#!/usr/bin/perl
use strict;
use warnings;
use WWW::Mechanize;
use HTML::TableContentParser;
use HTML::TokeParser::Simple;
use Data::Dumper;

use constant MODELS => 3;
use constant PICS   => 4;

-d "RESULTS" or mkdir "RESULTS", 0755 or die "cannot mkdir RESULTS: $!
+";
chdir 'RESULTS' or die "cannot chdir RESULTS: $!";

my $url = 'http://sportsillustrated.cnn.com/';
my $mech = WWW::Mechanize->new( autocheck => 1 );

$mech->get( $url . 'features/2005_swimsuit/' );
my $table = HTML::TableContentParser->new()->parse( $mech->content() )
+;

for my $row ( @{ $table->[MODELS]{rows} } ) {
    for my $cell ( @{ $row->{cells} } ) {
        my $link = get_link( $cell );
        if ( $link ) {
            $mech->get( $url . $link );
            my $t = HTML::TableContentParser->new()->parse( $mech->con
+tent() );
            for my $r ( @{ $t->[PICS]{rows} } ) {
                for my $c ( @{ $r->{cells} } ) {
                    my $p = get_pic( $c );
                    if ( $p ) {
                        my ($file) = $p =~ m|/([^/]+)$|;
                        print "Checking $file\n";
                        if (-e $file) {
                            print "skipping $file - already have\n";
                        }
                        else {
                            print "Downloading $file\n";
                            # Add error handling with $response if you
+ want
                            my $response = $mech->mirror($p, $file);
                            select(undef, undef, undef, .05);
                        }
                    }
                    else {
                        next;
                    }
                }
            }
        }
        else {
            print STDERR "No link found in ", $cell->{data}, "\n";
        }
    }
}

sub get_link {
    my $link = shift;
    my $p = HTML::TokeParser::Simple->new( \$link->{data} );
    my $token = $p->get_token;
    return $token->is_start_tag( 'a' ) ? $token->return_attr( 'href' )
+ : undef;
}

sub get_pic {
    my $link = shift;
    my $p = HTML::TokeParser::Simple->new( \$link->{data} );
    while ( my $token = $p->get_token ) {
        next if ! $token->is_start_tag( 'img' );
        my $src = $token->return_attr( 'src' );
        if ( $src =~ m|/05_[\w]+_0\dt\.jpg$|i ) {
            $src =~ s/t\.jpg$/.jpg/i;
            return $src;
        }
        else {
            return undef;
        }
    }
    return undef;
}
[download]