Can I reduce this Perl regex's atom's repetiton?

Question

^([rat])(?!\g1)([rat])(?!(\g2|\g1))[rat]$

If possible, I want to get rid of the repetition of [rat] in my regex. May be substitute it for something like \c or so.

Probably [`^(?!.*(.).*\1)[rat]{3}$`](https://regex101.com/r/pJFqCF/1), if all you need is to match a string consisting of non-repeating `r`, `a`, `t` chars. — Wiktor Stribiżew, Apr 20 '18 at 12:48
It fails the match if there is a repeating char inside the string. Does it work as you expect? — Wiktor Stribiżew, Apr 20 '18 at 12:59
It works as expected but can you please elaborate it in answer? Since I am new to assertions. — Prakash Bhattarai, Apr 20 '18 at 13:02

score 1 · Accepted Answer · answered Apr 20 '18 at 13:01

You are matching unique characters from character class. Simply saying input string should not contain a second occurrence of defined characters. Just check if the opposite happens:

([rat]).*\1

Or all in one go, validating if string follows ^[rat]{3}$ rule as well:

^(?!.*?(.).*\1)[rat]{3}$

Live demo

Breakdown:

^ Assert beginning of input string
(?! Construct a negative lookahead
- .*?(.).*\1 Look for duplicate characters
) End of negative lookahead
[rat]{3} Match a 3 char long string including defined chars
$ End of input string

Can I reduce this Perl regex's atom's repetiton?

1 Answers1