【MySQL】基础篇-单行函数、聚合函数

1、单行函数

1.数值函数

#基本的操作
SELECT ABS(-123),ABS(32),SIGN(-23),SIGN(43),PI(),CEIL(32.32),CEILING(-43.23),FLOOR(32.32),
FLOOR(-43.23),MOD(12,5),12 MOD 5,12 % 5
FROM DUAL;

#取随机数
SELECT RAND(),RAND(),RAND(10),RAND(10),RAND(-1),RAND(-1)
FROM DUAL;

#四舍五入，截断操作
SELECT ROUND(123.556),ROUND(123.456,0),ROUND(123.456,1),ROUND(123.456,2),
ROUND(123.456,-1),ROUND(153.456,-2)
FROM DUAL;

SELECT TRUNCATE(123.456,0),TRUNCATE(123.496,1),TRUNCATE(129.45,-1)
FROM DUAL;

#单行函数可以嵌套
SELECT TRUNCATE(ROUND(123.456,2),0)
FROM DUAL;

#角度与弧度的互换

SELECT RADIANS(30),RADIANS(45),RADIANS(60),RADIANS(90),
DEGREES(2*PI()),DEGREES(RADIANS(60))
FROM DUAL;


#三角函数
SELECT SIN(RADIANS(30)),DEGREES(ASIN(1)),TAN(RADIANS(45)),DEGREES(ATAN(1))
FROM DUAL;

#指数和对数
SELECT POW(2,5),POWER(2,4),EXP(2)
FROM DUAL;

SELECT LN(EXP(2)),LOG(EXP(2)),LOG10(10),LOG2(4)
FROM DUAL;

#进制间的转换
SELECT BIN(10),HEX(10),OCT(10),CONV(10,10,8)
FROM DUAL;

2. 字符串函数

SELECT ASCII('Abcdfsf'),CHAR_LENGTH('hello'),CHAR_LENGTH('我们'),
LENGTH('hello'),LENGTH('我们')
FROM DUAL;

# xxx worked for yyy
SELECT CONCAT(emp.last_name,' worked for ',mgr.last_name) "details"
FROM employees emp JOIN employees mgr
WHERE emp.`manager_id` = mgr.employee_id;

SELECT CONCAT_WS('-','hello','world','hello','beijing')
FROM DUAL;
#字符串的索引是从1开始的！
SELECT INSERT('helloworld',2,3,'aaaaa'),REPLACE('hello','lol','mmm')
FROM DUAL;

SELECT UPPER('HelLo'),LOWER('HelLo')
FROM DUAL;

SELECT last_name,salary
FROM employees
WHERE LOWER(last_name) = 'King';

SELECT LEFT('hello',2),RIGHT('hello',3),RIGHT('hello',13)
FROM DUAL;

# LPAD:实现右对齐效果
# RPAD:实现左对齐效果
SELECT employee_id,last_name,LPAD(salary,10,' ')
FROM employees;

SELECT CONCAT('---',LTRIM('    h  el  lo   '),'***'),
TRIM('oo' FROM 'ooheollo')
FROM DUAL;

SELECT REPEAT('hello',4),LENGTH(SPACE(5)),STRCMP('abc','abe')
FROM DUAL;


SELECT SUBSTR('hello',2,2),LOCATE('lll','hello')
FROM DUAL;

SELECT ELT(2,'a','b','c','d'),FIELD('mm','gg','jj','mm','dd','mm'),
FIND_IN_SET('mm','gg,mm,jj,dd,mm,gg')
FROM DUAL;

SELECT employee_id,NULLIF(LENGTH(first_name),LENGTH(last_name)) "compare"
FROM employees;

3. 日期和时间函数

3.1 获取日期、时间

SELECT CURDATE(),CURRENT_DATE(),CURTIME(),NOW(),SYSDATE(),
UTC_DATE(),UTC_TIME()
FROM DUAL;

SELECT CURDATE(),CURDATE() + 0,CURTIME() + 0,NOW() + 0
FROM DUAL;

3.2 日期与时间戳的转换

SELECT UNIX_TIMESTAMP(),UNIX_TIMESTAMP('2021-10-01 12:12:32'),
FROM_UNIXTIME(1635173853),FROM_UNIXTIME(1633061552)
FROM DUAL;

3.3 获取月份、星期、星期数、天数等函数

SELECT YEAR(CURDATE()),MONTH(CURDATE()),DAY(CURDATE()),
HOUR(CURTIME()),MINUTE(NOW()),SECOND(SYSDATE())
FROM DUAL;

SELECT MONTHNAME('2021-10-26'),DAYNAME('2021-10-26'),WEEKDAY('2021-10-26'),
QUARTER(CURDATE()),WEEK(CURDATE()),DAYOFYEAR(NOW()),
DAYOFMONTH(NOW()),DAYOFWEEK(NOW())
FROM DUAL;

3.4 日期的操作函数

SELECT EXTRACT(SECOND FROM NOW()),EXTRACT(DAY FROM NOW()),
EXTRACT(HOUR_MINUTE FROM NOW()),EXTRACT(QUARTER FROM '2021-05-12')
FROM DUAL;

3.5 时间和秒钟转换的函数

SELECT TIME_TO_SEC(CURTIME()),
SEC_TO_TIME(83355)
FROM DUAL;

3.6 计算日期和时间的函数

SELECT NOW(),DATE_ADD(NOW(),INTERVAL 1 YEAR),
DATE_ADD(NOW(),INTERVAL -1 YEAR),
DATE_SUB(NOW(),INTERVAL 1 YEAR)
FROM DUAL;

SELECT DATE_ADD(NOW(), INTERVAL 1 DAY) AS col1,DATE_ADD('2021-10-21 23:32:12',INTERVAL 1 SECOND) AS col2,
ADDDATE('2021-10-21 23:32:12',INTERVAL 1 SECOND) AS col3,
DATE_ADD('2021-10-21 23:32:12',INTERVAL '1_1' MINUTE_SECOND) AS col4,
DATE_ADD(NOW(), INTERVAL -1 YEAR) AS col5, #可以是负数
DATE_ADD(NOW(), INTERVAL '1_1' YEAR_MONTH) AS col6 #需要单引号
FROM DUAL;

SELECT ADDTIME(NOW(),20),SUBTIME(NOW(),30),SUBTIME(NOW(),'1:1:3'),DATEDIFF(NOW(),'2021-10-01'),
TIMEDIFF(NOW(),'2021-10-25 22:10:10'),FROM_DAYS(366),TO_DAYS('0000-12-25'),
LAST_DAY(NOW()),MAKEDATE(YEAR(NOW()),32),MAKETIME(10,21,23),PERIOD_ADD(20200101010101,10)
FROM DUAL;

3.7 日期的格式化与解析

# 格式化：日期 ---> 字符串
# 解析：  字符串 ----> 日期

#此时我们谈的是日期的显式格式化和解析

#之前，我们接触过隐式的格式化或解析
SELECT *
FROM employees
WHERE hire_date = '1993-01-13';

#格式化：
SELECT DATE_FORMAT(CURDATE(),'%Y-%M-%D'),
DATE_FORMAT(NOW(),'%Y-%m-%d'),TIME_FORMAT(CURTIME(),'%h:%i:%S'),
DATE_FORMAT(NOW(),'%Y-%M-%D %h:%i:%S %W %w %T %r')
FROM DUAL;

#解析：格式化的逆过程
SELECT STR_TO_DATE('2021-October-25th 11:37:30 Monday 1','%Y-%M-%D %h:%i:%S %W %w')
FROM DUAL;

SELECT GET_FORMAT(DATE,'USA')
FROM DUAL;

SELECT DATE_FORMAT(CURDATE(),GET_FORMAT(DATE,'USA'))
FROM DUAL;

4.流程控制函数

4.1 IF(VALUE,VALUE1,VALUE2)

SELECT last_name,salary,IF(salary >= 6000,'高工资','低工资') "details"
FROM employees;

SELECT last_name,commission_pct,IF(commission_pct IS NOT NULL,commission_pct,0) "details",
salary * 12 * (1 + IF(commission_pct IS NOT NULL,commission_pct,0)) "annual_sal"
FROM employees;

4.2 IFNULL(VALUE1,VALUE2):看做是IF(VALUE,VALUE1,VALUE2)的特殊情况

SELECT last_name,commission_pct,IFNULL(commission_pct,0) "details"
FROM employees;

4.3 CASE WHEN … THEN …WHEN … THEN … ELSE … END

# 类似于java的if ... else if ... else if ... else
SELECT last_name,salary,CASE WHEN salary >= 15000 THEN '白骨精' 
			     WHEN salary >= 10000 THEN '潜力股'
			     WHEN salary >= 8000 THEN '小屌丝'
			     ELSE '草根' END "details",department_id
FROM employees;

SELECT last_name,salary,CASE WHEN salary >= 15000 THEN '白骨精' 
			     WHEN salary >= 10000 THEN '潜力股'
			     WHEN salary >= 8000 THEN '小屌丝'
			     END "details"
FROM employees;

4.4 CASE … WHEN … THEN … WHEN … THEN … ELSE … END

/*
练习1
查询部门号为 10,20, 30 的员工信息, 
若部门号为 10, 则打印其工资的 1.1 倍, 
20 号部门, 则打印其工资的 1.2 倍, 
30 号部门,打印其工资的 1.3 倍数,
其他部门,打印其工资的 1.4 倍数
*/
SELECT employee_id,last_name,department_id,salary,CASE department_id WHEN 10 THEN salary * 1.1
								     WHEN 20 THEN salary * 1.2
								     WHEN 30 THEN salary * 1.3
								     ELSE salary * 1.4 END "details"
FROM employees;

/*

练习2
查询部门号为 10,20, 30 的员工信息, 
若部门号为 10, 则打印其工资的 1.1 倍, 
20 号部门, 则打印其工资的 1.2 倍, 
30 号部门打印其工资的 1.3 倍数

*/
SELECT employee_id,last_name,department_id,salary,CASE department_id WHEN 10 THEN salary * 1.1
								     WHEN 20 THEN salary * 1.2
								     WHEN 30 THEN salary * 1.3
								     END "details"
FROM employees
WHERE department_id IN (10,20,30);

5. 加密与解密的函数

# PASSWORD()在mysql8.0中弃用。
SELECT MD5('mysql'),SHA('mysql'),MD5(MD5('mysql'))
FROM DUAL;

#ENCODE()\DECODE() 在mysql8.0中弃用。
/*
SELECT ENCODE('atguigu','mysql'),DECODE(ENCODE('atguigu','mysql'),'mysql')
FROM DUAL;
*/

6. MySQL信息函数

SELECT VERSION(),CONNECTION_ID(),DATABASE(),SCHEMA(),
USER(),CURRENT_USER(),CHARSET('软件'),COLLATION('软件')
FROM DUAL;

7. 其他函数

#如果n的值小于或者等于0，则只保留整数部分
SELECT FORMAT(123.125,2),FORMAT(123.125,0),FORMAT(123.125,-2)
FROM DUAL;

SELECT CONV(16, 10, 2), CONV(8888,10,16), CONV(NULL, 10, 2)
FROM DUAL;
#以“192.168.1.100”为例，计算方式为192乘以256的3次方，加上168乘以256的2次方，加上1乘以256，再加上100。
SELECT INET_ATON('192.168.1.100'),INET_NTOA(3232235876)
FROM DUAL;

#BENCHMARK()用于测试表达式的执行效率
SELECT BENCHMARK(100000,MD5('mysql'))
FROM DUAL;
# CONVERT():可以实现字符集的转换
SELECT CHARSET('learndb'),CHARSET(CONVERT('learndb' USING 'gbk'))
FROM DUAL;

课后练习

# 1.显示系统时间(注：日期+时间)
SELECT NOW(),SYSDATE(),CURRENT_TIMESTAMP(),LOCALTIME(),LOCALTIMESTAMP() #大家只需要掌握一个函数就可以了
FROM DUAL;

# 2.查询员工号，姓名，工资，以及工资提高百分之20%后的结果（new salary）
SELECT employee_id,last_name,salary,salary * 1.2 "new salary"
FROM employees;


# 3.将员工的姓名按首字母排序，并写出姓名的长度（length）
SELECT last_name,LENGTH(last_name) "name_length"
FROM employees
#order by last_name asc;
ORDER BY name_length ASC;

# 4.查询员工id,last_name,salary，并作为一个列输出，别名为OUT_PUT
SELECT CONCAT(employee_id,',',last_name,',',salary) "OUT_PUT"
FROM employees;

# 5.查询公司各员工工作的年数、工作的天数，并按工作年数的降序排序
SELECT employee_id,DATEDIFF(CURDATE(),hire_date)/365 "worked_years",DATEDIFF(CURDATE(),hire_date) "worked_days",
TO_DAYS(CURDATE()) - TO_DAYS(hire_date) "worked_days1"
FROM employees
ORDER BY worked_years DESC;

# 6.查询员工姓名，hire_date , department_id，满足以下条件：
#雇用时间在1997年之后，department_id 为80 或 90 或110, commission_pct不为空
SELECT last_name,hire_date,department_id
FROM employees
WHERE department_id IN (80,90,110)
AND commission_pct IS NOT NULL
#and hire_date >= '1997-01-01';  #存在着隐式转换
#and  date_format(hire_date,'%Y-%m-%d') >= '1997-01-01';  # 显式转换操作，格式化：日期---> 字符串
#and  date_format(hire_date,'%Y') >= '1997';   # 显式转换操作，格式化
AND hire_date >= STR_TO_DATE('1997-01-01','%Y-%m-%d');# 显式转换操作，解析：字符串 ----> 日期

# 7.查询公司中入职超过10000天的员工姓名、入职时间
SELECT last_name,hire_date
FROM employees
WHERE DATEDIFF(CURDATE(),hire_date) >= 10000;

# 8.做一个查询，产生下面的结果
#<last_name> earns <salary> monthly but wants <salary*3> 
SELECT CONCAT(last_name,' earns ',TRUNCATE(salary,0), ' monthly but wants ',TRUNCATE(salary * 3,0)) "Dream Salary"
FROM employees;

# 9.使用case-when，按照下面的条件：
/*job                  grade
AD_PRES              	A
ST_MAN               	B
IT_PROG              	C
SA_REP               	D
ST_CLERK             	E

产生下面的结果:
*/

SELECT last_name "Last_name",job_id "Job_id",CASE job_id WHEN 'AD_PRES' THEN 'A'
							 WHEN 'ST_MAN' THEN 'B'
							 WHEN 'IT_PROG' THEN 'C'
							 WHEN 'SA_REP' THEN 'D'
							 WHEN 'ST_CLERK' THEN 'E'
							 END "Grade"
FROM employees;

SELECT last_name "Last_name",job_id "Job_id",CASE job_id WHEN 'AD_PRES' THEN 'A'
							 WHEN 'ST_MAN' THEN 'B'
							 WHEN 'IT_PROG' THEN 'C'
							 WHEN 'SA_REP' THEN 'D'
							 WHEN 'ST_CLERK' THEN 'E'
							 ELSE "undefined" END "Grade"
FROM employees;

2、聚合函数

1. 常见的几个聚合函数

1.1 AVG / SUM ：只适用于数值类型的字段（或变量）

SELECT AVG(salary),SUM(salary),AVG(salary) * 107
FROM employees;
#如下的操作没有意义
SELECT SUM(last_name),AVG(last_name),SUM(hire_date)
FROM employees;

1.2 MAX / MIN :适用于数值类型、字符串类型、日期时间类型的字段（或变量）

SELECT MAX(salary),MIN(salary)
FROM employees;

SELECT MAX(last_name),MIN(last_name),MAX(hire_date),MIN(hire_date)
FROM employees;

1.3 COUNT：

# ① 作用：计算指定字段在查询结构中出现的个数（不包含NULL值的）

SELECT COUNT(employee_id),COUNT(salary),COUNT(2 * salary),COUNT(1),COUNT(2),COUNT(*)
FROM employees ;

SELECT *
FROM employees;

#如果计算表中有多少条记录，如何实现？
#方式1：COUNT(*)
#方式2：COUNT(1)
#方式3：COUNT(具体字段) : 不一定对！

#② 注意：计算指定字段出现的个数时，是不计算NULL值的。
SELECT COUNT(commission_pct)
FROM employees;

SELECT commission_pct
FROM employees
WHERE commission_pct IS NOT NULL;

#③ 公式：AVG = SUM / COUNT
SELECT AVG(salary),SUM(salary)/COUNT(salary),
AVG(commission_pct),SUM(commission_pct)/COUNT(commission_pct),
SUM(commission_pct) / 107
FROM employees;

#需求：查询公司中平均奖金率
#错误的！
SELECT AVG(commission_pct)
FROM employees;

#正确的：
SELECT SUM(commission_pct) / COUNT(IFNULL(commission_pct,0)),
AVG(IFNULL(commission_pct,0))
FROM employees;

#如何需要统计表中的记录数，使用COUNT(*)、COUNT(1)、COUNT(具体字段) 哪个效率更高呢？
#如果使用的是MyISAM 存储引擎，则三者效率相同，都是O(1)
#如果使用的是InnoDB 存储引擎，则三者效率：COUNT(*) = COUNT(1)> COUNT(字段)


#其他：方差、标准差、中位数

2. GROUP BY 的使用

#需求：查询各个部门的平均工资，最高工资
SELECT department_id,AVG(salary),SUM(salary)
FROM employees
GROUP BY department_id

#需求：查询各个job_id的平均工资
SELECT job_id,AVG(salary)
FROM employees
GROUP BY job_id;

#需求：查询各个department_id,job_id的平均工资
#方式1：
SELECT department_id,job_id,AVG(salary)
FROM employees
GROUP BY  department_id,job_id;
#方式2：
SELECT job_id,department_id,AVG(salary)
FROM employees
GROUP BY job_id,department_id;


#错误的！
SELECT department_id,job_id,AVG(salary)
FROM employees
GROUP BY department_id;

#结论1：SELECT中出现的非组函数的字段必须声明在GROUP BY 中。
#反之，GROUP BY中声明的字段可以不出现在SELECT中。

#结论2：GROUP BY 声明在FROM后面、WHERE后面，ORDER BY 前面、LIMIT前面

#结论3：MySQL中GROUP BY中使用WITH ROLLUP

SELECT department_id,AVG(salary)
FROM employees
GROUP BY department_id WITH ROLLUP;

#需求：查询各个部门的平均工资，按照平均工资升序排列
SELECT department_id,AVG(salary) avg_sal
FROM employees
GROUP BY department_id
ORDER BY avg_sal ASC;

#说明：当使用ROLLUP时，不能同时使用ORDER BY子句进行结果排序，即ROLLUP和ORDER BY是互相排斥的。
#错误的：
SELECT department_id,AVG(salary) avg_sal
FROM employees
GROUP BY department_id WITH ROLLUP
ORDER BY avg_sal ASC;

3. HAVING的使用 (作用：用来过滤数据的)

#练习：查询各个部门中最高工资比10000高的部门信息
#错误的写法：
SELECT department_id,MAX(salary)
FROM employees
WHERE MAX(salary) > 10000
GROUP BY department_id;


#要求1：如果过滤条件中使用了聚合函数，则必须使用HAVING来替换WHERE。否则，报错。
#要求2：HAVING 必须声明在 GROUP BY 的后面。

#正确的写法：
SELECT department_id,MAX(salary)
FROM employees
GROUP BY department_id
HAVING MAX(salary) > 10000;

#要求3：开发中，我们使用HAVING的前提是SQL中使用了GROUP BY。


#练习：查询部门id为10,20,30,40这4个部门中最高工资比10000高的部门信息
#方式1：推荐，执行效率高于方式2.
SELECT department_id,MAX(salary)
FROM employees
WHERE department_id IN (10,20,30,40)
GROUP BY department_id
HAVING MAX(salary) > 10000;

#方式2：
SELECT department_id,MAX(salary)
FROM employees
GROUP BY department_id
HAVING MAX(salary) > 10000 AND department_id IN (10,20,30,40);

#结论：当过滤条件中有聚合函数时，则此过滤条件必须声明在HAVING中。
#当过滤条件中没有聚合函数时，则此过滤条件声明在WHERE中或HAVING中都可以。但是，建议大家声明在WHERE中。

/*
  WHERE 与 HAVING 的对比
1. 从适用范围上来讲，HAVING的适用范围更广。 
2. 如果过滤条件中没有聚合函数：这种情况下，WHERE的执行效率要高于HAVING
*/

4. SQL底层执行原理

4.1 SELECT 语句的完整结构

/*

#sql92语法：
SELECT ....,....,....(存在聚合函数)
FROM ...,....,....
WHERE 多表的连接条件 AND 不包含聚合函数的过滤条件
GROUP BY ...,....
HAVING 包含聚合函数的过滤条件
ORDER BY ....,...(ASC / DESC )
LIMIT ...,....


#sql99语法：
SELECT ....,....,....(存在聚合函数)
FROM ... (LEFT / RIGHT)JOIN ....ON 多表的连接条件 
(LEFT / RIGHT)JOIN ... ON ....
WHERE 不包含聚合函数的过滤条件
GROUP BY ...,....
HAVING 包含聚合函数的过滤条件
ORDER BY ....,...(ASC / DESC )
LIMIT ...,....

*/

4.2 SQL语句的执行过程：

#FROM ...,...-> ON -> (LEFT/RIGNT  JOIN) -> WHERE -> GROUP BY -> HAVING -> SELECT -> DISTINCT -> 
# ORDER BY -> LIMIT

课后练习

#1.where子句可否使用组函数进行过滤?  No!

#2.查询公司员工工资的最大值，最小值，平均值，总和
SELECT MAX(salary) max_sal ,MIN(salary) mim_sal,AVG(salary) avg_sal,SUM(salary) sum_sal
FROM employees;

#3.查询各job_id的员工工资的最大值，最小值，平均值，总和

SELECT job_id,MAX(salary),MIN(salary),AVG(salary),SUM(salary)
FROM employees
GROUP BY job_id;

#4.选择具有各个job_id的员工人数
SELECT job_id,COUNT(*)
FROM employees
GROUP BY job_id;

#5.查询员工最高工资和最低工资的差距（DIFFERENCE）  #DATEDIFF
SELECT MAX(salary) - MIN(salary) "DIFFERENCE"
FROM employees;

#6.查询各个管理者手下员工的最低工资，其中最低工资不能低于6000，没有管理者的员工不计算在内
SELECT manager_id,MIN(salary)
FROM employees
WHERE manager_id IS NOT NULL
GROUP BY manager_id
HAVING MIN(salary) >= 6000;

#7.查询所有部门的名字，location_id，员工数量和平均工资，并按平均工资降序 
SELECT d.department_name,d.location_id,COUNT(employee_id),AVG(salary)
FROM departments d LEFT JOIN employees e
ON d.`department_id` = e.`department_id`
GROUP BY department_name,location_id

#8.查询每个工种、每个部门的部门名、工种名和最低工资 
SELECT d.department_name,e.job_id,MIN(salary)
FROM departments d LEFT JOIN employees e
ON d.`department_id` = e.`department_id`
GROUP BY department_name,job_id