Semantic-Engine/EPrints/index.pl

#!/usr/bin/perl -w

use strict;
use Semantic::API;
use Data::Dump qw/dump/;

use EPrints qw/_x/;

use lib '/home/dpavlin/stem-hr/';
use StemHR;

my $debug = shift @ARGV;
my $use_score = 0;

my $dbh = EPrints->dbh;
my $sth = $dbh->prepare(qq{
SELECT
        archive_title.eprintid as id,
        title
FROM archive_title 
}) || die $dbh->errstr();
$sth->execute() || die $sth->errstr();

my $indexer = Semantic::API::Index->new(
        storage => 'sqlite',
        database => 'eprints.db',
        collection => 'EPrints'
);


$indexer->add_word_filters( minimum_length => 3,
                            too_many_numbers => 10,
                            maximum_word_length => 15 );

# use this encoding for any incoming text
$indexer->set_default_encoding( "iso-8859-2" ); 

my $total = 0;

while (my $row = $sth->fetchrow_hashref ) {
        EPrints->id( $row->{id} );
        my $parts = {
                title => [ _x( $row->{title} ), 4 ],
                keywords => [ EPrints->lookup( 'keywords' ), 3 ],
                abstract => [ EPrints->lookup( 'abstract' ), 2 ],
#               content => [ EPrints->fulltext_content, 1 ],
        };

        my $body = '';

        foreach my $part ( qw/title keywords abstract content/ ) {
                my $content = $parts->{$part}->[0];
                next unless defined $content;
                
#               $content = StemHR->stem( $content );
                $content = EPrints->slogovi( $content );

                if ( $use_score ) {
                        map { $body .= $content } 1 .. $parts->{$part}->[1];
                } else {
                        $body .= $content;
                }
        }

        $indexer->index( $row->{id}, $body );
        $total++;
        print STDERR "$total: ", $row->{id}, " ", _x( $row->{title} ), " - ", length($body), " bytes\n";
}


print STDERR "\nNow adding $total items to the database...";
$indexer->finish(); 
print STDERR "done!\n"; 

1	dpavlin	1	#!/usr/bin/perl -w
2
3			use strict;
4			use Semantic::API;
5			use Data::Dump qw/dump/;
6	dpavlin	13
7	dpavlin	1	use EPrints qw/_x/;
8	dpavlin	13
9	dpavlin	1	use lib '/home/dpavlin/stem-hr/';
10			use StemHR;
11
12			my $debug = shift @ARGV;
13	dpavlin	14	my $use_score = 0;
14	dpavlin	1
15			my $dbh = EPrints->dbh;
16			my $sth = $dbh->prepare(qq{
17			SELECT
18			archive_title.eprintid as id,
19			title
20			FROM archive_title
21			}) \|\| die $dbh->errstr();
22			$sth->execute() \|\| die $sth->errstr();
23
24			my $indexer = Semantic::API::Index->new(
25			storage => 'sqlite',
26			database => 'eprints.db',
27			collection => 'EPrints'
28			);
29
30
31			$indexer->add_word_filters( minimum_length => 3,
32			too_many_numbers => 10,
33			maximum_word_length => 15 );
34
35			# use this encoding for any incoming text
36	dpavlin	8	$indexer->set_default_encoding( "iso-8859-2" );
37	dpavlin	1
38			my $total = 0;
39
40			while (my $row = $sth->fetchrow_hashref ) {
41			EPrints->id( $row->{id} );
42	dpavlin	14	my $parts = {
43			title => [ _x( $row->{title} ), 4 ],
44			keywords => [ EPrints->lookup( 'keywords' ), 3 ],
45			abstract => [ EPrints->lookup( 'abstract' ), 2 ],
46			# content => [ EPrints->fulltext_content, 1 ],
47			};
48
49	dpavlin	8	my $body = '';
50	dpavlin	1
51	dpavlin	14	foreach my $part ( qw/title keywords abstract content/ ) {
52			my $content = $parts->{$part}->[0];
53			next unless defined $content;
54
55			# $content = StemHR->stem( $content );
56			$content = EPrints->slogovi( $content );
57	dpavlin	8
58	dpavlin	14	if ( $use_score ) {
59			map { $body .= $content } 1 .. $parts->{$part}->[1];
60			} else {
61			$body .= $content;
62			}
63			}
64	dpavlin	1
65	dpavlin	14	$indexer->index( $row->{id}, $body );
66	dpavlin	1	$total++;
67	dpavlin	14	print STDERR "$total: ", $row->{id}, " ", _x( $row->{title} ), " - ", length($body), " bytes\n";
68	dpavlin	1	}
69
70
71			print STDERR "\nNow adding $total items to the database...";
72			$indexer->finish();
73			print STDERR "done!\n";
74