oracle正则表达式

 正则表达式有几个优点优于常见的LIKE操作符和INSTR、SUBSTR及REPLACE 函数的。这些传统的SQL 函数不便于进行模式匹配。只有LIKE 操作符通过使用%和_字符匹配,但LIKE不支持表达式的重复、复杂的更替、字符范围、字符列表和POSIX 字符类等等。
 

元字符(Meta Character):

^            使表达式定位至一行的开头

$            使表达式定位至一行的末尾

*            匹配 0 次或更多次

?            匹配 0 次或 1 次

+            匹配 1 次或更多次

{m}          正好匹配 m 次

{m,}         至少匹配 m 次

{m, n}       至少匹配 m 次但不超过 n 次

[:alpha:]    字母字符

[:lower:]    小写字母字符

[:upper:]    大写字母字符

[:digit:]    数字

[:alnum:]    字母数字字符

[:space:]    空白字符(禁止打印),如回车符、换行符、竖直制表符和换页符[:punct:]    标点字符

[:cntrl:]    控制字符(禁止打印)

[:print:]    可打印字符 | 分隔替换选项,通常与分组操作符 () 一起使用

( )          将子表达式分组为一个替换单元、量词单元或后向引用单元

[char]       字符列表

 

Oracle 10g提供了四个regexp function: REGEXP_LIKE , REGEXP_REPLACE ,REGEXP_INSTR , REGEXP_SUBSTR 。

 

REGEXP_LIKE:比较一个字符串是否与正则表达式匹配

(srcstr, pattern [, match_option])

 

REGEXP_INSTR:在字符串中查找正则表达式,并且返回匹配的位置

(srcstr, pattern [, position [, occurrence [, return_option [, match_option]]]])

 

REGEXP_SUBSTR:返回与正则表达式匹配的子字符串

(srcstr, pattern [, position [, occurrence [, match_option]]])

 

REGEXP_REPLACE:搜索并且替换匹配的正则表达式

(srcstr, pattern [, replacestr [, position [, occurrence [, match_option]]]])

 

其中各参数的含义为:

srcstr:        被查找的字符数据。

pattern:       正则表达式。

occurrence:    出现的次数。默认为1。

position:      开始位置

return_option: 默认值为0,返回该模式的起始位置;值为1则返回符合匹配条件的下一个字符的起始位置。

replacestr:    用来替换匹配模式的字符串。

match_option:  匹配方式选项。缺省为c。

c:case sensitive

I:case insensitive

n:(.)匹配任何字符(包括newline)

m:字符串存在换行的时候被作为多行处理

 

下面通过一些具体的例子来说明如何使用这四个函数。首先创建一个测试数据表,

create table person
(  
  first_name varchar(20),  
  last_name varchar(20),  
  email varchar(100),  
  zip varchar(6)
);    

insert into person values ('Steven', 'Chen', '[email protected]', '123456');  
insert into person values ('James', 'Li', '[email protected]' || chr(10) || '[email protected]', '1b3d5f'); 

 

FIRST_NAME           LAST_NAME            EMAIL                                                                ZIP
-------------------- -------------------- ----------------------------------------------------------------- -----------
Steven               Chen                              [email protected]                                              123456
James                Li                                   [email protected]                                           ab23ef
                                                                  [email protected]                                                                                                              

REGEXP_LIKE

以数字开头

SQL> select * from person where regexp_like(zip,'^[0-9]');
 
FIRST_NAME           LAST_NAME            EMAIL                                                                            ZIP
-------------------- -------------------- -------------------------------------------------------------------------------- ------
Steven               Chen                 [email protected]                                                                    123456

 

以小写字母开头

SQL> select * from person where regexp_like(zip,'^[a-z]');
 
FIRST_NAME           LAST_NAME            EMAIL                            ZIP
-------------------- -------------------- ------------------------------              -----
James                Li                          [email protected]                ab23ef
                                                       [email protected]     

以a开头,f结尾

SQL> select * from person where regexp_like(zip,'^a.*f$');
 
FIRST_NAME           LAST_NAME            EMAIL                          ZIP
-------------------- --------------------     ------------------------------        ------
James                Li                           [email protected]              ab23ef
                                                           [email protected]   

匹配字母串

SQL> select * from person where regexp_like(first_name,'James');
 
FIRST_NAME           LAST_NAME            EMAIL                          ZIP
-------------------- -------------------- ------------------------------           ------
James                Li                        [email protected]                ab23ef
                                                        [email protected]   

匹配字母串(大小写不敏感)

SQL> select * from person where regexp_like(first_name,'james','i');
 
FIRST_NAME           LAST_NAME            EMAIL                          ZIP
-------------------- -------------------- ------------------------------            ------
James                Li                        [email protected]                ab23ef
                                                        [email protected]

匹配包含换行符字符串

SQL> select * from person where regexp_like(email,'^james.*.com$');
 
FIRST_NAME           LAST_NAME            EMAIL                          ZIP
-------------------- --------------------    ------------------------------         ------

SQL> select * from person where regexp_like(email,'^james.*.com$','n');
 
FIRST_NAME           LAST_NAME            EMAIL                          ZIP
-------------------- -------------------- ------------------------------             ------
James                Li                          [email protected]                ab23ef
                                                         [email protected]  

 

匹配换行的字符串

SQL> select * from person where regexp_like(email,'^Li.*.com$');
 
FIRST_NAME           LAST_NAME            EMAIL                          ZIP
-------------------- -------------------- ------------------------------            ------

SQL> select * from person where regexp_like(email,'^Li.*.com$','m');
 
FIRST_NAME           LAST_NAME            EMAIL                          ZIP
-------------------- -------------------- ------------------------------           ------
James                Li                            [email protected]             ab23ef
                                                           [email protected]  

注意上面分别测试了不同的match_option对结果的影响。

 

REGEXP_INSTR

查找zip中第一个数字字符的位置

SQL> select regexp_instr(zip,'[0-9]') from person;
 
REGEXP_INSTR(ZIP,'[0-9]')
-------------------------
                        1
                        3

从第三个字符开始,查找zip中第二个数字字符的位置

SQL> select regexp_instr(zip,'[0-9]',3,2) from person;
 
REGEXP_INSTR(ZIP,'[0-9]',3,2)
-----------------------------
                            4
                            4

  

REGEXP_REPLACE

把zip中所有非数字字符替换为0

SQL> update person set zip=regexp_replace(zip,'[a-z]','0') where regexp_like(zip,'[a-z]');

SQL> select * from person;
 
FIRST_NAME           LAST_NAME            EMAIL                          ZIP
-------------------- -------------------- ------------------------------           ------
Steven               Chen                   [email protected]                  123456
James                Li                        [email protected]               002300
                                                        [email protected]  

在DDL中也可以正则表达式,比如Constraint, index, view

SQL> alter table person add constraint constraint_zip check (regexp_like(zip, '^[[:digit:]]+$'));  

SQL> create index person_idx on person(regexp_substr(last_name, '^[[:upper:]]')); 

~

你可能感兴趣的:(oracle正则表达式)