Here's a possible solution: 
#!/usr/bin/perl -w

use strict;
use LWP::UserAgent;
use HTML::LinkExtor;
use URI::URL;
use HTTP::Request::Common qw(GET);

my $links = parse_page('http://www.perl.com/');

sub parse_page {
  my $url = shift;

  #Get the base URL
  my $base = url($url)->abs->base;

  my @links;

  #Push any matching links onto the @link array
  #Only put a relative link, so we don't store
  #excessive amounts of data
  my $callback = sub {
    my $tag = shift;
    my %attr = @_;

    if($tag eq 'a' && $attr{href}[0] =~ $base) {
      push @links, url($attr{href}[0], $base)->abs->rel;
    }
  };

  #Prepare the Link parser
  my $p = HTML::LinkExtor->new($callback, $base);

  #Fetch and Parse the web page
  my $ua       = LWP::UserAgent->new;
  my $response = $ua->request(GET($url), sub {$p->parse($_[0])});

  return \@links;
}

[download]

I'll explaination of what I did:

I tried to limit the amount of data that is stored into the @links array. So instead of holding the entire url in the array, then shortening it later, I just made it relative inside the callback subroutine.

Passing in the second argument to HTML::LinkExtor said to it "Add in this base url to any relative links automatically". This simplified the callback code, because every href link passed in was absolute, I didn't have to do any error checking similar to the map you used at the end of the original subroutine.

Notice how I am accessing the first element inside $attr{href}? It's an arrayref. I think this is where you may have had problems, I know I did too, at first. But liberal use of Data::Dumper showed me the error in my ways.

In order to simplify the LWP client block, I removed HTTP::Request. Instead I used a module called HTTP::Request::Common, which generally does the Right Thing, saving a few keystokes along the way.

And finally, I returned an array reference instead of a real array. When you return an array from a subroutine, you are literally copying over each element to a new place in memory. Not to mention, that when @links goes out of scope, perl's garbage collector will be invoked, and could cause a performance penalty. By passing an array reference, I have just passed the location of the @links array back to the caller, not the actual information. It's much lighter than the entire @array, and quite speedy to pass around. Your milage may vary, but I believe it's a good habit to get into.

In reply to Re: Why is this link parser only working the first time? by dkubb
in thread Why is this link parser only working the first time? by dave_aiello

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.