Similar questions have been asked quite a few times. (More bluntly, Try a [id://Super Search|little research] first.).

One way (probably the best) is to use the uniq function from List::MoreUtils. This is used something like this:

use List::MoreUtils;
use strict;
use warnings;
my @string = qw(a a a a b b b c d e f q q q r r r);
my @singletons = uniq(@string);

> where @singletons is a list of the elements in @string which occur only once. Note, however, that this operates on lists, not strings, so you'll have to split the string.

One A second way (guaranteed to be non-optimal &grin;) is to split the string into an array and count the occurences of each character.

use strict;use warnings;  # because it's a really good idea

my @string = split(//, 'aabbccddeeffgghhi');

my %count;

foreach(@string) {
    $count{$_} ++;
}

foreach (keys(%count)) {
    print "$_ is unique\n" if $count{$_} == 1;
}
[download]

I'm quite sure some of the regex gurus can do it with a single regex; I tend more to the "brute force and ignorance"™ school of programming than the "subtle and brilliant" school.

---

added in another update

As Hofmator pointed out, I screwed up: uniq doesn't return a list of elements that occur once in the original list; it returns a list with duplicates stripped out. That's why I've struck out a much of this post.

added in update

I knew that somebody would find a one-line solution.