FROM -> ON->(LEFT/RIGHT JOIN)->WHERE -> GROUP BY -> SELECT 的字段 -> HAVING -> DISTINCT -> ORDER BY -> LIMIT
SELECT ....,...,...,(存在聚合函数)
FROM ...（LEFT/RIGHT） JOIN...ON ...多表连接条件
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 ...,..
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 ...,..
MHERE子句在GROUP BY分组和聚合函数之前对数据行进行过滤；
HAVING子句在GROUP BY分组和聚合函数之后对数据行进行过滤。
SELECT dept_id, count(*)
FROM employee e
GROUP BY dept_id
HAVING count(*)>5
SELECT dept_id, count(*)
FROM employee e
WHERE salary > 10000
GROUP BY dept_id
HAVING count(*)>5
对于內连接查询，WHERE子句和ON子句等效：
对于外连接查询，ON子句在连接操作之前执行，
WHERE子句（逻辑上）在连接操作之后执行。
SELECT *
FROM department d, employee e
WHERE d.dept_id = e.dept_id
AND d.dept_name ='财务部'；
SELECT *
FROM department d
JOIN employee e ON (d.dept_id=e.dept_id AND d.dept_name ='财务部');
SELECT *
FROM department d
JOIN employee e ON (d.dept_id=e.dept_id)
WHERE d.dept_name ='财务部';