MySQL全面瓦解之查询的正则匹配详解

概述

上一章 查询的过滤条件,我们了解了MySQL可以通过 like % 通配符来进行模糊匹配。同样的,它也支持其他正则表达式的匹配,我们在MySQL中使用 REGEXP 操作符来进行正则表达式匹配。用法和like相

似,但又强大很多,能够实现一些很特殊的、复杂的规则匹配。正则表达式使用REGEXP命令进行匹配时,如果符合返回1,不符合返回0。如果 默认不加任何匹配规则REGEXP相当于like '%%'。在前面加上NOT(NOT REGEXP)相当于NOT LIKE。

匹配模式分析

下面有个表格 ,罗列了可应用于 REGEXP 操作符中正则匹配模式,描述相对比较详细了,后面我们一个一个来测试。

匹配模式

描述

^

匹配输入字符串的开始位置。如果设置了 REGEXP 对象的 Multiline 属性,^ 也匹配 '\n' 或 '\r' 之后的位置。

$

匹配输入字符串的结束位置。如果设置了REGEXP 对象的 Multiline 属性,$ 也匹配 '\n' 或 '\r' 之前的位置。

.

匹配除 "\n" 之外的任何单个字符。要匹配包括 '\n' 在内的任何字符,请使用 '[.\n]' 的模式。

[….]

字符集合。匹配所包含的任意一个字符。例如, '[abc]' 可以匹配 "plain" 中的 'a'。

[^...]

非匹配字符集合。匹配未包含的任意字符。例如, '[^abc]' 可以匹配 "plain" 中的'p'。

[n-m]

匹配m到n之间的任意单个字符,例如[0-9],[a-z],[A-Z]

*

匹配前面的子表达式零次或多次。例如,a* 能匹配 "a" 以及 "ab"。* 等价于{0,}。

+

匹配前面的子表达式一次或多次。例如,'a+' 能匹配 "ab" 以及 "abc",但不能匹配 "a"。+ 等价于 {1,}。

?

匹配前面的子表达式一次或多次。例如,'a?' 能匹配 "ab" 以及 "a"。? 等价于 {0,1}。

a1| a2|a3

匹配 a1 或 a2 或 a3。例如,'z|food' 能匹配 "z" 或 "food"。'(z|f)ood' 则匹配 "zood" 或 "food"。

{n}

n 是一个非负整数。匹配确定的 n 次。例如,'o{2}' 不能匹配 "Bob" 中的 'o',但是能匹配 "food" 中的两个 o。

{n,}

匹配前面的子表达式n次到多次。例如,'o{2,}' 不仅能匹配 "food" ,也能匹配 "foood"。

{n,m}

n 和 m 均为非负整数,其中n <= m。最少匹配 n 次且最多匹配 m 次。

{,m}

匹配前面的子表达式0次到m次

(….)

元素组合,即将模式元素组成单一元素,例如(do)*意思是匹配0个多或多个do

匹配模式^

从字符串首部分进行匹配,这边匹配s开头的,匹配符合返回1,不符合返回0。应用到表中,既符合返回匹配到的数据。

mysql> select 'selina' REGEXP '^s';
+----------------------+
| 'selina' REGEXP '^s' |
+----------------------+
|   1 |
+----------------------+
1 row in set

mysql> select 'aelina' REGEXP '^s';
+----------------------+
| 'aelina' REGEXP '^s' |
+----------------------+
|   0 |
+----------------------+
1 row in set
mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL | 0 |
+----+--------+-----+----------+-----+
5 rows in set

mysql> select * from user2 where name REGEXP '^s';
+----+--------+-----+---------+-----+
| id | name | age | address | sex |
+----+--------+-----+---------+-----+
| 3 | sol | 21 | xiamen | 0 |
| 5 | selina | 25 | NULL | 0 |
+----+--------+-----+---------+-----+
2 rows in set

匹配模式$

从字符串尾部进行匹配,这边匹配名称以d结尾的数据。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
5 rows in set

mysql> select * from user2 where name REGEXP 'd$';
+----+-------+-----+---------+-----+
| id | name | age | address | sex |
+----+-------+-----+---------+-----+
| 1 | brand | 21 | fuzhou | 1 |
+----+-------+-----+---------+-----+
1 row in set

匹配模式.

. 是匹配任意单个字符,下面脚本匹配 n并且后面带一个任意字符的条件

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
5 rows in set

mysql> select * from user2 where name REGEXP 'n.';
+----+--------+-----+---------+-----+
| id | name | age | address | sex |
+----+--------+-----+---------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL | 0 |
+----+--------+-----+---------+-----+
3 rows in set

匹配模式[...]

指匹配括号内的任意单个字符,只要有一个字符符合条件即可。下面例子能匹配到b、w、z的 只有brand、weng 两个名称。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
5 rows in set

mysql> select * from user2 where name REGEXP [bwz];
1064 - You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near '[bwz]' at line 1
mysql> select * from user2 where name REGEXP '[bwz]';
+----+-------+-----+---------+-----+
| id | name | age | address | sex |
+----+-------+-----+---------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 4 | weng | 33 | guizhou | 1 |
+----+-------+-----+---------+-----+
2 rows in set

匹配模式[^...]

[^...]取反的意思,指匹配未包含的任意字符。例如, '[^brand]' 可以匹配 "helen" 中的'h',"sol" 的 "s","weng" 的 "w","selina" 的 "s",但无法匹配"brand",所以被过滤了。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
5 rows in set

mysql> select * from user2 where name REGEXP '[^brand]';
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
4 rows in set

匹配模式[n-m]

匹配m到n之间的任意单个字符,例如[0-9],[a-z],[A-Z],下方代码中,任何元素不在a - e之间的"sol" 被过滤了。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
5 rows in set

mysql> select * from user2 where name REGEXP '[a-e]';
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
4 rows in set

匹配模式 *

匹配前面的子表达式零次或多次。例如,a* 能匹配 "a" 以及 "ab"。* 等价于{0,}。 下面的 "e*g" 可以匹配的只有 "weng" 这个名称。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
5 rows in set

mysql> select * from user2 where name REGEXP 'e*g';
+----+------+-----+---------+-----+
| id | name | age | address | sex |
+----+------+-----+---------+-----+
| 4 | weng | 33 | guizhou | 1 |
+----+------+-----+---------+-----+
1 row in set

匹配模式 +

匹配前面的子表达式一次或多次。例如,'a+' 能匹配 "ab" 以及 "abc",但不能匹配 "a"。+ 等价于 {1,}。如下方的脚本,符合条件的是1到多个的n加上一个d的组合,只有 "brand" 和 "annd" 符合。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+--------+-----+----------+-----+
7 rows in set

mysql> select * from user2 where name REGEXP 'n+d';
+----+-------+-----+----------+-----+
| id | name | age | address | sex |
+----+-------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 7 | annd | 24 | shanghai | 1 |
+----+-------+-----+----------+-----+
2 rows in set

匹配模式 ?

匹配前面的子表达式一次或多次。例如,'a?' 能匹配 "ab" 以及 "a"。? 等价于 {0,1}。e为1个或者0个,后面再用 l 限制,所以符合的只有三个。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+--------+-----+----------+-----+
7 rows in set

mysql> select * from user2 where name REGEXP 'e?l';
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
3 rows in set

匹配模式 a1| a2|a3

匹配 a1 或 a2 或 a3。例如下方,'nn|en' 能分别匹配到 "anny" 、"annd" 和 "helen"、"weng"。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+--------+-----+----------+-----+
7 rows in set

mysql> select * from user2 where name REGEXP 'nn|en';
+----+-------+-----+----------+-----+
| id | name | age | address | sex |
+----+-------+-----+----------+-----+
| 2 | helen | 20 | quanzhou | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+-------+-----+----------+-----+
4 rows in set

匹配模式 {n} {n,} {n,m} {,m} 

 n 和 m 均为非负整数,其中n <= m。最少匹配 n 次且最多匹配 m 次。m为空代表>=n的任意数,n为空代表0。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+--------+-----+----------+-----+
7 rows in set

mysql> select * from user2 where name REGEXP 'n{2}';
+----+------+-----+----------+-----+
| id | name | age | address | sex |
+----+------+-----+----------+-----+
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+------+-----+----------+-----+
2 rows in set

mysql> select * from user2 where name REGEXP 'n{1,2}';
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+--------+-----+----------+-----+
6 rows in set

mysql> select * from user2 where name REGEXP 'l{1,}';
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 5 | selina | 25 | NULL  | 0 |
+----+--------+-----+----------+-----+
3 rows in set

匹配模式(...)

假设括号内容为abc,则是将abc作为一个整体去匹配,符合这个规则的数据被过滤出来。下面以an为例子,配合上面学过的知识。

mysql> select * from user2;
+----+--------+-----+----------+-----+
| id | name | age | address | sex |
+----+--------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 2 | helen | 20 | quanzhou | 0 |
| 3 | sol | 21 | xiamen | 0 |
| 4 | weng | 33 | guizhou | 1 |
| 5 | selina | 25 | NULL  | 0 |
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+--------+-----+----------+-----+
7 rows in set

mysql> select * from user2 where name REGEXP '(an)+';
+----+-------+-----+----------+-----+
| id | name | age | address | sex |
+----+-------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+-------+-----+----------+-----+
3 rows in set

mysql> select * from user2 where name REGEXP '(ann)+';
+----+------+-----+----------+-----+
| id | name | age | address | sex |
+----+------+-----+----------+-----+
| 6 | anny | 23 | shanghai | 0 |
| 7 | annd | 24 | shanghai | 1 |
+----+------+-----+----------+-----+
2 rows in set

mysql> select * from user2 where name REGEXP '(an).*d{1,2}';
+----+-------+-----+----------+-----+
| id | name | age | address | sex |
+----+-------+-----+----------+-----+
| 1 | brand | 21 | fuzhou | 1 |
| 7 | annd | 24 | shanghai | 1 |
+----+-------+-----+----------+-----+
2 rows in set

匹配特殊字符 \\

正则表达式语言由具有特定含义的特殊字符构成。我们已经看到.、 []、|、*、+ 等, 那我们是怎么匹配这些字符的。如下示例,我们使用 \\ 来匹配特殊字符,\\为前导, \\-表示查找-, \\.表示查找.。

mysql> select * from user3;
+----+------+-------+
| id | age | name |
+----+------+-------+
| 1 | 20 | brand |
| 2 | 22 | sol |
| 3 | 20 | helen |
| 4 | 19.5 | diny |
+----+------+-------+
4 rows in set

mysql> select * from user3 where age REGEXP '[0-9]+\\.[0-9]+';
+----+------+------+
| id | age | name |
+----+------+------+
| 4 | 19.5 | diny |
+----+------+------+
1 row in set

总结

1.当我们需要用正则匹配数据的时候,可以使用REGEXP和NOT REGEXP操作符(类似LIKE和NOT LIKE);

2.REGEXP默认不区分大小写,可以使用BINARY关键词强制区分大小写; WHERE NAME REGEXP BINARY ‘^[A-Z]';

3.REGEXP默认是部分匹配原则,即有一个匹配上则返回真。例如:SELECT  'A123' REGEXP BINARY '[A-Z]',返回的是1;

4、如果使用 () 进行匹配,则是将括号内部的内容当作整体去匹配,比如 (ABC),则需要匹配整个ABC。

5、这边只是看介绍了正则的基础知识,想要更为透彻的了解可以参考 正则教程 ,我觉得写的不错。

相关推荐