Task — String reduction

You are given a word containing only alphabetic characters.

Write a function that repeatedly removes adjacent duplicate characters from the word until no adjacent duplicates remain and return the final word.

Examples

Example 1
Input: $word = 'aabbccdd'
Output: ''
Iteration 1: remove 'aa', 'bb', 'cc', 'dd' => ''

Example 2
Input: $word = 'abccba'
Output: ''
Iteration 1: remove 'cc' => 'abba'
Iteration 2: remove 'bb' => 'aa'
Iteration 3: remove 'aa' => ''

Example 3
Input: $word = 'abcdef'
Output: 'abcdef'
No duplicate found.

Example 4
Input: $word = 'aabbaeaccdd'
Output: 'aea'
Iteration 1: remove 'aa', 'bb', 'cc', 'dd' => 'aea'

Example 5
Input: $word = 'mississippi'
Output: 'm'
Iteration 1: Remove 'ss', 'ss', 'pp' => 'miiii'
Iteration 2: Remove 'ii', 'ii' => 'm'

Analysis

My solution is this:

$count = $string =~ s|(\w)\1||g while $count;

with $count initialised to 1.

It works, it's short, it's efficient but it probably fails my usual test of 'is it understandable by a non-Perl maintainer'?

It works because $string =~ s|(\w)\1||g eliminates all adjacent pairs of letters in the initial $string. But it only makes one pass, so, for example 'abba' will only be reduced to 'aa'.

It does however return (in scalar context) a $count of its deletions and if that isn't zero the while ensures that it tries again until no further pairs exist.

I am aware that it's possible to do that within the regex, but I think that makes it even more opaque.

I used \w rather than [a-z] to allow for non-Latin alphabets such as Greek or Cyrillic, and I also used lc() in the initialisation so that 'Ee' would be deleted, which might - or might not - be what Mohammad wanted.

Try it 

#!/usr/bin/perl

# Blog: http://ccgi.campbellsmiths.force9.co.uk/challenge

use v5.26;    # The Weekly Challenge - 2026-02-02
use utf8;     # Week 359 - task 2 - String reduction
use warnings; # Peter Campbell Smith
binmode STDOUT, ':utf8';
use Encode;

string_reduction('mississippi');
string_reduction('house');
string_reduction('abcdefggfedcba');
string_reduction('aaaaaaaaaaaaaaaaaa');
string_reduction('abcdeEDCBA');
string_reduction('αβγδεεδγβαω');

sub string_reduction {
    
    my ($string, $count);
    
    # initialise
    $string = lc($_[0]);
    $count = 1;
    
    # delete matched pairs
    $count = $string =~ s|(\w)\1||g while $count;
    
    say qq[\nInput:  '$_[0]'];
    say qq[Output: '$string'];
}

Output

Input:  'mississippi'
Output: 'm'

Input:  'house'
Output: 'house'

Input:  'abcdefggfedcba'
Output: ''

Input:  'aaaaaaaaaaaaaaaaaa'
Output: ''

Input:  'abcdeEDCBA'
Output: ''

Input:  'αβγδεεδγβαω'
Output: 'ω'