例子文件1:https://files.cnblogs.com/files/Vera-y/myemployees.zip
例子文件2:https://files-cdn.cnblogs.com/files/Vera-y/girls.zip
连接查询:又称多表查询,当查询的字段来自多个表时,将会运用到连接查询
连接查询的分类:
1.按年代分类:
sql92标准:92年推出的标准,仅仅支持内连接
语法: select 查询列表
from 表
where 连接条件、(筛选条件)
(group by 分组)
(having 筛选条件)
(order by 排序列表)
sql99标准:99年推出的标准,MySQL中99标准支持内连接,外连接(全外连接不支持),交叉连接
语法: select 查询列表
from 表1 (连接类型)
join 表2
on 连接条件
(where筛选条件)
(group by 分组)
(having 筛选条件)
(order by 排序列表)
2.按功能分类:
内连接:等值连接,非等值连接,自连接
外连接:左外连接,右外连接,全外连接
交叉连接
spl99 标准功能分类连接类型:
内连接:inner (可以省略)
外连接:左外:left (outer)、右外:right(outer)、全外:full(outer)
交叉连接:cross
壹:内连接
一、等值连接
1.简单基础的等值查询
例1.查询员工名和对应的部门名
sql92:
SELECT last_name,department_name
FROM employees,departments
WHERE employees.department_id = departments.department_id;
# 注意两个表是通过什么连接的,这里是通过department_id连接的(又称键)
sql99:
SELECT last_name,department_name
FROM employees INNER JOIN departments
ON employees.department_id = departments.department_id;
例2.查询员工名、工种号、工种名
SELECT last_name,employees.job_id,job_title
FROM employees,jobs # 在这里可以给表起别名
WHERE employees.job_id = jobs.job_id;
(注:虽然可以起别名,但是一旦起了别名,使用到表名的地方就一定要用别名代替,否则报错)
2. 加筛选条件的等值查询
例1.查询有奖金的员工名、部门名
sql92:
SELECT last_name, department_name
FROM employees,departments
WHERE commission_pct IS NOT NULL
AND employees.department_id=departments.department_id;
sql99:
SELECT last_name,department_name
FROM employees INNER JOIN departments
ON employees.department_id=departments.department_id
WHERE commission_pct IS NOT NULL;
例2.查询城市名中第二个字符为o的部门名和城市名
SELECT department_name,city
FROM departments,locations
WHERE departments.location_id=locations.location_id
AND city LIKE '_o%';
3.添加分组
例1.查询每个城市的部门个数
sql92:
SELECT COUNT(*),city
FROM departments,locations
WHERE departments.location_id=locations.location_id
GROUP BY city;
sql99:
SELECT COUNT(*),city
FROM departments INNER JOIN locations
ON departments.location_id=locations.location_id
GROUP BY city;
例2.查询有奖金的每个部门的部门名和部门的领导编号和该部门的最低工资
SELECT department_name,MIN(salary),departments.manager_id
FROM departments,employees
WHERE departments.department_id=employees.department_id
AND commission_pct IS NOT NULL
GROUP BY department_name,departments.manager_id;
可能出现的一个错误:https://www.cnblogs.com/Vera-y/p/10923850.html
3.添加排序
例1:查询每个工种的工种名和员工的个数,并且按员工个数排降序
SELECT job_title,COUNT(*)
FROM jobs, employees
WHERE jobs.job_id=employees.job_id
GROUP BY job_title
ORDER BY COUNT(*) DESC;
例2.查询那个部门的员工个数>3的部门名和员工个数,并按个数降序
sql92:
SELECT department_name,COUNT(*)
FROM departments,employees
WHERE departments.department_id=employees.department_id
GROUP BY department_name
HAVING COUNT(*)>3
ORDER BY COUNT(*) DESC;
sql99:
SELECT department_name,COUNT(*)
FROM departments INNER JOIN employees
ON departments.department_id=employees.department_id
GROUP BY department_name
HAVING COUNT(*)>3
ORDER BY COUNT(*) DESC;
4.三表连接
例:查询员工名、部门名、和所在城市
sql92:
SELECT last_name,department_name,city
FROM employees,departments,locations
WHERE employees.department_id=departments.department_id
AND departments.location_id=locations.location_id;
sql99:
SELECT last_name,department_name,city
FROM employees
INNER JOIN departments ON employees.department_id=departments.department_id
INNER JOIN locations ON departments.location_id=locations.location_id;
(注:顺序问题:两个表一定要有连接条件才能让两个表相连)
二、非等值连接
案例:查询员工的工资和工资级别
sql92:
SELECT salary,grade_level
FROM employees,job_grades
WHERE salary BETWEEN job_grades.lowest_sal AND job_grades.highest_sal; # 这里用between and 限制了范围
sql99:
SELECT salary,grade_level
FROM employees
JOIN job_grades # 省略了inner
ON employees.salary BETWEEN job_grades.lowest_sal AND job_grades.highest_sal;
三、自连接
自个儿的表和自个儿的表连接
案例:查询姓名中包含k的员工名和上级领导的名称
sql92:
SELECT e.employee_id, e.last_name,m.employee_id,m.last_name
FROM employees AS e,employees AS m # 这里的别名用作区分为两个表
WHERE e.manager_id=m.manager_id
AND e.last_name LIKE '%k%';
sql99:
SELECT e.employee_id, e.last_name,m.employee_id,m.last_name
FROM employees AS e
JOIN employees AS m # 这里省略了inner
ON e.manager_id=m.manager_id
WHERE e.last_name LIKE '%k%';
贰:外连接
(sql99标准的外连接)
1、外连接:一个表中有,另一个表中没有的
外连接的查询结果为主表中的所有记录
若从表中有和主表匹配的,则显示匹配值
若从表中没有和主表匹配的,则用null填充
即:外连接结果=内连接结果+主表中有而从表中没有的记录
2、左外连接:left join 左边的是主表
右外连接:right join 右边的是主表
左外和右外交换两个表的顺序可以实现同样的效果
3、全外连接=内连接的结果+表1中有但表2中没有+表2中有表1中没有
4、交叉连接:笛卡尔乘积
例1.查询男朋友不在男神表的女神名 # 最终查询的为女神名来自与女神表,所以主表为女神表
左外连接:
SELECT beauty.`name`,beauty.*
FROM beauty
LEFT OUTER JOIN boys
ON beauty.boyfriend_id=boys.id
WHERE boys.id IS NULL; # 主键是不可能本身就为空的,所以只能是匹配后为空
右外连接:
SELECT beauty.`name`,beauty.*
FROM boys
RIGHT OUTER JOIN beauty
ON beauty.boyfriend_id=boys.id
WHERE boys.id IS NULL;
例2.查询那个部门没有员工 # 最终查询的是部门的,所以部门表为主表
左外连接:
SELECT departments.*, employees.employee_id
FROM departments
LEFT JOIN employees
ON employees.department_id = departments.department_id
WHERE employees.department_id IS NULL;
右外连接:
SELECT departments.*, employees.employee_id
FROM employees
RIGHT JOIN departments
ON employees.department_id = departments.department_id
WHERE employees.department_id IS NULL;
例3.查询部门名为SAL 或IT的员工信息
SELECT departments.department_name,employees.*
FROM departments
LEFT JOIN employees
ON departments.department_id = employees.department_id
WHERE departments.department_name IN('SAL','IT');
例4、交叉连接
SELECT beauty.*,boys.*
FROM beauty
CROSS JOIN boys;
有关几个查询的小统计: