摘自:html
http://www.educity.cn/wenda/590849.htmlmysql
http://blog.csdn.net/hguisu/article/details/7106159sql
問:oracle
不是說,一條sql語句只能用一個索引麼post
但SELECT * FROM `comment` WHERE `toconuid` = '10' or `tocomuid` = '10'ui
其中 toconuid列 和 tocomuid列 分別爲單列索引spa
explain後 顯示兩個索引都用了,extra爲 Using union(toconuid,tocomuid); Using where.net
答:htm
凡事總有特列。blog
而MYSQL能夠理解爲把這個語句拆成了兩條語句SELECT * FROM `comment` WHERE `toconuid` = '10'unionSELECT * FROM `comment` WHERE `tocomuid` = '10'
在某些狀況下,or條件能夠避免全表掃描的。
1)myisam表:
CREATE TABLE IF NOT EXISTS `a` (
`id` int(1) NOT NULL AUTO_INCREMENT,
`uid` int(11) NOT NULL,
`aNum` char(20) DEFAULT NULL,
PRIMARY KEY (`id`),
KEY `uid` (`uid`)
) ENGINE=MyISAM DEFAULT CHARSET=latin1 AUTO_INCREMENT=6 ;
mysql> explain select * from a where id=1 or uid =2;
+----+-------------+-------+-------------+---------------+-------------+---------+------+------+---------------------------------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+-------------+---------------+-------------+---------+------+------+---------------------------------------+
| 1 | SIMPLE | a | index_merge | PRIMARY,uid | PRIMARY,uid | 4,4 | NULL | 2 | Using union(PRIMARY,uid); Using where |
+----+-------------+-------+-------------+---------------+-------------+---------+------+------+---------------------------------------+
1 row in set (0.00 sec)
2)innodb表:
CREATE TABLE IF NOT EXISTS `a` (
`id` int(1) NOT NULL AUTO_INCREMENT,
`uid` int(11) NOT NULL,
`aNum` char(20) DEFAULT NULL,
PRIMARY KEY (`id`),
KEY `uid` (`uid`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1 AUTO_INCREMENT=6 ;
mysql> explain select * from a where id=1 or uid =2;
+----+-------------+-------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | a | ALL | PRIMARY,uid | NULL | NULL | NULL | 5 | Using where |
+----+-------------+-------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
+-------+----------------------------------------------------------------------------------------------------------------------
| Table | Create Table
+-------+----------------------------------------------------------------------------------------------------------------------
| a | CREATE TABLE `a` (
`id` int(1) NOT NULL AUTO_INCREMENT,
`uid` int(11) NOT NULL,
`aNum` char(20) DEFAULT NULL,
PRIMARY KEY (`id`)
) ENGINE=MyISAM AUTO_INCREMENT=6 DEFAULT CHARSET=latin1 |
+-------+----------------------------------------------------------------------------------------------------------------------
1 row in set (0.00 sec)
explain查看:
mysql> explain select * from a where id=1 or uid =2;
+----+-------------+-------+------+---------------+------+---------+------+------+-------------+
| id | select_type | table | type | possible_keys | key | key_len | ref | rows | Extra |
+----+-------------+-------+------+---------------+------+---------+------+------+-------------+
| 1 | SIMPLE | a | ALL | PRIMARY | NULL | NULL | NULL | 5 | Using where |
+----+-------------+-------+------+---------------+------+---------+------+------+-------------+
1 row in set (0.00 sec)
全表掃描了。
一般狀況下, 用UNION替換WHERE子句中的OR將會起到較好的效果. 對索引列使用OR將形成全表掃描.
注意, 以上規則只針對多個索引列有效. 若是有column沒有被索引, 查詢效率可能會由於你沒有選擇OR而下降.
在下面的例子中, LOC_ID 和REGION上都建有索引.
高效:
低效:
若是你堅持要用OR, 那就須要返回記錄最少的索引列寫在最前面.
這是一條簡單易記的規則,可是實際的執行效果還須檢驗,在oracle8i下,二者的執行路徑彷佛是相同的.
低效:
select…. from location where loc_id = 10 or loc_id = 20 or loc_id = 30
高效
select… from location where loc_in in (10,20,30);