正则表达式不包括子模式

Question

I have got codes to match with a very simple string pattern: XXnnnnnnnnn (2 alphanumeric, 9 numeric) .

I am using this regex: \\w{2}\\d{9} .

Now I am required to excude from matching any string that begins with the constant token ' AY ', and any string with 11 repeated character (eg ' 11111111111 ' or ' 00000000000 ').

How may I exlude subpatterns using regular expressions?

Answer 1

Try this

\b(?!AY)(?!(\w)\1{10})\w{2}\d{9}\b

See it here on Regexr

Basically I added only the \\b word boundaries to your regex to avoid partial matches.

Your restrictions are achieved by using negative lookahead assertions.

(?!AY) The assertion fails if the pattern starts with "AY"

(?!(\\w)\\1{10}) The assertion fails if the first word character is repeated 10 more times.

Lookaround assertions on regular-expressions.info

Answer 2

You can use zero-width look-ahead matchers to exclude those two cases.

Not matching AY is easy:

 (?!AY)

Not matching 11 of the same characters is more difficult. You either need to use back references (which means the regexp isn't strictly regular anymore), or you can match explicitly to each digit.

(?!0{11}|1{11}|2{11}|…|9{11})  // 11 zeros, or 11 ones, or 11 twos, …
(?!(.)\1{10})  // A character, followed by itself 10 more times.

So that would make the full regexp:

(?!AY)(?!(.)\1{10})\w{2}\d{9}