MySQL - use of aggregate functions and group by groups

MySQL—— Aggregate functions and group by The use of groups

1、 Aggregate function introduction

SQL The aggregate function calculates a set of data and returns a single value .

except COUNT outside , The aggregate function ignores null values , If COUNT The application object of the function is a certain column name , And the column has a null value , here COUNT Null values are still ignored .

Because the aggregate function operates on a set of values , So it usually works with SELECT Of the statement GROUP BY Clauses are used together , To calculate the metrics that provide information for each group .

2、GROUP BY grouping

Grouping is one of the most important tasks that must be handled when using a database . To group rows , Use GROUP BY Clause .

GROUP BY Clause is SELECT Optional clause of the statement , It groups rows based on matching values in the specified column , Each group returns to a row .

GROUP BY Syntax of clause ：

SELECT
    column1,
    column2,
    GROUP_FUNCTION (column3)
FROM
    table1
WHERE
    a = b
GROUP BY
    column1,
    column2
HAVING
    c = d
ORDER BY
    column2 DESC;

stay SELECT The inclusion of aggregate functions in clauses is not mandatory . however , If you use aggregate functions , It will calculate the sum of each group .

It's important to note that , Apply before grouping rows WHERE Clause , After grouping the rows, apply HAVING Clause . let me put it another way ,WHERE Clause applies to the row , and HAVING Clause applies to grouping .

To sort groups , Please be there. GROUP BY Clause followed by ORDER BY Clause .

GROUP BY The columns appearing in the clause are called grouping columns . If the grouping column contains NULL value , Then all NULL The values are summarized into a group , because GROUP BY The clause says NULL The values are equal .

3、 Common aggregate functions

The name of the function	effect
MAX	Query the maximum value of the specified column
MIN	Query the minimum value of the specified column
COUNT	Count the number of rows in the query result
SUM	Sum up , Returns the sum of the specified columns
AVG	averaging , Returns the average of the specified column data

AVG/ SUM Only applicable to fields of numeric type ( Or variable )

MAX / MIN For numeric types 、 String type 、 Fields of date time type （ Or variable )

Use COUNT(*)、COUNT(1)、COUNT( Specific fields ) Which is more efficient ：

If you are using MyISAM Storage engine , Then the efficiency of the three is the same , All are O(1)

If you are using InnoDB Storage engine , Then the efficiency of the three COUNT(*) = COUNT(1) >COUNT( Field )

Using examples

-- AVG  Calculate the average wage per department 
SELECT e.department_id,department_name, ROUND(AVG(salary), 0) avg_salary
FROM employees e
JOIN departments d on e.department_id = d.department_id
GROUP BY department_name
ORDER BY department_name;

-- SUM  Return the total salary of all employees in each department 
SELECT department_id, SUM(salary)
FROM employees
GROUP BY department_id;

-- MAX / MIN  Return the minimum and maximum salary of employees in each department 
SELECT department_name, MIN(salary) min_salary,MAX(salary) max_salary
FROM employees e
JOIN departments d on e.department_id = d.department_id
GROUP BY department_name
ORDER BY department_name;

-- COUNT  Return the number of people in each department in ascending order according to the Department name 
SELECT department_name, COUNT(*) headcount
FROM employees e
JOIN departments d on e.department_id = d.department_id
GROUP BY department_name
ORDER BY department_name;

--  Query minimum salary greater than 6000 Information of various departments 、 Maximum salary and average salary 
select e.department_id,department_name,min(salary) min_salary,max(salary) max_salary,round(avg(salary),2) average_salary
from employees e
join departments d on e.department_id = d.department_id
GROUP BY e.department_id
having min_salary > 6000
order by department_id ;

--  The number of searchers is greater than 5 The department in charge of the 
SELECT e.department_id,department_name,COUNT(employee_id) headcount
FROM employees e
JOIN departments d ON d.department_id = e.department_id
GROUP BY e.department_id
HAVING headcount > 5
ORDER BY headcount DESC;

#1.where Clause can be filtered using group functions ?
--  Can not be 

#2. Query the maximum salary of employees in the company , minimum value , Average , The sum of the 
select max(salary),min(salary),avg(salary),sum(salary) from employees;

#3. Query each job_id It's the maximum wage of employees , minimum value , Average , The sum of the 
select job_id, max(salary),min(salary),avg(salary),sum(salary)
from employees
group by job_id;

#4. Select each job_id Number of employees 
select job_id,count(*)
from employees
group by job_id;

#5. Find out the difference between the maximum wage and the minimum wage (DIFFERENCE)
select max(salary) - min(salary) DIFFERENCE
from employees;

#6. Check the minimum wage of employees under each manager , The minimum wage should not be lower than 6000, Employees without managers are not counted 
select emp.employee_id,emp.manager_id, min(emp.salary) min_salary
from employees emp
join employees mang
on emp.manager_id = mang.employee_id
group by emp.manager_id
having min_salary >= 6000;
--  or 
select employee_id,manager_id, min(salary) min_salary
from employees
where manager_id is not null
group by manager_id
having min_salary >= 6000;

#7. Query the names of all departments ,location_id, Number of employees and average salary , And in descending order of average wage 
select department_name,location_id,count(employee_id),round(avg(salary),2) avg_salary
from  departments d
left join employees e on e.department_id = d.department_id
group by department_name
order by avg_salary desc;

#8. Query each type of work 、 The name of each department 、 The name of the type of work and the minimum wage 
select department_name,job_id,min(salary)
from employees e
right join departments d on e.department_id = d.department_id
group by job_id, d.department_id;

4、SQL Execution order

SELECT The complete structure of the statement （SQL99）

select  duplicate removal     Fields to query     from    surface （ Be careful ： Tables and fields can be aliased ）
xxxx  join  Table to connect      on    Equivalent judgment ( The order ： First on Again where)
where （ Specific value / Subquery , Filter conditions that do not contain aggregate functions ）
group by( Group by that sub segment )
having （ Filter the grouped information , Conditions and where equally , Different location , Filter conditions containing aggregate functions ）
order by（ By which field ）
limit （ Pagination ）

SQL The order in which statements are executed