MySQL分组聚合

where 与 having 的区别 & order by

假如我们有一张表，表名为 sales，如下所示：

+----------------+-------------+--------------+---------+----------+---------------+
| transaction_id | customer_id | product_name | price   | quantity | purchase_date |
+----------------+-------------+--------------+---------+----------+---------------+
|              1 |         101 | Laptop       | 1200.00 |        2 | 2024-06-01    |
|              2 |         102 | Phone        |  800.00 |        1 | 2024-06-02    |
|              3 |         101 | Tablet       |  500.00 |        3 | 2024-06-03    |
|              4 |         103 | Laptop       | 1200.00 |        1 | 2024-06-04    |
|              5 |         102 | Headphones   |  100.00 |        2 | 2024-06-05    |
|              6 |         101 | Phone        |  800.00 |        1 | 2024-06-06    |
|              7 |         104 | Tablet       |  500.00 |        2 | 2024-06-07    |
|              8 |         102 | Laptop       | 1200.00 |        1 | 2024-06-08    |
|              9 |         101 | Phone        |  800.00 |        2 | 2024-06-09    |
|             10 |         103 | Headphones   |  100.00 |        1 | 2024-06-10    |
+----------------+-------------+--------------+---------+----------+---------------+

having子句

having 子句用于在 SQL 查询中对分组后的结果进行筛选，用于过滤分组后的数据。

where 子句对行数据进行筛选，用于过滤行数据。

想要找出购买商品数量大于 1 且总价值超过 1000 的客户。我们可以这样做：

select customer_id, sum(price * quantity) as total_purchase_value
from sales
group by customer_id
having sum(quantity) > 1 and sum(price * quantity) > 1000;+-------------+----------------------+
| customer_id | total_purchase_value |
+-------------+----------------------+
|         101 |              6300.00 |
|         102 |              2200.00 |
|         103 |              1300.00 |
+-------------+----------------------+
3 rows in set (0.00 sec)

在这个例子中只有满足这两个条件的客户才会包含在结果中。并且像上面having后面的就是分组后的数据。

查找购买了至少三种不同产品的客户。我们可以这样查询：

select customer_id, count(distinct product_name) as distinct_products_count
from sales
group by customer_id
having count(distinct product_name) >= 3;+-------------+-------------------------+
| customer_id | distinct_products_count |
+-------------+-------------------------+
|         101 |                       3 |
|         102 |                       3 |
+-------------+-------------------------+
2 rows in set (0.00 sec)

这里关于where就不在进行举例子说明了。

如果where和having都能输出我们想要的结果，那我们就首选where会提高我们查询的效率。

order by 子句

order by 子句用于在 SQL 查询结果中指定排序顺序。它允许按照一个或多个列的值对查询结果进行排序，默认情况下，order by会按照升序（asc）排列数据，但也可以指定降序（desc）。

以下是 order by 子句的一般语法结构：

select 列名1, 列名2, ...
from 表名
order by 列名1 [asc|desc], 列名2 [asc|desc], ...;

按客户ID和销售额排序：

select customer_id, sum(price * quantity) as total_sales
from sales
group by customer_id
order by customer_id asc, total_sales desc;+-------------+-------------+
| customer_id | total_sales |
+-------------+-------------+
|         101 |     6300.00 |
|         102 |     2200.00 |
|         103 |     1300.00 |
|         104 |     1000.00 |
+-------------+-------------+
4 rows in set (0.00 sec)

在这个例子中，order by customer_id asc, total_sales desc 指定了首先按客户ID升序排序，然后在同一客户ID内按总销售额降序排序。

按产品名称排序：

select product_name, sum(quantity) as total_quantity_sold
from sales
group by product_name
order by product_name asc;+--------------+---------------------+
| product_name | total_quantity_sold |
+--------------+---------------------+
| Headphones   |                   3 |
| Laptop       |                   4 |
| Phone        |                   4 |
| Tablet       |                   5 |
+--------------+---------------------+
4 rows in set (0.00 sec)

在这个例子中，order by product_name asc 指定了按产品名称的字母顺序进行升序排序。

按销售数量和单价排序：

select product_name, sum(quantity) as total_quantity, avg(price) as average_price
from sales
group by product_name
order by total_quantity desc, average_price desc;+--------------+----------------+---------------+
| product_name | total_quantity | average_price |
+--------------+----------------+---------------+
| Tablet       |              5 |    500.000000 |
| Laptop       |              4 |   1200.000000 |
| Phone        |              4 |    800.000000 |
| Headphones   |              3 |    100.000000 |
+--------------+----------------+---------------+
4 rows in set (0.00 sec)