How are nested capturing groups numbered in regular expressions?

Question

Is there a defined behavior for how regular expressions should handle the capturing behavior of nested parentheses? More specifically, can you reasonably expect that different engines will capture the outer parentheses in the first position, and nested parentheses in subsequent positions?

Consider the following PHP code (using PCRE regular expressions)

<?php
  $test_string = 'I want to test sub patterns';
  preg_match('{(I (want) (to) test) sub (patterns)}', $test_string, $matches);
  print_r($matches);
?>

Array
(
    [0] => I want to test sub patterns  //entire pattern
    [1] => I want to test           //entire outer parenthesis
    [2] => want             //first inner
    [3] => to               //second inner
    [4] => patterns             //next parentheses set
)

The entire parenthesized expression is captured first (I want to test), and then the inner parenthesized patterns are captured next ("want" and "to"). This makes logical sense, but I could see an equally logical case being made for first capturing the sub parentheses, and THEN capturing the entire pattern.

So, is this "capture the entire thing first" defined behavior in regular expression engines, or is it going to depend on the context of the pattern and/or the behavior of the engine (PCRE being different than C#'s being different than Java's being different than etc.)?

Answer 1

From perlrequick

If the groupings in a regex are nested, $1 gets the group with the leftmost opening parenthesis, $2 the next opening parenthesis, etc.

Caveat: Excluding non-capture group opening parenthesis (?=)

Update

I don't use PCRE much, as I generally use the real thing ;), but PCRE's docs show the same as Perl's:

SUBPATTERNS

2. It sets up the subpattern as a capturing subpattern. This means that, when the whole pattern matches, that portion of the subject string that matched the subpattern is passed back to the caller via the ovector argument of pcre_exec(). Opening parentheses are counted from left to right (starting from 1) to obtain number for the capturing subpatterns.

For example, if the string "the red king" is matched against the pattern

the ((red|white) (king|queen))

the captured substrings are "red king", "red", and "king", and are numbered 1, 2, and 3, respectively.

If PCRE is drifting away from Perl regex compatibility, perhaps the acronym should be redefined--"Perl Cognate Regular Expressions", "Perl Comparable Regular Expressions" or something. Or just divest the letters of meaning.