【MySQL】 语法顺序和执行顺序

【一、语法顺序】

写sql的关键字顺序

select [distinct]
from
join（如left join）
on
where
group by
having
union
order by
limit

---

【二、执行顺序】

在数据库底层执行时sql按照下面的顺序进行执行

from
on
join
where
group by
having
select
distinct
union
order by

---

【三、实战篇】

建立如下表格orders：

注：下面所有语句符合语法顺序，只分析其执行顺序：(join和on属于多表查询，放在最后展示)

---

普通查询：

select a.Customer 
from orders a 
where a.Customer='Bush' or a.Customer = 'Adams'

分析：首先是from语句找到表格，然后根据where得到符合条件的记录，最后select出需要的字段，结果如下：

---

groupby：

groupby要和聚合函数一起使用

select a.Customer,sum(a.OrderPrice)
from orders a
where a.Customer='Bush' or a.Customer = 'Adams'
group by a.Customer

分析：在from，where执行后，执行group by，同时也根据group by的字段，执行sum这个聚合函数。这样的话得到的记录对group by的字段来说是不重复的，结果如下：

---

having：

select a.Customer,sum(a.OrderPrice)
from orders a
where a.Customer='Bush' or a.Customer = 'Adams'
group by a.Customer
having sum(a.OrderPrice) > 2000

分析：由于where是在group之前执行，那么如何对group by的结果进行筛选，就用到了having，结果如下：

---

distinct：

为测试，先把数据库中Adams那条记录的OrderPrice改为3000

select distinct sum(a.OrderPrice)
from orders a
where a.Customer='Bush' or a.Customer = 'Adams' or a.Customer = 'Carter'
group by a.Customer
having sum(a.OrderPrice) > 1700

分析：将得到一条记录（没有distinct，将会是两条同样的记录）：

---

union：

完全是对select的结果进行合并（默认去掉重复的记录）

select distinct sum(a.OrderPrice) As Order1
from orders a
where a.Customer='Bush' or a.Customer = 'Adams' or a.Customer = 'Carter'
group by a.Customer
having sum(a.OrderPrice) > 1500
union
select distinct sum(a.OrderPrice) As Order1
from orders a
where a.Customer='Bush' or a.Customer = 'Adams' or a.Customer = 'Carter'
group by a.Customer
having sum(a.OrderPrice) > 2000

分析：默认去掉重复记录（想保留重复记录使用union all），结果如下：

---

order by：

排序

select distinct sum(a.OrderPrice) As order1
from orders a
where a.Customer='Bush' or a.Customer = 'Adams' or a.Customer = 'Carter'
group by a.Customer
having sum(a.OrderPrice) > 1500
union
select distinct sum(a.OrderPrice) As order1
from orders a
where a.Customer='Bush' or a.Customer = 'Adams' or a.Customer = 'Carter'
group by a.Customer
having sum(a.OrderPrice) > 2000
order by order1

分析：升序排序，结果如下：

---

limit：

取指定条记录

select distinct sum(a.OrderPrice) As order1
from orders a
where a.Customer='Bush' or a.Customer = 'Adams' or a.Customer = 'Carter'
group by a.Customer
having sum(a.OrderPrice) > 1500
union
select distinct sum(a.OrderPrice) As order1
from orders a
where a.Customer='Bush' or a.Customer = 'Adams' or a.Customer = 'Carter'
group by a.Customer
having sum(a.OrderPrice) > 2000
order by order1
limit 1

分析：取出结果中的前1条记录，结果如下：

---

join on：

多表联查

select distinct sum(a.OrderPrice) As order1,sum(d.OrderPrice) As order2
from orders a
left join (select c.* from Orders c) d 
on a.O_Id = d.O_Id
where a.Customer='Bush' or a.Customer = 'Adams' or a.Customer = 'Carter'
group by a.Customer
having sum(a.OrderPrice) > 1500
union
select distinct sum(a.OrderPrice) As order1,sum(e.OrderPrice) As order2
from orders a
left join (select c.* from Orders c) e 
on a.O_Id = e.O_Id
where a.Customer='Bush' or a.Customer = 'Adams' or a.Customer = 'Carter'
group by a.Customer
having sum(a.OrderPrice) > 2000
order by order1
limit 1

分析：上述语句其实join on就是多连接了一张表，而且是两张一样的表，都是Orders。 执行过程是，在执行from关键字之后根据on指定的条件，把left join指定的表格数据附在from指定的表格后面，然后再执行where字句。

---

注：

• 使用distinct要写在所有要查询字段的前面，后面有几个字段，就代表修饰几个字段，而不是紧随distinct的字段；
• group by执行后(有聚合函数)，group by后面的字段在结果中一定是唯一的，也就不需要针对这个字段用distinct；