Reusing a character class in a regular expression

Question

In order to keep a regular expression more brief, is there a shorthand way to refer to a character class that occurs earlier in the same regular expression?

Example

Is there a way to shorten the following:

[acegikmoqstz@#&].*[acegikmoqstz@#&].*[acegikmoqstz@#&]

Dynamically creating the regex with a variable might be your best bet. For example in Python you could do `cc = r'[acegikmoqstz@#&]'; regex = r'{0}.*{0}.*{0}'.format(cc)`. — Andrew Clark, Nov 20 '13 at 23:37
In most programming languages, there isn't; but in languages where regexps are strings, you can get the same effect by building your regexp out of previously defined strings. — alexis, Nov 20 '13 at 23:37
@NoBrainer Java, but I'm also interested as to whether or not this can be done only with regex — Joel Christophel, Nov 20 '13 at 23:38

score 1 · Accepted Answer · edited Nov 21 '13 at 17:56

1

Keep in mind that regex features are dependant on the language being used.

With Java, you can do this:

[acegikmoqstz@#&](?:.*[acegikmoqstz@#&]){2}

But that's all, with java you can't refer to named subpattern.

With PHP you can do that:

(?(DEFINE)(?<a>[acegikmoqstz@#&]))\g<a>(?:.*\g<a>){2}

edited Nov 21 '13 at 17:56

Joel Christophel

2,604
4
30
49

answered Nov 20 '13 at 23:58

Casimir et Hippolyte

88,009
5
94
125

Reusing a character class in a regular expression

1 Answers1

Linked