Regular Expressions 101

\ban ASCII word boundary

Non-Capturing Group(?:(?:([a-z])(?=[a-z]*\1))+(?!\1))?

?repeats the group contents below at most once:

Non-Capturing Group(?:([a-z])(?=[a-z]*\1))+

+repeats the group contents below at least once:

Group 1([a-z])

Character Class[a-z]

a-zone character from a (97₁₀) to z (122₁₀) (case insensitive)

Positive Lookahead(?=[a-z]*\1)

Character Class[a-z]*

*matches any number of characters from the set below:

a-zone character from a (97₁₀) to z (122₁₀) (case insensitive)

\1the text saved by group 1

Negative Lookahead(?!\1)

\1the text saved by group 1

Group 2([a-z])

Character Class[a-z]

a-zone character from a (97₁₀) to z (122₁₀) (case insensitive)

Negative Lookahead(?![a-z]*\2)

Character Class[a-z]*

*matches any number of characters from the set below:

a-zone character from a (97₁₀) to z (122₁₀) (case insensitive)

\2the text saved by group 2

Global pattern flags

i modifier: insensitive. Matches without distinguishing uppercase and lowercase letters

g modifier: global. Finds all matches instead of stopping after the first

Detailed match information will be displayed here automatically.

Search reference

Regular Expression

\b(?:(?:([a-z])(?=[a-z]*\1))+(?!\1))?([a-z])(?![a-z]*\2)

New

'NUNUMUNN'        // expected output 'M'
'LOMING'          // expected output 'L'

I think it's possible to do this using general string operations, but my preference is really for pure regex.

AAAAAAAAAAAAAAAAAAAAAA

AXXXXXA

XXXXXXA