File Coverage

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

Criterion	Covered	Total	%
statement	13	15	86.6
branch			n/a
condition			n/a
subroutine	5	5	100.0
pod			n/a
total	18	20	90.0

package WWW::Scraper::ISBN::Bookstore_Driver;

use strict;
use warnings;

use vars qw($VERSION @ISA);
$VERSION = '0.02';

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

=head1 NAME

WWW::Scraper::ISBN::Bookstore_Driver - Search driver for The Bookstore online book catalog.

=head1 SYNOPSIS

See parent class documentation (L)

=head1 DESCRIPTION

Searches for book information from Bookstore online book catalog

=cut

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

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

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

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

use WWW::Mechanize;

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

use constant    SEARCH  => 'http://www.bookstore.co.uk/TBP.Direct/PurchaseProduct/OrderProduct/CustomerSelectProduct/SearchProducts.aspx?d=bookstore&s=C&r=10000046&ui=0&bc=0&productGroupId=&keywordSearch=';
use constant    PRODUCT => 'http://www.bookstore.co.uk';

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

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

=head1 METHODS

=over 4

=item C

Creates a query string, then passes the appropriate form fields to the
Bookstore 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)
  pages         (if known)
  width         (if known) (in millimetres)
  height        (if known) (in millimetres)
  depth         (if known) (in millimetres)

The book_link and image_link refer back to the Bookstore 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 $self->handler("Invalid ISBN specified [$isbn]")
        unless($ean);

#print STDERR "\n# isbn=[$isbn] => ean=[$ean]\n";
    $isbn = $ean;

    my $mech = WWW::Mechanize->new();
    $mech->agent_alias( 'Linux Mozilla' );

#print STDERR "\n# url=[".(SEARCH . $isbn)."]\n";

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

    # The Results page
    my $html = $mech->content();
#print STDERR "\n# html=[\n$html\n]\n";

    my ($link) = $html =~ m!\s*

#print STDERR "\n# link=[".(PRODUCT . $link)."]\n";

    return $self->handler("Failed to find that book on The Bookstore website. [$isbn]")
        unless($link);

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

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

    return $self->handler("Failed to find that book on The Bookstore website. [$isbn]")
        if($html =~ m!Sorry, we couldn't find any matches for!si);

#print STDERR "\n# html=[\n$html\n]\n";

    my $data;

    ($data->{isbn13})       = $html =~ m!
    ($data->{isbn10})       = $html =~ m!
    ($data->{title})        = $html =~ m!
    ($data->{author})       = $html =~ m!
    ($data->{publisher})    = $html =~ m!
    ($data->{pubdate})      = $html =~ m!
    ($data->{format})       = $html =~ m!
    ($data->{description})  = $html =~ m!

    ($data->{image})        = $html =~ m!
    ($data->{thumb})        = $data->{image};

    ($data->{binding},$data->{height},$data->{width},$data->{depth},$data->{pages})
                            = $data->{format} =~ m!([^;]+);\s*H:(\d+);\s*W:(\d+);\s*D:(\d+);\s*(\d+)p\.!si;

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

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

    # trim top and tail
    foreach (keys %$data) { next unless(defined $data->{$_});$data->{$_} =~ 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},
        'width'         => $data->{width},
        'height'        => $data->{height},
        'depth'         => $data->{depth},
        'html'          => $html
    };

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

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

1;

__END__