1 |
dpavlin |
1 |
#!/usr/bin/perl -w |
2 |
|
|
|
3 |
|
|
use strict; |
4 |
|
|
use Semantic::API; |
5 |
|
|
use Data::Dump qw/dump/; |
6 |
|
|
use EPrints qw/_x/; |
7 |
|
|
use lib '/home/dpavlin/stem-hr/'; |
8 |
|
|
use StemHR; |
9 |
|
|
|
10 |
|
|
warn dump( StemHR->stem('kuæni') ); |
11 |
|
|
|
12 |
|
|
my $debug = shift @ARGV; |
13 |
|
|
|
14 |
|
|
my $dbh = EPrints->dbh; |
15 |
|
|
my $sth = $dbh->prepare(qq{ |
16 |
|
|
SELECT |
17 |
|
|
archive_title.eprintid as id, |
18 |
|
|
title |
19 |
|
|
FROM archive_title |
20 |
|
|
}) || die $dbh->errstr(); |
21 |
|
|
$sth->execute() || die $sth->errstr(); |
22 |
|
|
|
23 |
|
|
my $indexer = Semantic::API::Index->new( |
24 |
|
|
storage => 'sqlite', |
25 |
|
|
database => 'eprints.db', |
26 |
|
|
collection => 'EPrints' |
27 |
|
|
); |
28 |
|
|
|
29 |
|
|
|
30 |
|
|
$indexer->add_word_filters( minimum_length => 3, |
31 |
|
|
too_many_numbers => 10, |
32 |
|
|
maximum_word_length => 15 ); |
33 |
|
|
|
34 |
|
|
# use this encoding for any incoming text |
35 |
|
|
$indexer->set_default_encoding( "utf8"); |
36 |
|
|
|
37 |
|
|
my $total = 0; |
38 |
|
|
|
39 |
|
|
while (my $row = $sth->fetchrow_hashref ) { |
40 |
|
|
EPrints->id( $row->{id} ); |
41 |
|
|
my ( $title, $keywords, $abstract ) = ( |
42 |
|
|
_x( $row->{title} ), |
43 |
|
|
EPrints->lookup( 'keywords' ), |
44 |
|
|
EPrints->lookup( 'abstract' ) |
45 |
|
|
); |
46 |
|
|
my @body = split( /\W*\s+\W*/, "$title $title $title $keywords $keywords $abstract" ); |
47 |
|
|
my $body; |
48 |
|
|
foreach my $word ( @body ) { |
49 |
|
|
$body .= StemHR->stem( $word ) . ' '; |
50 |
|
|
} |
51 |
|
|
|
52 |
|
|
warn "body: $body\n" if $debug; |
53 |
|
|
|
54 |
|
|
$indexer->index( $row->{id}, join(" ", @body, $body ) ); |
55 |
|
|
$total++; |
56 |
|
|
print STDERR _x( $row->{id}, " ", $row->{title} ), "\n"; |
57 |
|
|
} |
58 |
|
|
|
59 |
|
|
|
60 |
|
|
print STDERR "\nNow adding $total items to the database..."; |
61 |
|
|
$indexer->finish(); |
62 |
|
|
print STDERR "done!\n"; |
63 |
|
|
|