First of all, a ++ is in order because you shown another approach to the problem.

I guess your example might not work as expected, because the regexp is greedy, so if the XML contains more than one CDATA sections, only the first will be seen. I made slight adjustments as shown below...

#!/usr/bin/perl

use strict;
use warnings;

my $cdata = join('', <DATA>);

while ($cdata =~ s/<!\[CDATA\[(.*?)\]\]>/$1/ms)
{
    print "cdata = $1\n";
}

__DATA__
<?xml version="1.0" encoding="UTF-8" ?> 
<![CDATA[This is the first cdata]]>
<![CDATA[This is the second cdata]]>
<some/>
[download]

So that it outputs...

cdata = This is the first cdata
cdata = This is the second cdata
[download]

Also, probably hating dot star is in order, but I can't think about a better regexp right now :) .

Best regards

-lem, but some call me fokat