Efficiency in capturing groups (?: => work(?:\s+)?group VERSUS work(\s+)?group

Question

Both expressions work for me:

E1=> work(?:\s+)?group 

E2=> work(\s+)?group

I need to capture workgroup OR work group considering the space could be a line break (\s+)?

However, the first expression has a non-capture groups (?: and I am wondering if it is worse or better in the performance/fast output of the Regex. In other words, in terms of performance, what is the best expression?

Answer 1

The answer actually depends on the internals of the regex engine you are using.

In Javascript, I don't know which is faster.

In PHP, a capture group can be a bit faster. Here is a simple test with a simplified version of your regex.

<?php
$string = "WORD1".str_repeat(" someword",100000);
$regex1="~WORD1(?:\s+\w+){0,2}~";
$regex2="~WORD1(\s+\w+){0,2}~";

$start=microtime(TRUE);
for ($i=1;$i<1000000;$i++) preg_match($regex1,$string);
$noncapend=microtime(TRUE);
for ($i=1;$i<1000000;$i++) preg_match($regex2,$string);
$withcapend=microtime(TRUE);
$noncap = $noncapend-$start;
$withcap = $withcapend-$noncapend;
$diff = 100*($withcap-$noncap)/$noncap;
echo "Non-Capture Group: ".$noncap."<br />";
echo "Capture Group: ".$withcap."<br />";
echo "difference: ".$diff." percent longer<br />";

?>

The Output:

Note that you will get different results every time.

Non-Capture Group: 1.092001914978
Capture Group: 1.0608019828796
difference: -2.857131628658 percent longer