package MyAggregator;
use Moose;
extends 'Dezi::Aggregator';

use Dezi::Doc;

sub crawl {
    my ( $self, $inputfile ) = @_;

    open( RF1, $inputfile ) or die "Can't open < $inputfile: $!";
    my $header = <RF1>;    # read out header line

    my $count = 0;
    while ( my $line = <RF1> ) {
        chomp $line;
        $count++;
        my @array = split( /\t/, $line );

        my $dezi_doc = Dezi::Doc->new( uri => $count, );
        $dezi_doc->set_field( 'shopid'   => $array[0] );
        $dezi_doc->set_field( 'prodtype' => $array[1] );
        $dezi_doc->set_field( 'prodid'   => $array[2] );
        $dezi_doc->set_field( 'prodname' => $array[3] );

        my $xml = $dezi_doc->as_string_ref;

        my $doc = $self->doc_class->new(
            content => $$xml,
            url     => $count,
            modtime => time(),
            parser  => 'XML*',
            type    => 'application/xml',
            size    => length $$xml,
        );

        $self->indexer->process($doc);
    }
    close(RF1);
    return $count;
}

1;