Match 6 digits in a row, no repeats in any of the digits

[dsaba]

These work for matching 6 digits in a row with none repeating, is there a shorter/better way to write it?

~(\d)(?!\1)(\d)(?!\1|\2)(\d)(?!\1|\2|\3)(\d)(?!\1|\2|\3|\4)(\d)(?!\1|\2|\3|\4|5)(\d)~
~(?\d)(?!.*\1)){6}~

 

checked against:

 

123456

 

in PCRE

[effigy]

<pre>
<?php
$tests = array(
	'1234567890',
	'1213456789',
	'1234561234',
	'1122334455',
	'0000000000',
	'ab987654cd'
);
function check_digits ($matches) {
	$unique_digits = array_fill_keys(str_split($matches[0]), 0);
	echo $matches[0], count($unique_digits) == 6 ?
		' passed.' :
		' failed.' ;
	echo '<br>';
	return $matches[0];
}
foreach ($tests as $test) {
	preg_replace_callback('/(\d{6})/', 'check_digits', $test);
}
?>
</pre>

[dsaba]

is there a shorter/better way to write it?

I thought "it" was understood to be the regex pattern. However, you responded with a non-regex solution below.

 

$unique_digits = array_fill_keys(str_split($matches[0]), 0);

I see this is the essence of your shorter/better solution. Its a great way to solve the problem, but not with regex.

 

Regardless of this still, how/why is your non-regex solution a shorter/better way to do it?

[effigy]

The ability of regular expressions to use callbacks extends their functionality. Using a pattern alone, no, I don't think there is an easier way of doing this because you cannot "expand" a back reference in a character class. In terms of benchmarks, I'm not sure that it is better; however, to me, it seems easier to understand and scale.