File Coverage

blib/lib/WWW/Scraper/ISBN/Foyles_Driver.pm

Criterion	Covered	Total	%
statement	71	71	100.0
branch	9	18	50.0
condition	10	24	41.6
subroutine	7	7	100.0
pod	1	1	100.0
total	98	121	80.9

package WWW::Scraper::ISBN::Foyles_Driver;

use strict;
use warnings;

use vars qw($VERSION @ISA);
$VERSION = '0.18';

#--------------------------------------------------------------------------

=head1 NAME

WWW::Scraper::ISBN::Foyles_Driver - Search driver for the Foyles online book catalog.

=head1 SYNOPSIS

See parent class documentation (L)

=head1 DESCRIPTION

Searches for book information from the Foyles online book catalog.

=cut

#--------------------------------------------------------------------------

###########################################################################
# Inheritence

use base qw(WWW::Scraper::ISBN::Driver);

###########################################################################
# Modules

use WWW::Mechanize;

###########################################################################
# Constants

my $REFERER	    = 'http://www.foyles.co.uk';
my $FORMNAME    = 'aspnetForm';

#--------------------------------------------------------------------------

###########################################################################
# Public Interface

=head1 METHODS

=over 4

=item C

Creates a query string, then passes the appropriate form fields to the
Foyles server.

The returned page should be the correct catalog page for that ISBN. If not the
function returns zero and allows the next driver in the chain to have a go. If
a valid page is returned, the following fields are returned via the book hash:

  isbn          (now returns isbn13)
  isbn10
  isbn13
  ean13         (industry name)
  author
  title
  book_link
  image_link
  thumb_link
  description
  pubdate
  publisher
  binding       (if known)

The book_link, image_link and thumb_link all refer back to the Foyles website.

=back

=cut

sub search {
	my $self = shift;
	my $isbn = shift;
	$self->found(0);
	$self->book(undef);

    # validate and convert into EAN13 format
    my $ean = $self->convert_to_ean13($isbn);
    return $self->handler("Invalid ISBN specified")
        if(!$ean || (length $isbn == 13 && $isbn ne $ean)
                 || (length $isbn == 10 && $isbn ne $self->convert_to_isbn10($ean)));

	my $mech = WWW::Mechanize->new();
    $mech->agent_alias( 'Windows IE 6' );
    $mech->add_header( 'Accept-Encoding' => undef );

    eval { $mech->get( $REFERER ) };
    return $self->handler("The Foyles website appears to be unavailable.")
	    if($@ || !$mech->success() || !$mech->content());

    my @forms = $mech->forms;
    my %forms = map {$_->attr('name') => 1} @forms;

    return $self->handler("The Foyles website appears to be broken [".join(',',keys %forms)."].")
	    unless($forms{$FORMNAME});

    $mech->form_name( $FORMNAME );
    $mech->field( 'ctl00$txtTerm', $ean );
    $mech->field( '__EVENTTARGET', 'ctl00$LinkBtnQuickSearchBy' );
    $mech->field( '__EVENTARGUMENT', '' );

    eval { $mech->submit(); };
    return $self->handler("The Foyles website appears to be unavailable.")
	    if($@ || !$mech->success() || !$mech->content());

	# The Book page
    my $html = $mech->content();

    return $self->handler("The Foyles website appears to be unavailable.")
		if($html =~ m!I'm sorry we have encountered a problem with this page!si);

	return $self->handler("Failed to find that book on the Foyles website. [$isbn]")
		if($html =~ m!Sorry, there are no results for|This item is not currently listed on the Foyles Website!si);

    $html =~ s/&/&/g;
    $html =~ s/ / /g;
#print STDERR "\n# content2=[\n$html\n]\n";

    my $data;
    ($data->{title})        = $html =~ m!\s*([^<]+)!si;
    ($data->{author})       = $html =~ m!\s*([^<]+)!si;
    ($data->{binding})      = $html =~ m!Type: ([^<]+)!si;
    ($data->{publisher})    = $html =~ m!Publisher: ([^<]+)!si;
    ($data->{pubdate})      = $html =~ m!Publication Date: ([^<]+)!si;
    ($data->{isbn13})       = $html =~ m!ISBN-13: !si;
    ($data->{description})  = $html =~ m!([^<]+)!si;
    ($data->{image})        = $html =~ m!\s*]+>!;

    $data->{thumb}          = $data->{image};
    $data->{isbn10}         = $self->convert_to_isbn10($ean);
    $data->{isbn13}         = $ean;

    for(qw(publisher)) {
        next    unless($data->{$_});
        $data->{$_} =~ s/�?39;/'/g;
    }

#use Data::Dumper;
#print STDERR "\n# " . Dumper($data);

	return $self->handler("Could not extract data from the Foyles result page.")
		unless(defined $data);

	# trim top and tail
	foreach (keys %$data) {
        next unless(defined $data->{$_});
        $data->{$_} =~ s! ! !g;
        $data->{$_} =~ s/^\s+//;
        $data->{$_} =~ s/\s+$//;
    }

	my $bk = {
		'ean13'		    => $data->{isbn13},
		'isbn13'		=> $data->{isbn13},
		'isbn10'		=> $data->{isbn10},
		'isbn'			=> $data->{isbn13},
		'author'		=> $data->{author},
		'title'			=> $data->{title},
		'book_link'		=> $mech->uri(),
		'image_link'	=> $data->{image},
		'thumb_link'	=> $data->{thumb},
		'description'	=> $data->{description},
		'pubdate'		=> $data->{pubdate},
		'publisher'		=> $data->{publisher},
		'binding'	    => $data->{binding},
		'pages'		    => $data->{pages},
        'html'          => $html
	};

#use Data::Dumper;
#print STDERR "\n# book=".Dumper($bk);

    $self->book($bk);
	$self->found(1);
	return $self->book;
}

1;

__END__