Can I use Perl regular expressions to match balanced text?

Question

I would like to match text enclosed in brackets etc in Perl. How can I do that?

---

This is a question from the official perlfaq. We\'re importing the perlfaq to S

Answer 1

This is the official FAQ answer minus any subsequent edits.

Your first try should probably be the Text::Balanced module, which is in the Perl standard library since Perl 5.8. It has a variety of functions to deal with tricky text. The Regexp::Common module can also help by providing canned patterns you can use.

As of Perl 5.10, you can match balanced text with regular expressions using recursive patterns. Before Perl 5.10, you had to resort to various tricks such as using Perl code in (??{}) sequences.

Here's an example using a recursive regular expression. The goal is to capture all of the text within angle brackets, including the text in nested angle brackets. This sample text has two "major" groups: a group with one level of nesting and a group with two levels of nesting. There are five total groups in angle brackets:

I have some <brackets in <nested brackets> > and
<another group <nested once <nested twice> > >
and that's it.

The regular expression to match the balanced text uses two new (to Perl 5.10) regular expression features. These are covered in perlre and this example is a modified version of one in that documentation.

First, adding the new possessive + to any quantifier finds the longest match and does not backtrack. That's important since you want to handle any angle brackets through the recursion, not backtracking. The group [^<>]++ finds one or more non-angle brackets without backtracking.

Second, the new (?PARNO) refers to the sub-pattern in the particular capture group given by PARNO. In the following regex, the first capture group finds (and remembers) the balanced text, and you need that same pattern within the first buffer to get past the nested text. That's the recursive part. The (?1) uses the pattern in the outer capture group as an independent part of the regex.

Putting it all together, you have:

#!/usr/local/bin/perl5.10.0

my $string =<<"HERE";
I have some <brackets in <nested brackets> > and
<another group <nested once <nested twice> > >
and that's it.
HERE

my @groups = $string =~ m/
        (                   # start of capture group 1
        <                   # match an opening angle bracket
            (?:
                [^<>]++     # one or more non angle brackets, non backtracking
                  |
                (?1)        # found < or >, so recurse to capture group 1
            )*
        >                   # match a closing angle bracket
        )                   # end of capture group 1
        /xg;

$" = "\n\t";
print "Found:\n\t@groups\n";

The output shows that Perl found the two major groups:

Found:
    <brackets in <nested brackets> >
    <another group <nested once <nested twice> > >

With a little extra work, you can get the all of the groups in angle brackets even if they are in other angle brackets too. Each time you get a balanced match, remove its outer delimiter (that's the one you just matched so don't match it again) and add it to a queue of strings to process. Keep doing that until you get no matches:

#!/usr/local/bin/perl5.10.0

my @queue =<<"HERE";
I have some <brackets in <nested brackets> > and
<another group <nested once <nested twice> > >
and that's it.
HERE

my $regex = qr/
        (                   # start of bracket 1
        <                   # match an opening angle bracket
            (?:
                [^<>]++     # one or more non angle brackets, non backtracking
                  |
                (?1)        # recurse to bracket 1
            )*
        >                   # match a closing angle bracket
        )                   # end of bracket 1
        /x;

$" = "\n\t";

while( @queue )
    {
    my $string = shift @queue;

    my @groups = $string =~ m/$regex/g;
    print "Found:\n\t@groups\n\n" if @groups;

    unshift @queue, map { s/^<//; s/>$//; $_ } @groups;
    }

The output shows all of the groups. The outermost matches show up first and the nested matches so up later:

Found:
    <brackets in <nested brackets> >
    <another group <nested once <nested twice> > >

Found:
    <nested brackets>

Found:
    <nested once <nested twice> >

Found:
    <nested twice>