderline_product_quantity) AS `Order Line Total`
FROM order-lines-table olt
LEFT JOIN staff-table st ON olt.order_staff_member_id = st.staff_id
LEFT JOIN products-table pt ON olt.orderline_product_id = pt.product_id
LEFT JOIN categories-table ct ON pt.categories_id = ct.categories_id

通过3个左联接加快MySQL查询

1 个答案: