【原创】MySQL 返回更新值（RETURNING）

在写SQL中，经常会有诸如更新了一行记录，之后要获取更新过的这一行。 本身从程序来说，没啥难度，大不了把这行缓存起来，完了直接访问。 但是从数据库的角度出发，怎么能快速的拿出来，而又不对原表进行二次扫描？ 比如其他数据库提供了如下的语法来实现：


返回更新掉的行：

t_girl=# update t1 set log_time = now() where id in (1,2,3) returning *;
 id | log_time
----+----------------------------
  1 | 2014-11-26 11:06:53.555217
  2 | 2014-11-26 11:06:53.555217
  3 | 2014-11-26 11:06:53.555217
(3 rows)


UPDATE 3
Time: 6.991 ms

返回删除掉的行：

t_girl=# delete from t1 where id < 2 returning *;
 id | log_time
----+----------------------------
  1 | 2014-11-26 11:06:53.555217
(1 row)


DELETE 1
Time: 6.042 ms

返回插入后的行：

t_girl=# insert into t1 select 1,now() returning *;
 id | log_time
----+----------------------------
  1 | 2014-11-26 11:07:40.431766
(1 row)


INSERT 0 1
Time: 6.107 ms
t_girl=#

那在MySQL里如何实现呢？ 
我可以创建几张内存表来来保存这些返回值，如下：

CREATE TABLE t1_insert ENGINE MEMORY SELECT * FROM t1 WHERE FALSE;
CREATE TABLE t1_update ENGINE MEMORY SELECT * FROM t1 WHERE FALSE;
CREATE TABLE t1_delete ENGINE MEMORY SELECT * FROM t1 WHERE FALSE;


ALTER TABLE t1_insert ADD PRIMARY KEY (id);
ALTER TABLE t1_update ADD PRIMARY KEY (id);
ALTER TABLE t1_delete ADD PRIMARY KEY (id);

以上建立了三张表来存放对应的操作。 t1_insert 保存插入；t1_update 保存更新；t1_delete 保存删除。


那这样的话，我来创建对应的触发器完成。

DELIMITER $$


USE `t_girl`$$


DROP TRIGGER /*!50032 IF EXISTS */ `tr_t1_insert_after`$$


CREATE
    /*!50017 DEFINER = 'root'@'localhost' */
    TRIGGER `tr_t1_insert_after` AFTER INSERT ON `t1`
    FOR EACH ROW BEGIN
      REPLACE INTO t1_insert VALUES (new.id,new.log_time);
    END;
$$


DELIMITER ;




DELIMITER $$


USE `t_girl`$$


DROP TRIGGER /*!50032 IF EXISTS */ `tr_t1_update_after`$$


CREATE
    /*!50017 DEFINER = 'root'@'localhost' */
    TRIGGER `tr_t1_update_after` AFTER UPDATE ON `t1`
    FOR EACH ROW BEGIN
      REPLACE INTO t1_update VALUES (new.id,new.log_time);
    END;
$$


DELIMITER ;




DELIMITER $$


USE `t_girl`$$


DROP TRIGGER /*!50032 IF EXISTS */ `tr_t1_delete_after`$$


CREATE
    /*!50017 DEFINER = 'root'@'localhost' */
    TRIGGER `tr_t1_delete_after` AFTER DELETE ON `t1`
    FOR EACH ROW BEGIN
      REPLACE INTO t1_delete VALUES (old.id,old.log_time);；
    END;
$$


DELIMITER ;

创建好了以上的表和触发器后， 拿到返回值就非常容易了， 我直接从以上几张表来查询就是。


我现在来演示：
更新：

mysql> truncate table t1_update;
Query OK, 0 rows affected (0.00 sec)


mysql> UPDATE t1 SET log_time = NOW() WHERE id < 15;
Query OK, 3 rows affected (0.01 sec)
Rows matched: 3  Changed: 3  Warnings: 0

获取更新记录：

mysql> select * from t1_update;
+----+----------------------------+
| id | log_time |
+----+----------------------------+
| 12 | 2014-11-26 13:38:06.000000 |
| 13 | 2014-11-26 13:38:06.000000 |
| 14 | 2014-11-26 13:38:06.000000 |
+----+----------------------------+
3 rows in set (0.00 sec)

插入：

mysql> truncate table t1_insert;
Query OK, 0 rows affected (0.00 sec)


mysql> INSERT INTO t1 VALUES (1,NOW());
Query OK, 1 row affected (0.08 sec)

获取插入记录：

mysql> select * from t1_insert;
+----+----------------------------+
| id | log_time |
+----+----------------------------+
| 1 | 2014-11-26 13:38:06.000000 |
+----+----------------------------+
1 row in set (0.00 sec)

删除：

mysql> truncate table t1_delete;
Query OK, 0 rows affected (0.00 sec)


mysql> DELETE FROM t1 WHERE id < 15;
Query OK, 4 rows affected (0.01 sec)

获取删除记录：

mysql> select * from t1_delete;
+----+----------------------------+
| id | log_time |
+----+----------------------------+
| 1 | 2014-11-26 13:38:06.000000 |
| 12 | 2014-11-26 13:38:06.000000 |
| 13 | 2014-11-26 13:38:06.000000 |
| 14 | 2014-11-26 13:38:06.000000 |
+----+----------------------------+
4 rows in set (0.00 sec)