今日sql学习 - 技术栈

[一,union all](#一,union all)

二，GROUP_CONCAT函数

三，字符串函数：RIGHT()函数

[四，字符串函数：LENGTH() 函数 / REPLACE（）函数](#四，字符串函数：LENGTH() 函数 / REPLACE（）函数)

一,union all

对查询结果进行连接与 union 不同 union 不会去重

sql 复制代码

SELECT 
AVG(salary)
FROM `salaries8` 
WHERE to_date = '9999-01-01' AND salary NOT IN (

  SELECT MAX(salary) FROM `salaries8` AND to_date = '9999-01-01'
  UNION ALL 
  SELECT MIN(salary) FROM `salaries8` AND to_date = '9999-01-01'


)

二，GROUP_CONCAT函数

可以将分组后的相同字段连接

sql 复制代码

select
    dept_no,
    group_concat(emp_no) employees
from dept_emp
group by dept_no;

三，字符串函数：RIGHT()函数

从右向左获取字段的值可以指定获取的个数

sql 复制代码

SELECT
first_name
FROM `employees`
ORDER BY RIGHT(first_name,2)

四，字符串函数：LENGTH() 函数 / REPLACE（）函数

LENGTH()函数用来获取字符串的长度 / replace（字段，"被替换值"，"替换的值"）函数可以替换值

sql 复制代码

SELECT id, (LENGTH(string) - LENGTH(REPLACE(string,",",''))) AS cnt  FROM `strings`

sql 复制代码

update titles_test
set titles_test.emp_no = replace(titles_test.emp_no, 10001, 10005)
where titles_test.id = '5';

五，条件更新表内的值

UPDATE "table_name" SET `字段` = "value" WHERE

sql 复制代码

-- 更新数据库
UPDATE `salaries6` SET salary = 1.1 * salary
WHERE salaries6.emp_no IN (

SELECT emp_no FROM `emp_bonus`

) AND salaries6.to_date = '9999-01-01'

update titles_test set to_date = NULL, from_date = "2001-01-01"

where id in

(

select * from (

select id

from titles_test

where to_date = "9999-01-01"

) as t

)

六，创建外键

sql 复制代码

ALTER TABLE audit ADD FOREIGN KEY (emp_no) REFERENCES employees_test(ID);

七，修改表名

sql 复制代码

ALTER TABLE `titles_test` RENAME AS `titles_2017`

八，删除记录

sql 复制代码

delete from titles_test
where id not in(
select min_id from (select min(id) as min_id from titles_test group by emp_no)t1);

九，构建触发器

sql 复制代码

create trigger audit_log
after insert on employees_test
for each row
begin
    insert into audit values(new.id,new.name);
end

十，增加属性字段

sql 复制代码

ALTER TABLE actor ADD DELETE_date DATETIME NOT NULL DEFAULT '0000 00:00:00'