File Coverage

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

Criterion	Covered	Total	%
statement	18	64	28.1
branch	0	12	0.0
condition	0	6	0.0
subroutine	6	7	85.7
pod	1	1	100.0
total	25	90	27.7

package WWW::Scraper::ISBN::Wheelers_Driver;

use strict;
use warnings;

use vars qw($VERSION @ISA);
$VERSION = '0.15';

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

=head1 NAME

WWW::Scraper::ISBN::Wheelers_Driver - Search drivers for the Wheelers' online book catalog.

=head1 SYNOPSIS

See parent class documentation (L)

=head1 DESCRIPTION

Searches for book information from the Wheelers' online book catalog.

=cut

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

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

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

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

use WWW::Mechanize;

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

use constant SEARCH => 'https://www.wheelers.co.nz/search/results/?query=';

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

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

=head1 METHODS

=over 4

=item C

Creates a query string, then passes the appropriate form fields to the
Wheelers 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
  description
  pubdate
  publisher
  binding       (if known)
  pages         (if known)
  weight        (if known) (in grammes)
  width         (if known) (in millimetres)
  height        (if known) (in millimetres)

The book_link and image_link refer back to the Wheelers website.

=back

=cut

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

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

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

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

    return $self->handler("Failed to find that book on Wheelers website.")
        if($html =~ m!Your search returned 0 results!si);

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

    my $data;
    ($data->{image})                 = $html =~ m!href="(.*?/large/\d+/\d+.jpg)!i;
    ($data->{thumb})                 = $html =~ m!src="(.*?/small/\d+/\d+.jpg)!i;
    ($data->{author})                = $html =~ m!\s*.*?by ([^-]+)- ISBN:[^<]+
    ($data->{title})                 = $html =~ m!\s*([^,<]+)!i;
    ($data->{publisher})             = $html =~ m!
    ($data->{pubdate})               = $html =~ m!
    ($data->{isbn13})                = $html =~ m!
    ($data->{isbn10})                = $html =~ m!
    ($data->{binding})               = $html =~ m!
    ($data->{pages})                 = $html =~ m!
    ($data->{width},$data->{height}) = $html =~ m!
    ($data->{weight})                = $html =~ m!
    ($data->{description})           = $html =~ m!Description of this book\s*([^<]+)!i;

    $data->{binding} =~ s/,.*//;
    $data->{weight}  =~ s/,//g;
    for(qw(image thumb)) {
        next unless(defined $data->{$_});
        next if($data->{$_} =~ m!^https?://!);
        $data->{$_} =~ s!^//!https://!;
    }

    # top 'n' tail trim
    for(keys %$data) {
        $data->{$_} =~ s/'/'/g;
        $data->{$_} =~ s/^\+//;
        $data->{$_} =~ s/\+$//;
    }

    my $base = $mech->uri();
    $base =~ s!(http://[^/]+).*!$1!;

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

    return $self->handler("Could not extract data from Wheelers 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},
        'weight'      => $data->{weight},
        'width'       => $data->{width},
        'height'      => $data->{height},
        'html'        => $html
    };

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

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

1;
__END__