作者简介：C/C++领域新星创作者,为C++和java奋斗中
个人社区：微凉秋意社区
系列专栏：MySql一点通
推荐一款模拟面试、刷题神器注册免费刷题

前言

书接上文,Advanced operation learning queries today,包括函数、分组、分页、子查询和表连接.内容比较实用,Recommended collection or subscribe to the column is convenient to study and review.

一、MySQL函数的使用

• mysql中内置了很多函数,Each function represents a specific functions

1、单行函数

• 特点：Each row of data will produce a corresponding results
  • 表中有多少行,How many rows will be the result
• 位置：select之后,from之前 Or as a condition of screening

1. concat（…,列名,…,列名）
   Used for splicing of multiple columns values

   • 例如：以firstname+lastnameShow all employees, in the form of name
     select concat(first_name,last_name) as 全名 from employees

2. mod(值1,值2)
   求值1和值2的取余结果 ,Equivalent in mathematics %（值1%值2）

   • 标准SQLThe specification does not exist%Take more than writing,So take more operation should be usedmod()
   • 例如：select mod(10,3) from dual
     注： dual为虚拟表,In order to make the currentSQL符合语法规范,When they don't write to do automatic filling

3. length(列名|数据)
   获取长度

   • 查询所有firstname长度>6的员工信息
     select * from employees where length(first_name)>6

4. now() | sysdate()
   获取系统当前时间

   • select now()
     select SYSDATE()
     select now() from dual

2、多行函数(组函数)

• As a group for the unit to obtain the results,A group to get a result
• If you do not manually to group,则默认整张表为一组

1. sum(列名)
   • For the sum of a column
2. avg(列名)
   • 求某一列的平均值
3. max(列名)
   • 求某一列的最大值
4. min(列名)
   • 求某一列的最小值
5. count(列名)
   • For a number of column values(非空)

示例：

-- Set of functions operatingsalary
select 
SUM(salary) 总和,AVG(salary) 平均值,MAX(salary) 最大值,min(salary) 最小值,COUNT(salary) 非空值的数量
from employees
-- Computational performance are listed in the table is not the number of employees is empty
select COUNT(commission_pct) from employees
-- For the rows in the table
select COUNT(*) from employees
select COUNT(employee_id) from employees

二、MySQL的分组

• 在MySQLIn a group is an operating unit

1. 简单分组
   select 列名 from 表名 group by 列名
   • An average salary of each department by department as the unit
     select department_id,AVG(salary) 平均薪资 from employees group by department_id
2. where + 分组

-- 查询部门id为10、20、30的部门的平均薪资、
select department_id,AVG(salary) 平均薪资 
from employees
where department_id in(10,20,30) -- 先筛选
GROUP BY department_id -- After grouping query

3. having + 分组

select department_id,AVG(salary) 平均薪资 
from employees
group by department_id -- 先分组
having department_id in(10,20,30)  -- 后筛选

4. 分组中 where 和 having 的区别
   • whereIs to perform in front of the group,havingAfter group perform
   • whereNot in the group with a set of functions,但是having可以
   • When both are available,优先使用where,效率更高

三、Paging and query keyword execution order

• limit：Used to display the query results limit article number,通常用于分页查询
  select 列名 from 表名 limit According to the starting index,显示条数

  • Query all employees data before ten
    select * from employees limit 0,10
select * from employees limit 10下标为0时可以省略

• 当limitAnd other query keyword combination,limitThe last execution must be

查询的执行顺序：

语法顺序		执行顺序
select			5
from			1
where			2
group by		3
having			4
order by		6
limit			7

四、子查询

• 当一个SQLRely on anotherSQL的执行结果,在当前SQL中嵌套另一个SQL,This method is called the subquery

1、where子查询

1.1、单值子查询

• 嵌套的子SQLReturns only a line of a column of the result

-- 查询员工id为100的员工的部门id
select department_id from employees where employee_id=100

-- 子查询：：查询员工id为100的员工所在的部门信息
select * from departments 
where department_id=(select department_id from employees where employee_id=100)

• Prioritise the innerSQL,再执行外层SQL
• 子SQLNeed to write the brackets

1.2、多值子查询

• 嵌套的子SQLCan return multiple results

--查询firstname中包含s的员工所在的部门信息
select * from departments 
where department_id in(select department_id from employees where first_name like '%s%')

2、from子查询

• 将子SQLThe query results as a temporary table for subsequent operation

-- To get the top ten employees' average salary
-- To get the highest salary first top ten employees salary
select salary from employees ORDER BY salary desc LIMIT 10
-- 子查询：再根据子SQLResults temporary query through the alias to construct a data table
select avg(salary) from (select salary from employees ORDER BY salary desc LIMIT 10) as e

五、表连接

• When the query results need to be obtained from multiple tables,You need to connect multiple tables for novelty search operation

The premise of table joins：

1. There is a connection between multiple tables must be relationship（外键）
2. A foreign key is usually connected to another table primary key
3. In the table as a condition of connection using

1、内连接

语法：

select 表1.列名,表2.列名,... 
from 表1 inner join 表2
on 连接条件

• 特点：The two tables at the same time constraint,Only when all the tables are conform to join condition,Will display information
• 使用：inner可省

2、(左)外连接

语法：

select 表1.列名,表2.列名,... 
from 表1(左表) left outer join 表2（右表）
on 连接条件

• 特点：Only to the right table do constraint,Left the data in the table will show,Right in the table only if it meets the requirements of connection will show
• 使用：outer可省
• Left outer join and right outer join isleft和right的不同,位置不同,Usually use the left outer join
• 使用：

-- Check all employee information and meet the requirements for the connection of department information
select e.*,d.*
from employees e LEFT JOIN departments d
ON e.department_id=d.department_id

全外连接：

• Merge the two queries results show
  查询语句1 union 查询语句2
• 使用：

-- 内连接
select e.*,d.*  
from employees e inner join departments d
on  e.department_id=d.department_id

union -- Internal connection and left the connection of the query results to merge

-- 左连接
select e.*,d.*
from employees e LEFT JOIN departments d
ON e.department_id=d.department_id

• unionWill the merger results are去重
• union all不会对结果去重
• 特点：Merge both sides do not restrain the
• Merge both the number of query results in field、顺序必须一致

实际开发中,Using the high frequency is the internal connection and left outer join

3、自连接

• 概念：Is a special table join,Participate in the connection is the same table

具体使用：

1. A certain relationship between the two fields in the table

-- 查询：员工id、员工姓名-firstname、Direct leadership ofid、Direct leadership of姓名
select e1.employee_id 员工id,e1.first_name 员工姓名,e2.employee_id 领导id,e2.first_name 领导姓名
from employees e1 LEFT JOIN employees e2-- e1On behalf of the employee information table,e2On behalf of the leading information table
ON e1.manager_id=e2.employee_id -- The two correlation fields as a condition of connection

2. Judgment of the same field,作为连接条件

-- Query the same wages employee information
select e1.employee_id 员工1的id,e1.salary 员工1的工资,e2.employee_id 员工2的id,e2.salary 员工2的工资
from employees e1 LEFT JOIN employees e2
on e1.salary=e2.salary -- 连接条件
where e1.employee_id<e2.employee_id -- Both sides to prevent judgment is the same person,并去重

写在最后
This blog isMySQLIn the blockbuster content,I hope you can properly absorb and correct use of,一起加油！！！