in reply to Trouble capturing multiple groupings in regex

Here's one way to do it. It creates a new array which contains all matches for each line of input:

use warnings;
use strict;

while (<DATA>){
    my @matches;
    while (/%([^%]+)%/g){
        push @matches, $1;
    }
    print join ' ', @matches;
    print "\n";
}

__DATA__
<span color="#231f20" textOverprint="false">%PN1%</span>
<span color="#231f20" textOverprint="false">%DIMMM%%DIMINCH%</span>


__END__
PN1
DIMMM DIMINCH

[download]

Replies are listed 'Best First'.
Re^2: Trouble capturing multiple groupings in regex
by reverendphil (Initiate) on Dec 09, 2015 at 15:07 UTC

    I've updated my post. I should've included examples of the %variable% tag being found in places where it is to be skipped, highlighting the importance of my bounding < and > in the original matching pattern. I have a main bounding area delimited by the >< chars, and a set of fields within that bounded by %'s.

[reply]

      After thinking a bit more about this, the following approach using look-around assertion works:

      use warnings;
use strict;

while (<DATA>){
    my @matches;
    @matches = (/(?<=[%>])%([^%]+)%(?=[%<])/g);
    print join ' ', @matches;
    print "\n";
}

__DATA__
<span color="#231f20" someattr="%do_not_match%" textOverprint="false">
+%PN1%</span>
<span color="#231f20" someattr="%do_not_match%" textOverprint="false">
+%DIMMM%%DIMINCH%</span>

__END__
PN1
DIMMM DIMINCH

      [download]
[reply]
[d/l]

        That's it! Fantastic! Thanks so much. I'm going to finish up this report, and then figure out what exactly you did so I can understand this, but that worked absolutely perfectly. Here's hoping I don't come up with another strange scenario when I run this on 250 templates instead of 5. :)

[reply]

In Section Seekers of Perl Wisdom