G'day choeppner,

In my opinion, a regex is not the right tool for this job. The string-related functions length, index and substr can provide all the functionality you need. I'd also expect them to be much faster than any regex solution (Benchmark against any you receive).

Here's my test code (with some additional test data):

#!/usr/bin/env perl

use strict;
use warnings;

use Test::More;

my @tests = (
    [qw{ABC DFGABCKBG ABC}], 
    [qw{ABC DFGAXBHCY AXBHC}],
    [qw{ABC DFGAXBHY AXB}],
    [qw{ABC DFGCBAKBG AKB}],
    [qw{ABC DFGCBAKbG A}],
    ['', 'DFGCBAKbG', ''],
    ['ABC', '', ''],
    [qw{ABC AXBHCY AXBHC}],
    [qw{ABC DFGAXBHC AXBHC}],
    [qw{abc DFGABCKBG}, ''],
);

plan tests => scalar @tests;

for my $test (@tests) {
    my ($start, $end) = (-1, -1),
    my ($rfid, $ocr, $exp) = @$test;

    for (0 .. length($rfid) - 1) {
        my $pos = index $ocr, substr($rfid, $_, 1), $end + 1;
        next if $pos == -1;
        $start = $pos if $start == -1;
        $end = $pos;
    }

    my $got = $start == -1 ? '' : substr $ocr, $start, $end - $start +
+ 1;

    is($got, $exp, "Find '$rfid' in '$ocr' as '$exp'");
}
[download]

Output:

1..10
ok 1 - Find 'ABC' in 'DFGABCKBG' as 'ABC'
ok 2 - Find 'ABC' in 'DFGAXBHCY' as 'AXBHC'
ok 3 - Find 'ABC' in 'DFGAXBHY' as 'AXB'
ok 4 - Find 'ABC' in 'DFGCBAKBG' as 'AKB'
ok 5 - Find 'ABC' in 'DFGCBAKbG' as 'A'
ok 6 - Find '' in 'DFGCBAKbG' as ''
ok 7 - Find 'ABC' in '' as ''
ok 8 - Find 'ABC' in 'AXBHCY' as 'AXBHC'
ok 9 - Find 'ABC' in 'DFGAXBHC' as 'AXBHC'
ok 10 - Find 'abc' in 'DFGABCKBG' as ''
[download]

See also: Test::More