【背景】

　　某某同學執行了一下Explain結果結果發現數據庫有了一條寫入操作，恭喜這位同學你的鍋到貨了，你簽收一下；

　　對！ 你沒有聽錯，在一種場景下就算是Explain也會引發數據的寫操作，就這是外層查詢訪問任意表，內層查詢調用function

　　在function有寫入動作的情況下會發生寫入。

【硬生生的套上一個場景】

　　假設我們有一個Person表，每訪問一次Person表都記錄一次在什麽時候，訪問了哪一行，表結構設計如下

create table if not exists person(
    id int not null auto_increment primary
 
 key,
    name varchar(16)
);

create table if not exists person_opration_log(
    id int not null auto_increment primary key,
    pid int not null,
    access_datetime datetime
);

delimiter //
create function fun_person_log(pid int) returns int
    begin
        insert into person_opration_log(pid,access_datetime) values
 
(pid,now());
        return pid;
    end //

delimiter ;

　　正常的數據訪問SQL如下，但是它並不寫日誌

mysql> select
    ->     id,
    ->     name 
    -> from person 
    -> where id = 1;
+----+--------+
| id | name   |
+----+--------+
|  1 | 項羽   |
+----+--------+
1 row in set (0.00 sec)

　　如果我們要寫日誌可以分兩步走，先訪問再計一筆日誌

mysql> select
    ->     id,
    ->     name 
    -> from person 
    -> where id = 1;
+----+--------+
| id | name   |
+----+--------+
|  1 | 項羽   |
+----+--------+
1 row in set (0.00 sec)

mysql> 
mysql> select fun_person_log(1);
+-------------------+
| fun_person_log(1) |
+-------------------+
|                 1 |
+-------------------+
1 row in set (0.05 sec)

mysql> 
mysql> select * from person_opration_log ;
+----+-----+---------------------+
| id | pid | access_datetime     |
+----+-----+---------------------+
|  1 |   1 | 2018-10-06 17:12:31 |
+----+-----+---------------------+
1 row in set (0.00 sec)

【牛人想出的新點子把兩步合成一步】

　　牛人的新點子

mysql> select 
    ->     fun_person_log(100) as id , 
    ->     name
    -> from person
    -> where id = (select fun_person_log(100));
Empty set (0.04 sec)

mysql> 
mysql> select * from person_opration_log;
+----+-----+---------------------+
| id | pid | access_datetime     |
+----+-----+---------------------+
|  1 |   1 | 2018-10-06 17:12:31 |
|  2 | 100 | 2018-10-06 17:15:29 |
+----+-----+--------------------

　　牛人的新點子剛好入坑，我們可以explain一下

mysql> explain select 
    ->     fun_person_log(250) as id , 
    ->     name
    -> from person
    -> where id = (select fun_person_log(250));
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+--------------------------------+
| id | select_type | table | partitions | type | possible_keys | key  | key_len | ref  | rows | filtered | Extra                          |
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+--------------------------------+
|  1 | PRIMARY     | NULL  | NULL       | NULL | NULL          | NULL | NULL    | NULL | NULL |     NULL | no matching row in const table |
|  2 | SUBQUERY    | NULL  | NULL       | NULL | NULL          | NULL | NULL    | NULL | NULL |     NULL | No tables used                 |
+----+-------------+-------+------------+------+---------------+------+---------+------+------+----------+--------------------------------+
2 rows in set, 1 warning (0.08 sec)

mysql> 
mysql> select * from person_opration_log;
+----+-----+---------------------+
| id | pid | access_datetime     |
+----+-----+---------------------+
|  1 |   1 | 2018-10-06 17:12:31 |
|  2 | 100 | 2018-10-06 17:15:29 |
|  3 | 250 | 2018-10-06 17:17:23 |
+----+-----+---------------------+
3 rows in set (0.00 sec)

　　看吧！ explain引發了寫入操作！

【參考連接】

Derived Tables

【學習交流】

-----------------------------http://www.sqlpy.com-------------------------------------------------

-----------------------------http://www.sqlpy.com-------------------------------------------------

MySQL通過Explain查看select語句的執行計劃結果觸發寫操作