#!/usr/bin/perl -w
#######################
# findlinks4
#
# If one really wanted to extract parts 
# of an html document, though, there's no
# reason to re-invent the wheel.
#
# See, for example, HTML::LinkExtor, which is
# part of the HTML::Parser package.
# Documentation and examples may be found on the CPAN,
# for example at
#   http://theoryx5.uwinnipeg.ca/CPAN/data/HTML-Parser/HTML/LinkExtor.html
#
########################
use strict;

# Complain to the user if not called properly.
unless (@ARGV==1) {
  die " Wrong number of arguments.  Usage: 'findfile file.html' \n";
}

# Tell the nice user what we're up to.
my $filename = $ARGV[0];
print "Analyzing '$filename' for links...\n";

#  -- Now let the professionals do the work -- 

# Extract all this links.
use HTML::LinkExtor;			# Import a package.
my $parser = HTML::LinkExtor->new;	# Create an instance of an object.
$parser->parse_file($filename);		# Call one of that object's methods.
my @results = $parser->links;		# Call another method.

# Print the data structure using the Dumper.
use Data::Dumper;
print Dumper(@results);

#   or
#
## Print the data structure @results = ( $link1, $link2, $link3, ...)
## where $link = [ $tag, $attr1 => $uri1, $att2 => $uri2, ... ]
##
#foreach my $link (@results) {
#  my ($tag, %stuff) = @$link;
#  print " tag: '$tag' ";
#  foreach my $key (keys %stuff) {
#    print ", $key => " . $stuff{$key}
#  }
#  print "\n";
#}