比如有如下的sql查询语句,其中的执行顺序是什么样子的呢?
select * from help_topic as t join help_relation as r on t.help_topic_id = r.help_topic_id
where t.help_topic_id < 200
having r.help_topic_id<150
order by t.help_topic_id asc;
简单来说,上述语句是我们所写的SQL语句,这个是人理解的顺序,当然对于初学者,这个顺序是有点绕的,机器执行的顺序和我们手写SQL的顺序是不一致的!其执行顺序如下:
1. FROM <left_table>
2. ON <join_codition>
3. <join_type> JOIN <right_table>
4. WHERE <where_condition>
5. GROUP BY <group_by_list>
6. HAVING <HAVING_condition>
7. SELECT
8. DISTINCT <select_list>
9. ORDER BY <order_by_condition>
10. LIMIT <limit_number>
总体来说,mysql的查询语句,可以分为三个部分:
总体来说,MYSQL的查询语句,可以分为三个部分:
- 一个是表之间的各种连接,这部分是join ...on....形成了一个新的表(可以这么理解);
select * from help_topic as t join help_relation as r on t.help_topic_id = r.help_topic_id;
- 第二是按照条件去筛选,此处使用的是where条件,也可以理解成形成了一个新的表;
select * from help_topic as t join help_relation as r on t.help_topic_id = r.help_topic_id where t.help_topic_id < 200;
- 第三是按照上述形成的新的临时表,然后去通过分组(group by),筛选(having),排列(order by),限定行数(limit)等操作。
select * from help_topic as t join help_relation as r on t.help_topic_id = r.help_topic_id where t.help_topic_id < 200 having r.help_topic_id < 150 order by t.help_topic_id asc;
正如上述描述,SQL查询就可以分成这么三个步骤。形象的解释如下图所示: