you already had nice answers and a useful thread to read, just to add my 2 cents.

bliako is right with:

> Caveat: the motto "only Perl can parse Perl" very frequently pops up in these sort of questions.

But a perl document can be parsed by PPI and pod and comments easily removed

# the entire document:

>perl -MPPI -e "print PPI::Document->new('example.pl')"

use strict;
use warnings;
# safety net loaded

my %ha = (  # dont use one letter variables
'ha' => 1,  # this stand for..
'he' => 2,  # and this other for
);
=pod

=h4

documentation

=cut
print $ha{'ha'}."\n"; # other unuseful comment
__END__


# the stripped document

>perl -MPPI -e "$doc = PPI::Document->new('example.pl'); $doc->prune('
+PPI::Token::Pod'); $doc->prune('PPI::Token::Comment'); print $doc->se
+rialize"

use strict;
use warnings;

my %ha = (
'ha' => 1,
'he' => 2,
);
print $ha{'ha'}."\n";
__END__
[download]

So you can take a big enough perl document ( 1Mb ?) with lot of comments and pod, setup a non persistent weberver (with my limited experience I mean: a server which load the content at each request) and use ab to spot differences between serving the stripped and the complete document. Have fun ;)

L*