File Coverage

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

Criterion	Covered	Total	%
statement	31	44	70.4
branch	0	4	0.0
condition			n/a
subroutine	9	9	100.0
pod	1	1	100.0
total	41	58	70.6

# ex:ts=8

package WWW::Scraper::ISBN::TWTenlong_Driver;

use strict;
use warnings;

use vars qw($VERSION @ISA);
$VERSION = '0.01';

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

=head1 NAME

WWW::Scraper::ISBN::TWTenlong_Driver - Search driver for TWTenlong's online catalog.

=head1 SYNOPSIS

See parent class documentation (L)

=head1 DESCRIPTION

Searches for book information from the TWTenlong's online catalog.

=cut

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

###########################################################################
#Library Modules                                                          #
###########################################################################

use WWW::Scraper::ISBN::Driver;
use WWW::Mechanize;
use Template::Extract;

use Data::Dumper;

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

use constant	TENLONG	=> 'http://www.tenlong.com.tw';

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

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

@ISA = qw(WWW::Scraper::ISBN::Driver);

###########################################################################
#Interface Functions                                                      #
###########################################################################

=head1 METHODS

=over 4

=item C

Creates a query string, then passes the appropriate form fields to the Tenlong
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
  title
  author
  pages
  book_link
  image_link
  pubdate
  publisher
  price_list
  price_sell

The book_link and image_link refer back to the Tenlong website.

=back

=cut

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

	my $mechanize = WWW::Mechanize->new();
	$mechanize->get(TENLONG);

	$mechanize->submit_form(
		form_name	=> 'BookSearchForm',
		fields		=> {
			fKeyword	=> $isbn,
		},
	);

	# The Search Results page
	my $template = <
ÃöÁä¦r¬d¸ßµ²ªG[% ... %]
END

	my $extract = Template::Extract->new;
	my $data = $extract->extract($template, $mechanize->content());

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

	my $book = $data->{book};
	$mechanize->get($book);

	$template = <
[% ... %]
[% title %][% ... %]
 by [% author %]

ISBN :[% ... %] [% isbn %]
¥Xª©°Ó :[% ... %] [% publisher %]
¥Xª©¤é´Á :[% ... %] [% pubdate %]
­¶¼Æ :[% ... %] [% pages %]
©w»ù :[% ... %]">[% price_list %][% ... %]
°â»ù :[% ... %]">[% price_sell %]
END

	$data = $extract->extract($template, $mechanize->content());

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

	my $bk = {
		'isbn'		=> $data->{isbn},
		'title'		=> $data->{title},
		'author'	=> $data->{author},
		'pages'		=> $data->{pages},
		'book_link'	=> TENLONG.$book,
		'image_link'	=> TENLONG."/BookSearch/".$data->{image_link},
		'pubdate'	=> $data->{pubdate},
		'publisher'	=> $data->{publisher},
		'price_list'	=> $data->{price_list},
		'price_sell'	=> $data->{price_sell},
	};

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

1;
__END__