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 $type = 'slogovi'; |
15 |
|
16 |
my $dbh = EPrints->dbh; |
17 |
my $sth = $dbh->prepare(qq{ |
18 |
SELECT |
19 |
archive_title.eprintid as id, |
20 |
title |
21 |
FROM archive_title |
22 |
}) || die $dbh->errstr(); |
23 |
$sth->execute() || die $sth->errstr(); |
24 |
|
25 |
my $indexer = Semantic::API::Index->new( |
26 |
storage => 'sqlite', |
27 |
database => 'eprints.db', |
28 |
collection => 'EPrints' |
29 |
); |
30 |
|
31 |
|
32 |
$indexer->add_word_filters( minimum_length => 3, |
33 |
too_many_numbers => 10, |
34 |
maximum_word_length => 15 ); |
35 |
|
36 |
# use this encoding for any incoming text |
37 |
#$indexer->set_default_encoding( "utf8"); |
38 |
|
39 |
my $total = 0; |
40 |
|
41 |
while (my $row = $sth->fetchrow_hashref ) { |
42 |
EPrints->id( $row->{id} ); |
43 |
my ( $title, $keywords, $abstract ) = ( |
44 |
_x( $row->{title} ), |
45 |
EPrints->lookup( 'keywords' ), |
46 |
EPrints->lookup( 'abstract' ) |
47 |
); |
48 |
my @body = split( /\W*\s+\W*/, "$title $title $title $keywords $keywords $abstract" ); |
49 |
my $body; |
50 |
foreach my $word ( @body ) { |
51 |
# $body .= StemHR->stem( $word ) . ' '; |
52 |
$body .= join(" ",EPrints::slogovi( $word )) . ' '; |
53 |
} |
54 |
|
55 |
warn "body: $body\n" if $debug; |
56 |
|
57 |
$indexer->index( $row->{id}, join(" ", @body, $body ) ); |
58 |
$total++; |
59 |
print STDERR _x( $row->{id}, " ", $row->{title} ), "\n"; |
60 |
} |
61 |
|
62 |
|
63 |
print STDERR "\nNow adding $total items to the database..."; |
64 |
$indexer->finish(); |
65 |
print STDERR "done!\n"; |
66 |
|