use Modern::Perl;
-use MARC::Record;
use Koha::SearchEngine;
use Koha::SearchEngine::Search;
use Koha::SearchEngine::QueryBuilder;
-use Koha::Util::Normalize qw/legacy_default remove_spaces upper_case lower_case/;
+use Koha::Util::Normalize qw(
+ ISBN
+ legacy_default
+ lower_case
+ remove_spaces
+ upper_case
+);
=head1 NAME
$sth->execute($id);
my $row = $sth->fetchrow_hashref;
$sth->finish();
- return undef unless defined $row;
+ return unless defined $row;
my $self = {};
$self->{'id'} = $row->{'matcher_id'};
foreach my $component (@{ $matchpoint->{'components'} }) {
$seqnum++;
$sth = $dbh->prepare_cached("INSERT INTO matchpoint_components
- (matchpoint_id, sequence, tag, subfields, offset, length)
+ (matchpoint_id, sequence, tag, subfields, `offset`, length)
VALUES (?, ?, ?, ?, ?, ?)");
$sth->bind_param(1, $matchpoint_id);
$sth->bind_param(2, $seqnum);
my ($source_record, $max_matches) = @_;
my $matches = {};
- my $marcframework_used = ''; # use the default framework
- my $QParser;
- $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser'));
foreach my $matchpoint ( @{ $self->{'matchpoints'} } ) {
my @source_keys = _get_match_keys( $source_record, $matchpoint );
next if scalar(@source_keys) == 0;
- # FIXME - because of a bug in QueryParser, an expression ofthe
- # format 'isbn:"isbn1" || isbn:"isbn2" || isbn"isbn3"...'
- # does not get parsed correctly, so we will not
- # do AggressiveMatchOnISBN if UseQueryParser is on
@source_keys = C4::Koha::GetVariationsOfISBNs(@source_keys)
if ( $matchpoint->{index} =~ /^isbn$/i
- && C4::Context->preference('AggressiveMatchOnISBN') )
- && !C4::Context->preference('UseQueryParser');
+ && C4::Context->preference('AggressiveMatchOnISBN') );
@source_keys = C4::Koha::GetVariationsOfISSNs(@source_keys)
if ( $matchpoint->{index} =~ /^issn$/i
- && C4::Context->preference('AggressiveMatchOnISSN') )
- && !C4::Context->preference('UseQueryParser');
+ && C4::Context->preference('AggressiveMatchOnISSN') );
# build query
my $query;
my $total_hits;
if ( $self->{'record_type'} eq 'biblio' ) {
- #NOTE: The QueryParser can't handle the CCL syntax of 'qualifier','qualifier', so fallback to non-QueryParser.
- #NOTE: You can see this in C4::Search::SimpleSearch() as well in a different way.
- if ($QParser && $matchpoint->{'index'} !~ m/\w,\w/) {
- $query = join( " || ",
- map { "$matchpoint->{'index'}:$_" } @source_keys );
- }
- else {
- my $phr = ( C4::Context->preference('AggressiveMatchOnISBN') || C4::Context->preference('AggressiveMatchOnISSN') ) ? ',phr' : q{};
- $query = join( " or ",
- map { "$matchpoint->{'index'}$phr=\"$_\"" } @source_keys );
- #NOTE: double-quote the values so you don't get a "Embedded truncation not supported" error when a term has a ? in it.
- }
+ my $phr = ( C4::Context->preference('AggressiveMatchOnISBN') || C4::Context->preference('AggressiveMatchOnISSN') ) ? ',phr' : q{};
+ $query = join( " OR ",
+ map { "$matchpoint->{'index'}$phr=\"$_\"" } @source_keys );
+ #NOTE: double-quote the values so you don't get a "Embedded truncation not supported" error when a term has a ? in it.
- my $searcher = Koha::SearchEngine::Search->new({index => $Koha::SearchEngine::BIBLIOS_INDEX});
+ # Use state variables to avoid recreating the objects every time.
+ # With Elasticsearch this also avoids creating a massive amount of
+ # ES connectors that would eventually run out of file descriptors.
+ state $searcher = Koha::SearchEngine::Search->new({index => $Koha::SearchEngine::BIBLIOS_INDEX});
( $error, $searchresults, $total_hits ) =
$searcher->simple_search_compat( $query, 0, $max_matches, undef, skip_normalize => 1 );
warn "search failed ($query) $error";
}
else {
- if ( C4::Context->preference('SearchEngine') eq 'Elasticsearch' ) {
- foreach my $matched ( @{$searchresults} ) {
- my ( $biblionumber_tag, $biblionumber_subfield ) = C4::Biblio::GetMarcFromKohaField( "biblio.biblionumber", $marcframework_used );
- my $id = ( $biblionumber_tag > 10 ) ?
- $matched->field($biblionumber_tag)->subfield($biblionumber_subfield) :
- $matched->field($biblionumber_tag)->data();
- $matches->{$id}->{score} += $matchpoint->{score};
- $matches->{$id}->{record} = $matched;
- }
- }
- else {
- foreach my $matched ( @{$searchresults} ) {
- $matches->{$matched}->{score} += $matchpoint->{'score'};
- $matches->{$matched}->{record} = $matched;
- }
+ foreach my $matched ( @{$searchresults} ) {
+ my $target_record = C4::Search::new_record_from_zebra( 'biblioserver', $matched );
+ my ( $biblionumber_tag, $biblionumber_subfield ) = C4::Biblio::GetMarcFromKohaField( "biblio.biblionumber" );
+ my $id = ( $biblionumber_tag > 10 ) ?
+ $target_record->field($biblionumber_tag)->subfield($biblionumber_subfield) :
+ $target_record->field($biblionumber_tag)->data();
+ $matches->{$id}->{score} += $matchpoint->{score};
+ $matches->{$id}->{record} = $target_record;
}
}
push @operator, 'exact';
push @value, $key;
}
- my $builder = Koha::SearchEngine::QueryBuilder->new({index => $Koha::SearchEngine::AUTHORITIES_INDEX});
- my $searcher = Koha::SearchEngine::Search->new({index => $Koha::SearchEngine::AUTHORITIES_INDEX});
+ # Use state variables to avoid recreating the objects every time.
+ # With Elasticsearch this also avoids creating a massive amount of
+ # ES connectors that would eventually run out of file descriptors.
+ state $builder = Koha::SearchEngine::QueryBuilder->new({index => $Koha::SearchEngine::AUTHORITIES_INDEX});
+ state $searcher = Koha::SearchEngine::Search->new({index => $Koha::SearchEngine::AUTHORITIES_INDEX});
my $search_query = $builder->build_authorities_query_compat(
\@marclist, \@and_or, \@excluding, \@operator,
\@value, undef, 'AuthidAsc'
# get rid of any that don't meet the required checks
$matches = {
map {
- _passes_required_checks( $source_record, $_, $self->{'required_checks'} )
+ _passes_required_checks( $source_record, $matches->{$_}->{'record'}, $self->{'required_checks'} )
? ( $_ => $matches->{$_} )
: ()
} keys %$matches
};
foreach my $id ( keys %$matches ) {
- my $target_record = C4::Search::new_record_from_zebra( 'biblioserver', $matches->{$id}->{record} );
- my $result = C4::Biblio::TransformMarcToKoha( $target_record, $marcframework_used );
push @results, {
- record_id => $result->{biblionumber},
+ record_id => $id,
score => $matches->{$id}->{score}
};
}
}
sub _passes_required_checks {
- my ($source_record, $target_blob, $matchchecks) = @_;
- my $target_record = MARC::Record->new_from_usmarc($target_blob); # FIXME -- need to avoid parsing record twice
+ my ($source_record, $target_record, $matchchecks) = @_;
# no checks supplied == automatic pass
return 1 if $#{ $matchchecks } == -1;
for (my $i = 0; $i <= $#{ $matchpoint->{'components'} }; $i++) {
my $component = $matchpoint->{'components'}->[$i];
my $j = -1;
- FIELD: foreach my $field ($source_record->field($component->{'tag'})) {
+
+ my @fields = ();
+ my $tag = $component->{'tag'};
+ if ($tag && $tag eq 'LDR'){
+ $fields[0] = $source_record->leader();
+ }
+ else {
+ @fields = $source_record->field($tag);
+ }
+
+ FIELD: foreach my $field (@fields) {
$j++;
last FIELD if $j > 0 and $check_only_first_repeat;
last FIELD if $i > 0 and $j > $#keys;
my $string;
- if ( $field->is_control_field() ) {
+ if ( ! ref $field ){
+ $string = "$field";
+ }
+ elsif ( $field->is_control_field() ) {
$string = $field->data();
- } else {
+ } elsif ( defined $component->{subfields} && keys %{$component->{subfields}} ){
$string = $field->as_string(
join('', keys %{ $component->{ subfields } }), ' ' # ' ' as separator
);
+ } else {
+ $string = $field->as_string();
}
if ($component->{'length'}>0) {
elsif ( $norm eq 'legacy_default' ) {
$key = legacy_default($key);
}
+ elsif ( $norm eq 'ISBN' ) {
+ $key = ISBN($key);
+ }
} else {
warn "Invalid normalization routine required ($norm)"
unless $norm eq 'none';
'remove_spaces',
'upper_case',
'lower_case',
- 'legacy_default'
+ 'legacy_default',
+ 'ISBN'
);
}
projects / koha-ffzg.git / blobdiff