Oracle 正则表达式详解(regexp | 您所在的位置:网站首页 › place和replace的关系 › Oracle 正则表达式详解(regexp |
文章目录 1 概述 2 匹配规则 3 正则函数 3.1 regexp_substr() 3.2 regexp_instr() 3.3 regexp_replace() 3.4 regexp_like() 1 概述 1. 作用:处理字符时,很强大 1 2. 分类:与下列相似,但功能更加强大(‘支持正则表达式’) (1) regexp_like : 同 like 功能相似(模糊 ‘匹配’) (2) regexp_instr : 同 instr 功能相似(返回字符所在 ‘下标’) (3) regexp_substr : 同 substr 功能相似(‘截取’ 字符串) (4) regexp_replace: 同 replace 功能相似( ‘替换’ 字符串) – 使用了 ‘正则表达式’ 替代了老的 ‘百分比 %’ 和 ‘通配符 _’ – 上述 四个 函数,正则用法相似,知晓一个,其它皆可参考 2 匹配规则 主要的类型如下: 分组 元字符 描述 特殊的 \ 转义字符。“\n”:换行,"\\n":文本 ‘\n’ . 匹配除转义字符外的 任何单字符 定位元字符 ^ 使表达式定位至一行的 开头 $ 使表达式定位至一行的 末尾 量词或重复操作符 * 匹配 0 次或多次(任意次数) ? 匹配 0 次或 1 次(最多 1 次) + 匹配 1 次或多次(至少 1 次) {m} 正好 匹配 m 次 {m, } 至少 匹配 m 次 {m, n} 匹配 m 到 n 次 表达式的替换和分组 | 替换,通常和分组操作符 () 一起使用 () 分组,并且 的关系 [char] 字符列表,或者 的关系 预定义的 posix 字符类 [:alpah:] 任何字母 [:lower:] 任何小写字母 [:upper:] 任何大写字母 [:digit:] 任何数字,相当于 [0-9] [:xdigit:] 任何 16 进制的数字,相当于 [0-9a-fA-F] [:alnum:] 任何字符或数字字符 [:space:] 空白字符,如:回车符,换行符、竖直制表符和换页符 [:punct:] 任何标点符号 示例:此案例来源于网络,略作修改,挺适合参考的 1. 查询 value 中以 1 开头 60 结尾的记录并且长度是 7 位 select * from table_name where value like '1____60'; -- 4 个 _ select * from table_name where regexp_like(value, '1....60'); 2. 查询 value 中以 1 开头 60 结尾的记录并且长度是 7 位且全部是数字的记录 此时,用 like 就不是很好实现了 select * from table_name where regexp_like(value, '^1[0-9]{4}60$'); select * from table_name where regexp_like(value, '^1[[:digit:]]{4}60$'); 3. 查询 value 中不包含任何数字的记录 select * from table_name where regexp_like(value, '^[^[:digit:]]+$'); -- [] 内的 ^ 表示 '非' 4. 查询 value 中不是纯数字的记录 select * from table_name where not regexp_like(value, '^[[:digit:]]+$'); 5. 查询以 12 或 1b 开头的记录,区分大小写 select * from table_name where regexp_like(value, '^(12)|^(1b)'); select * from table_name where regexp_like(value, '^1[2b]'); 6. 查询以 12 或 1b 开头的记录,不区分大小写 select * from table_name where regexp_like(value, '^1[2b]', 'i'); 7. 查询所有均为 小写字母 或 数字的记录 select * from table_name where regexp_like(value, '^([a-z]+|[0-9]+)$'); 8. 查询包含 空白 的记录 select * from table_name where regexp_like(value, '[[:space:]]'); 9. 查询包含 标点符号 的记录 select * from table_name where regexp_like(value, '[[:punct:]]'); 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 3 正则函数 扩展:Oracle 官方文档 - regexp_substr 3.1 regexp_substr() 参数解释: srcstr : 源字符串 -- source string pattern : 正则表达式 position : 起始位置,默认 1 occurrence : 第几次出现,默认 1(匹配成功的次数,依次递增) modifier : 模式('i': 不区分大小写,'c': 区分大小写。 默认 'c') subexpression: 含有子表达式 0-9, 默认 0:不含子表达式,1:第一个子表达式,以此类推 1 2 3 4 5 6 例1: select regexp_substr('abc,CBA121ABC,cba', ',[^,]+,') rs1, -- ,CBA121ABC, regexp_substr('1234567890', '(123)(4(56)(78))', 1, 1, 'i', 1) rs2 -- 123 from dual; 1 2 3 例2:拆分邮箱 with temp_email as ( select 1 user_no, '[email protected]' email from dual union all select 2 user_no, '[email protected]' email from dual union all select 3 user_no, '[email protected]' email from dual ) select t.user_no 用户编号, regexp_substr(t.email,'[[:alnum:]]+') 用户名, regexp_substr(t.email, '\@[[:alnum:]]+\.[[:alnum:]]+') 邮箱后缀 -- 两者等同 -- regexp_substr(t.email,'[a-zA-Z0-9]+') 用户名2, -- regexp_substr(t.email, '\@[a-zA-Z0-9]+\.[a-zA-Z0-9]+') 邮箱后缀2 from temp_email t; 1 2 3 4 5 6 7 8 9 10 11 12 测试结果: 用户编号 用户名 邮箱后缀 1 a1 @qq.com 2 bb2 @sina.com 3 ccc3 @aliyun.com 1 2 3 4 例3:匹配固定电话号码和手机号码 --************************************************************* -- 固定电话格式:0开头 + 2到3位任意数字 + '-' + 6到7位任意数字 -- 手机号码格式:1开头 + 10位任意数字(11位) --************************************************************* with temp_phone as ( select 1 user_no, '0755-11112222' phone_no from dual union all select 2 user_no, '055-33334444' phone_no from dual union all select 3 user_no, '0755 55556666' phone_no from dual union all select 4 user_no, 'AE86-77778888' phone_no from dual union all select 5 user_no, '0755-888999' phone_no from dual union all select 6 user_no, '15866668888' phone_no from dual union all select 7 user_no, '015866668888' phone_no from dual ) select t.user_no 用户编号, t.phone_no 联系方式, regexp_substr(t.phone_no,'^0[0-9]{2,3}-[0-9]{6,7}') 固定电话, regexp_substr(t.phone_no,'^1[0-9]{10}') 手机号码 from temp_phone t; 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 测试结果: 3.2 regexp_instr() 参数解释: srcstr : 源字符串 -- source string pattern : 正则表达式 position : 起始位置,默认 1 occurrence : 第几次出现,默认 1(匹配成功的次数,依次递增) returnparam : 返回参数对应的下标(0:第一个位置的下标,非0:最后一个位置的下标) modifier : 模式('i': 不区分大小写,'c': 区分大小写。 默认 'c') subexpression: 含有子表达式 0-9, 默认 0:不含子表达式,1:第一个子表达式,以此类推 1 2 3 4 5 6 7 示例: select regexp_instr('1,23,456,7890', '[0-9]+', 1, 3) 匹配到的第一个字符下标, -- 6 regexp_instr('1,23,456,7890', '[0-9]+', 1, 3, 1) 最后一个字符后一位的下标, -- 9 regexp_instr('1234567890', '(123)(4)(56)(78)', 1, 1, 0, 'i', 4) 匹配到的第四个子表达式 -- 7 from dual; 1 2 3 4 查询结果: 匹配到的第一个字符下标 最后一个字符后一位的下标 匹配到的第四个子表达式 6 9 7 1 2 3.3 regexp_replace() 示例: with temp_strings as ( select 'abc123' str from dual union all select '123abc' str from dual union all select 'a1b2c3' str from dual ) select t.str 源字符串, regexp_replace(t.str, '[0-9]', '', 1) 无数字字符串 from temp_strings t;查询结果: 源字符串 无数字字符串 1 abc123 abc 2 123abc abc 3 a1b2c3 abc示例: with temp_strings as ( select 'abc123' str from dual union all select '12abcd' str from dual union all select 'a1b2c3' str from dual ) select t.str 连续的三个数字字符 from temp_strings t where regexp_like(t.str, '[0-9]{3}');查询结果: 连续的三个数字字符 abc123 |
CopyRight 2018-2019 实验室设备网 版权所有 |