Here's the solution using XML::Rules:

#!/usr/bin/perl
use warnings;
use strict;

use XML::Rules;
use Data::Dumper;

my $xml = do { local $/; <DATA>};

my $parser = XML::Rules->new(
    rules => [
        _default => 'content',
        page => 'no content array',
        story => 'pass no content',
    ]
);

my $data = $parser->parse($xml);
print Dumper $data->{page};

__DATA__
<story>
           <page>
                <author>Author 1 </author>
                <keywords>Keyword1 </keywords>
                <headline>Headline1 </headline>
                <image> Image1 </image>
                <description>Desc 1 </description>
            </page>
            <page>
                <author>Author 2</author>
                <keywords>Keyword2 </keywords>
                <headline>Headline 2 </headline>
               <image> Image2 </image>
               <description>Decs 2 </description>
            </page>
</story>
[download]

If you wanted to process each <page> separately, as the file is parsed, you could do something like this:

#!/usr/bin/perl
use warnings;
use strict;

use XML::Rules;
use Data::Dumper;

my $xml = do { local $/; <DATA>};

my $parser = XML::Rules->new(
    rules => [
        _default => 'content trim',
        page => sub {
            print <<"*END*"
"$_[1]->{headline}" by $_[1]->{author}
Keywords: $_[1]->{keywords}
Description: $_[1]->{description}

*END*
        },
        story => '',
    ]
);

$parser->parse($xml);

__DATA__
...
[download]