3 # This file is part of Koha.
5 # Koha is free software; you can redistribute it and/or modify it
6 # under the terms of the GNU General Public License as published by
7 # the Free Software Foundation; either version 3 of the License, or
8 # (at your option) any later version.
10 # Koha is distributed in the hope that it will be useful, but
11 # WITHOUT ANY WARRANTY; without even the implied warranty of
12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 # GNU General Public License for more details.
15 # You should have received a copy of the GNU General Public License
16 # along with Koha; if not, see <http://www.gnu.org/licenses>.
20 use C4::Biblio qw( TransformMarcToKoha GetMarcFromKohaField GetFrameworkCode GetAuthorisedValueDesc GetBiblioData );
21 use C4::Koha qw( getFacets GetVariationsOfISBN GetNormalizedUPC GetNormalizedEAN GetNormalizedOCLCNumber GetNormalizedISBN getitemtypeimagelocation );
26 use C4::XSLT qw( XSLTParse4Display );
27 use C4::Reserves qw( GetReserveStatus );
28 use C4::Charset qw( SetUTF8Flag );
29 use Koha::AuthorisedValues;
34 use Koha::RecordProcessor;
40 our (@ISA, @EXPORT_OK);
51 enabled_staff_search_views
60 C4::Search - Functions for searching the Koha catalog.
64 See opac/opac-search.pl or catalogue/search.pl for example of usage
68 This module provides searching functions for Koha's bibliographic databases
74 # make all your functions, whether exported or not;
78 ($biblionumber,$biblionumber,$title) = FindDuplicate($record);
80 This function attempts to find duplicate records using a hard-coded, fairly simplistic algorithm
86 my $dbh = C4::Context->dbh;
87 my $result = TransformMarcToKoha( $record, '' );
91 # search duplicate on ISBN, easy and fast..
93 if ( $result->{isbn} ) {
94 $result->{isbn} =~ s/\(.*$//;
95 $result->{isbn} =~ s/\s+$//;
96 $query = "isbn:$result->{isbn}";
100 my $titleindex = 'ti,ext';
101 my $authorindex = 'au,ext';
104 $result->{title} =~ s /\\//g;
105 $result->{title} =~ s /\"//g;
106 $result->{title} =~ s /\(//g;
107 $result->{title} =~ s /\)//g;
109 $query = "$titleindex:\"$result->{title}\"";
110 if ( $result->{author} ) {
111 $result->{author} =~ s /\\//g;
112 $result->{author} =~ s /\"//g;
113 $result->{author} =~ s /\(//g;
114 $result->{author} =~ s /\)//g;
116 $query .= " $op $authorindex:\"$result->{author}\"";
120 my $searcher = Koha::SearchEngine::Search->new({index => $Koha::SearchEngine::BIBLIOS_INDEX});
121 my ( $error, $searchresults, undef ) = $searcher->simple_search_compat($query,0,50);
123 if (!defined $error) {
124 foreach my $possible_duplicate_record (@{$searchresults}) {
125 my $marcrecord = new_record_from_zebra(
127 $possible_duplicate_record
130 my $result = TransformMarcToKoha( $marcrecord, '' );
132 # FIXME :: why 2 $biblionumber ?
134 push @results, $result->{'biblionumber'};
135 push @results, $result->{'title'};
144 ( $error, $results, $total_hits ) = SimpleSearch( $query, $offset, $max_results, [@servers], [%options] );
146 This function provides a simple search API on the bibliographic catalog
152 * $query can be a simple keyword or a complete CCL query
153 * @servers is optional. Defaults to biblioserver as found in koha-conf.xml
154 * $offset - If present, represents the number of records at the beginning to omit. Defaults to 0
155 * $max_results - if present, determines the maximum number of records to fetch. undef is All. defaults to undef.
156 * %options is optional. (e.g. "skip_normalize" allows you to skip changing : to = )
161 Returns an array consisting of three elements
162 * $error is undefined unless an error is detected
163 * $results is a reference to an array of records.
164 * $total_hits is the number of hits that would have been returned with no limit
166 If an error is returned the two other return elements are undefined. If error itself is undefined
167 the other two elements are always defined
169 =item C<usage in the script:>
173 my ( $error, $marcresults, $total_hits ) = SimpleSearch($query);
175 if (defined $error) {
176 $template->param(query_error => $error);
177 warn "error: ".$error;
178 output_html_with_http_headers $input, $cookie, $template->output;
182 my $hits = @{$marcresults};
185 for my $r ( @{$marcresults} ) {
186 my $marcrecord = MARC::File::USMARC::decode($r);
187 my $biblio = TransformMarcToKoha($marcrecord,q{});
189 #build the iarray of hashs for the template.
191 title => $biblio->{'title'},
192 subtitle => $biblio->{'subtitle'},
193 biblionumber => $biblio->{'biblionumber'},
194 author => $biblio->{'author'},
195 publishercode => $biblio->{'publishercode'},
196 publicationyear => $biblio->{'publicationyear'},
201 $template->param(result=>\@results);
206 my ( $query, $offset, $max_results, $servers, %options ) = @_;
208 return ( 'No query entered', undef, undef ) unless $query;
209 # FIXME hardcoded value. See catalog/search.pl & opac-search.pl too.
210 my @servers = defined ( $servers ) ? @$servers : ( 'biblioserver' );
217 # Initialize & Search Zebra
218 for ( my $i = 0 ; $i < @servers ; $i++ ) {
220 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
221 $query =~ s/:/=/g unless $options{skip_normalize};
222 $zoom_queries[$i] = ZOOM::Query::CCL2RPN->new( $query, $zconns[$i]);
223 $tmpresults[$i] = $zconns[$i]->search( $zoom_queries[$i] );
227 $zconns[$i]->errmsg() . " ("
228 . $zconns[$i]->errcode() . ") "
229 . $zconns[$i]->addinfo() . " "
230 . $zconns[$i]->diagset();
232 return ( $error, undef, undef ) if $zconns[$i]->errcode();
236 # caught a ZOOM::Exception
240 . $@->addinfo() . " "
242 warn $error." for query: $query";
243 return ( $error, undef, undef );
252 my $first_record = defined($offset) ? $offset + 1 : 1;
253 my $hits = $tmpresults[ $i - 1 ]->size();
254 $total_hits += $hits;
255 my $last_record = $hits;
256 if ( defined $max_results && $offset + $max_results < $hits ) {
257 $last_record = $offset + $max_results;
260 for my $j ( $first_record .. $last_record ) {
262 $tmpresults[ $i - 1 ]->record( $j - 1 )->raw()
265 push @{$results}, $record if defined $record;
270 foreach my $zoom_query (@zoom_queries) {
271 $zoom_query->destroy();
274 return ( undef, $results, $total_hits );
279 ( undef, $results_hashref, \@facets_loop ) = getRecords (
281 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
282 $results_per_page, $offset, $branches, $itemtypes,
283 $query_type, $scan, $opac
286 The all singing, all dancing, multi-server, asynchronous, scanning,
287 searching, record nabbing, facet-building
289 See verbose embedded documentation.
295 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
296 $results_per_page, $offset, $branches, $itemtypes,
297 $query_type, $scan, $opac
300 my @servers = @$servers_ref;
301 my @sort_by = @$sort_by_ref;
302 $offset = 0 if $offset < 0;
304 # Initialize variables for the ZOOM connection and results object
307 my $results_hashref = ();
309 # TODO simplify this structure ( { branchcode => $branchname } is enought) and remove this parameter
310 $branches ||= { map { $_->branchcode => { branchname => $_->branchname } } Koha::Libraries->search };
312 # Initialize variables for the faceted results objects
313 my $facets_counter = {};
314 my $facets_info = {};
315 my $facets = getFacets();
317 my @facets_loop; # stores the ref to array of hashes for template facets loop
319 ### LOOP THROUGH THE SERVERS
320 for ( my $i = 0 ; $i < @servers ; $i++ ) {
321 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
323 # perform the search, create the results objects
324 # if this is a local search, use the $koha-query, if it's a federated one, use the federated-query
325 my $query_to_use = ($servers[$i] =~ /biblioserver/) ? $koha_query : $simple_query;
327 Koha::Logger->get->debug($simple_query) if $scan;
329 # Check if we've got a query_type defined, if so, use it
332 if ($query_type =~ /^ccl/) {
333 $query_to_use =~ s/\:/\=/g; # change : to = last minute (FIXME)
334 $results[$i] = $zconns[$i]->search(ZOOM::Query::CCL2RPN->new($query_to_use, $zconns[$i]));
335 } elsif ($query_type =~ /^cql/) {
336 $results[$i] = $zconns[$i]->search(ZOOM::Query::CQL->new($query_to_use, $zconns[$i]));
337 } elsif ($query_type =~ /^pqf/) {
338 $results[$i] = $zconns[$i]->search(ZOOM::Query::PQF->new($query_to_use, $zconns[$i]));
340 warn "Unknown query_type '$query_type'. Results undetermined.";
343 $results[$i] = $zconns[$i]->scan( ZOOM::Query::CCL2RPN->new($query_to_use, $zconns[$i]));
345 $results[$i] = $zconns[$i]->search(ZOOM::Query::CCL2RPN->new($query_to_use, $zconns[$i]));
349 warn "WARNING: query problem with $query_to_use " . $@;
352 # Concatenate the sort_by limits and pass them to the results object
353 # Note: sort will override rank
355 foreach my $sort (@sort_by) {
356 if ( $sort eq "author_az" || $sort eq "author_asc" ) {
357 $sort_by .= "1=1003 <i ";
359 elsif ( $sort eq "author_za" || $sort eq "author_dsc" ) {
360 $sort_by .= "1=1003 >i ";
362 elsif ( $sort eq "popularity_asc" ) {
363 $sort_by .= "1=9003 <i ";
365 elsif ( $sort eq "popularity_dsc" ) {
366 $sort_by .= "1=9003 >i ";
368 elsif ( $sort eq "call_number_asc" ) {
369 $sort_by .= "1=8007 <i ";
371 elsif ( $sort eq "call_number_dsc" ) {
372 $sort_by .= "1=8007 >i ";
374 elsif ( $sort eq "pubdate_asc" ) {
375 $sort_by .= "1=31 <i ";
377 elsif ( $sort eq "pubdate_dsc" ) {
378 $sort_by .= "1=31 >i ";
380 elsif ( $sort eq "acqdate_asc" ) {
381 $sort_by .= "1=32 <i ";
383 elsif ( $sort eq "acqdate_dsc" ) {
384 $sort_by .= "1=32 >i ";
386 elsif ( $sort eq "title_az" || $sort eq "title_asc" ) {
387 $sort_by .= "1=4 <i ";
389 elsif ( $sort eq "title_za" || $sort eq "title_dsc" ) {
390 $sort_by .= "1=4 >i ";
393 warn "Ignoring unrecognized sort '$sort' requested" if $sort_by;
396 if ( $sort_by && !$scan && $results[$i] ) {
397 if ( $results[$i]->sort( "yaz", $sort_by ) < 0 ) {
398 warn "WARNING sort $sort_by failed";
401 } # finished looping through servers
403 # The big moment: asynchronously retrieve results from all servers
408 my ( $i, $size ) = @_;
411 # loop through the results
412 $results_hash->{'hits'} = $size;
414 if ( $offset + $results_per_page <= $size ) {
415 $times = $offset + $results_per_page;
421 for ( my $j = $offset ; $j < $times ; $j++ ) {
424 ## Check if it's an index scan
426 my ( $term, $occ ) = $results[ $i - 1 ]->display_term($j);
428 # here we create a minimal MARC record and hand it off to the
429 # template just like a normal result ... perhaps not ideal, but
431 my $tmprecord = MARC::Record->new();
432 $tmprecord->encoding('UTF-8');
436 # the minimal record in author/title (depending on MARC flavour)
437 if ( C4::Context->preference("marcflavour") eq
440 $tmptitle = MARC::Field->new(
445 $tmprecord->append_fields($tmptitle);
449 MARC::Field->new( '245', ' ', ' ', a => $term, );
451 MARC::Field->new( '100', ' ', ' ', a => $occ, );
452 $tmprecord->append_fields($tmptitle);
453 $tmprecord->append_fields($tmpauthor);
455 $results_hash->{'RECORDS'}[$j] =
456 $tmprecord->as_usmarc();
461 $record = $results[ $i - 1 ]->record($j)->raw();
462 # warn "RECORD $j:".$record;
463 $results_hash->{'RECORDS'}[$j] = $record;
467 $results_hashref->{ $servers[ $i - 1 ] } = $results_hash;
469 # Fill the facets while we're looping, but only for the
470 # biblioserver and not for a scan
471 if ( !$scan && $servers[ $i - 1 ] =~ /biblioserver/ ) {
472 $facets_counter = GetFacets( $results[ $i - 1 ] );
473 $facets_info = _get_facets_info( $facets );
477 if ( $servers[ $i - 1 ] =~ /biblioserver/ ) {
479 sort { $a cmp $b } keys %$facets_counter
482 my @this_facets_array;
485 $facets_counter->{$link_value}
486 ->{$b} <=> $facets_counter->{$link_value}
488 } keys %{ $facets_counter->{$link_value} }
491 # Sanitize the link value : parenthesis, question and exclamation mark will cause errors with CCL
492 my $facet_link_value = $one_facet;
493 $facet_link_value =~ s/[()!?¡¿؟]/ /g;
495 # fix the length that will display in the label,
496 my $facet_label_value = $one_facet;
497 my $facet_max_length = C4::Context->preference(
498 'FacetLabelTruncationLength')
501 substr( $one_facet, 0, $facet_max_length )
503 if length($facet_label_value) >
506 # if it's a branch, label by the name, not the code,
507 if ( $link_value =~ /branch/ ) {
508 if ( defined $branches
509 && ref($branches) eq "HASH"
510 && defined $branches->{$one_facet}
511 && ref( $branches->{$one_facet} ) eq
515 $branches->{$one_facet}
519 $facet_label_value = "*";
523 # if it's a itemtype, label by the name, not the code,
524 if ( $link_value =~ /itype/ ) {
525 if ( defined $itemtypes
526 && ref($itemtypes) eq "HASH"
527 && defined $itemtypes->{$one_facet}
528 && ref( $itemtypes->{$one_facet} ) eq
532 $itemtypes->{$one_facet}
533 ->{translated_description};
537 # also, if it's a location code, use the name instead of the code
538 if ( $link_value =~ /location/ ) {
539 # TODO Retrieve all authorised values at once, instead of 1 query per entry
540 my $av = Koha::AuthorisedValues->search({ category => 'LOC', authorised_value => $one_facet });
541 $facet_label_value = $av->count ? $av->next->opac_description : '';
544 # also, if it's a collection code, use the name instead of the code
545 if ( $link_value =~ /ccode/ ) {
546 # TODO Retrieve all authorised values at once, instead of 1 query per entry
547 my $av = Koha::AuthorisedValues->search({ category => 'CCODE', authorised_value => $one_facet });
548 $facet_label_value = $av->count ? $av->next->opac_description : '';
551 # but we're down with the whole label being in the link's title.
552 push @this_facets_array,
555 $facets_counter->{$link_value}
557 facet_label_value => $facet_label_value,
558 facet_title_value => $one_facet,
559 facet_link_value => $facet_link_value,
560 type_link_value => $link_value,
562 if ($facet_label_value);
567 type_link_value => $link_value,
568 type_id => $link_value . "_id",
570 . $facets_info->{$link_value}->{'label_value'} =>
572 facets => \@this_facets_array,
576 $facets_info->{$link_value}->{'label_value'} =~
579 and ( Koha::Libraries->search->count == 1 )
586 # This sorts the facets into alphabetical order
588 foreach my $f (@facets_loop) {
589 $f->{facets} = [ sort { uc($a->{facet_label_value}) cmp uc($b->{facet_label_value}) } @{ $f->{facets} } ];
593 return ( undef, $results_hashref, \@facets_loop );
601 my $use_zebra_facets = C4::Context->config('use_zebra_facets') // 0;
603 if ( $use_zebra_facets ) {
604 $facets = _get_facets_from_zebra( $rs );
606 $facets = _get_facets_from_records( $rs );
612 sub _get_facets_from_records {
616 my $facets_maxrecs = C4::Context->preference('maxRecordsForFacets') // 20;
617 my $facets_config = getFacets();
619 my $size = $rs->size();
620 my $jmax = $size > $facets_maxrecs
624 for ( my $j = 0 ; $j < $jmax ; $j++ ) {
626 my $marc_record = new_record_from_zebra (
628 $rs->record( $j )->raw()
631 if ( ! defined $marc_record ) {
632 warn "ERROR DECODING RECORD - $@: " .
633 $rs->record( $j )->raw();
637 _get_facets_data_from_record( $marc_record, $facets_config, $facets );
643 =head2 _get_facets_data_from_record
645 C4::Search::_get_facets_data_from_record( $marc_record, $facets, $facets_counter );
647 Internal function that extracts facets information from a MARC::Record object
648 and populates $facets_counter for using in getRecords.
650 $facets is expected to be filled with C4::Koha::getFacets output (i.e. the configured
655 sub _get_facets_data_from_record {
657 my ( $marc_record, $facets, $facets_counter ) = @_;
659 for my $facet (@$facets) {
663 foreach my $tag ( @{ $facet->{ tags } } ) {
665 # tag number is the first three digits
666 my $tag_num = substr( $tag, 0, 3 );
667 # subfields are the remainder
668 my $subfield_letters = substr( $tag, 3 );
670 my @fields = $marc_record->field( $tag_num );
671 foreach my $field (@fields) {
672 # If $field->indicator(1) eq 'z', it means it is a 'see from'
673 # field introduced because of IncludeSeeFromInSearches, so skip it
674 next if $field->indicator(1) eq 'z';
676 my $data = $field->as_string( $subfield_letters, $facet->{ sep } );
677 $data =~ s/\s*(?<!\p{Uppercase})[.\-,;]*\s*$//;
679 unless ( grep { $_ eq $data } @used_datas ) {
680 push @used_datas, $data;
681 $facets_counter->{ $facet->{ idx } }->{ $data }++;
688 =head2 _get_facets_from_zebra
690 my $facets = _get_facets_from_zebra( $result_set )
692 Retrieves facets for a specified result set. It loops through the facets defined
693 in C4::Koha::getFacets and returns a hash with the following structure:
703 sub _get_facets_from_zebra {
707 # save current elementSetName
708 my $elementSetName = $rs->option( 'elementSetName' );
710 my $facets_loop = getFacets();
711 my $facets_data = {};
712 # loop through defined facets and fill the facets hashref
713 foreach my $facet ( @$facets_loop ) {
715 my $idx = $facet->{ idx };
716 my $sep = $facet->{ sep };
717 my $facet_values = _get_facet_from_result_set( $idx, $rs, $sep );
718 if ( $facet_values ) {
719 # we've actually got a result
720 $facets_data->{ $idx } = $facet_values;
723 # set elementSetName to its previous value to avoid side effects
724 $rs->option( elementSetName => $elementSetName );
729 =head2 _get_facet_from_result_set
732 C4::Search::_get_facet_from_result_set( $facet_idx, $result_set, $sep )
734 Internal function that extracts facet information for a specific index ($facet_idx) and
735 returns a hash containing facet values and count:
738 $facet_value => $count ,
742 Warning: this function has the side effect of changing the elementSetName for the result
743 set. It is a helper function for the main loop, which takes care of backing it up for
748 sub _get_facet_from_result_set {
750 my $facet_idx = shift;
754 my $internal_sep = '<*>';
755 my $facetMaxCount = C4::Context->preference('FacetMaxCount') // 20;
757 return if ( ! defined $facet_idx || ! defined $rs );
758 # zebra's facet element, untokenized index
759 my $facet_element = 'zebra::facet::' . $facet_idx . ':0:' . $facetMaxCount;
760 # configure zebra results for retrieving the desired facet
761 $rs->option( elementSetName => $facet_element );
762 # get the facet record from result set
763 my $facet = $rs->record( 0 )->raw;
764 # if the facet has no restuls...
765 return if !defined $facet;
766 # TODO: benchmark DOM vs. SAX performance
767 my $facet_dom = XML::LibXML->load_xml(
770 my @terms = $facet_dom->getElementsByTagName('term');
774 foreach my $term ( @terms ) {
775 my $facet_value = $term->textContent;
776 $facet_value =~ s/\s*(?<!\p{Uppercase})[.\-,;]*\s*$//;
777 $facet_value =~ s/\Q$internal_sep\E/$sep/ if defined $sep;
778 $facets->{ $facet_value } += $term->getAttribute( 'occur' );
784 =head2 _get_facets_info
786 my $facets_info = C4::Search::_get_facets_info( $facets )
788 Internal function that extracts facets information and properly builds
789 the data structure needed to render facet labels.
793 sub _get_facets_info {
797 my $facets_info = {};
799 for my $facet ( @$facets ) {
800 $facets_info->{ $facet->{ idx } }->{ label_value } = $facet->{ label };
807 sub _detect_truncation {
808 my ( $operand, $index ) = @_;
809 my ( @nontruncated, @righttruncated, @lefttruncated, @rightlefttruncated,
812 my @wordlist = split( /\s/, $operand );
813 foreach my $word (@wordlist) {
814 if ( $word =~ s/^\*([^\*]+)\*$/$1/ ) {
815 push @rightlefttruncated, $word;
817 elsif ( $word =~ s/^\*([^\*]+)$/$1/ ) {
818 push @lefttruncated, $word;
820 elsif ( $word =~ s/^([^\*]+)\*$/$1/ ) {
821 push @righttruncated, $word;
823 elsif ( index( $word, "*" ) < 0 ) {
824 push @nontruncated, $word;
827 push @regexpr, $word;
831 \@nontruncated, \@righttruncated, \@lefttruncated,
832 \@rightlefttruncated, \@regexpr
837 sub _build_stemmed_operand {
838 my ($operand,$lang) = @_;
839 require Lingua::Stem::Snowball ;
840 my $stemmed_operand=q{};
842 # Stemmer needs language
843 return $operand unless $lang;
845 # If operand contains a digit, it is almost certainly an identifier, and should
846 # not be stemmed. This is particularly relevant for ISBNs and ISSNs, which
847 # can contain the letter "X" - for example, _build_stemmend_operand would reduce
848 # "014100018X" to "x ", which for a MARC21 database would bring up irrelevant
849 # results (e.g., "23 x 29 cm." from the 300$c). Bug 2098.
850 return $operand if $operand =~ /\d/;
852 # FIXME: the locale should be set based on the user's language and/or search choice
854 # Make sure we only use the first two letters from the language code
855 $lang = lc(substr($lang, 0, 2));
856 # The language codes for the two variants of Norwegian will now be "nb" and "nn",
857 # none of which Lingua::Stem::Snowball can use, so we need to "translate" them
858 if ($lang eq 'nb' || $lang eq 'nn') {
861 my $stemmer = Lingua::Stem::Snowball->new( lang => $lang,
862 encoding => "UTF-8" );
864 my @words = split( / /, $operand );
865 my @stems = $stemmer->stem(\@words);
866 for my $stem (@stems) {
867 $stemmed_operand .= "$stem";
868 $stemmed_operand .= "?"
869 unless ( $stem =~ /(and$|or$|not$)/ ) || ( length($stem) < 3 );
870 $stemmed_operand .= " ";
873 Koha::Logger->get->debug("STEMMED OPERAND: $stemmed_operand");
874 return $stemmed_operand;
878 sub _build_weighted_query {
880 # FIELD WEIGHTING - This is largely experimental stuff. What I'm committing works
881 # pretty well but could work much better if we had a smarter query parser
882 my ( $operand, $stemmed_operand, $index ) = @_;
883 my $stemming = C4::Context->preference("QueryStemming") || 0;
884 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
885 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
886 $operand =~ s/"/ /g; # Bug 7518: searches with quotation marks don't work
888 my $weighted_query .= "(rk=("; # Specifies that we're applying rank
890 # Keyword, or, no index specified
891 if ( ( $index eq 'kw' ) || ( !$index ) ) {
893 "Title-cover,ext,r1=\"$operand\""; # exact title-cover
894 $weighted_query .= " or ti,ext,r2=\"$operand\""; # exact title
895 $weighted_query .= " or Title-cover,phr,r3=\"$operand\""; # phrase title
896 $weighted_query .= " or ti,wrdl,r4=\"$operand\""; # words in title
897 #$weighted_query .= " or any,ext,r4=$operand"; # exact any
898 #$weighted_query .=" or kw,wrdl,r5=\"$operand\""; # word list any
899 $weighted_query .= " or wrdl,fuzzy,r8=\"$operand\""
900 if $fuzzy_enabled; # add fuzzy, word list
901 $weighted_query .= " or wrdl,right-Truncation,r9=\"$stemmed_operand\""
902 if ( $stemming and $stemmed_operand )
903 ; # add stemming, right truncation
904 $weighted_query .= " or wrdl,r9=\"$operand\"";
906 # embedded sorting: 0 a-z; 1 z-a
907 # $weighted_query .= ") or (sort1,aut=1";
910 # Barcode searches should skip this process
911 elsif ( $index eq 'bc' ) {
912 $weighted_query .= "bc=\"$operand\"";
915 # Authority-number searches should skip this process
916 elsif ( $index eq 'an' ) {
917 $weighted_query .= "an=\"$operand\"";
920 # If the index is numeric, don't autoquote it.
921 elsif ( $index =~ /,st-numeric$/ ) {
922 $weighted_query .= " $index=$operand";
925 # If the index already has more than one qualifier, wrap the operand
926 # in quotes and pass it back (assumption is that the user knows what they
927 # are doing and won't appreciate us mucking up their query
928 elsif ( $index =~ ',' ) {
929 $weighted_query .= " $index=\"$operand\"";
932 #TODO: build better cases based on specific search indexes
934 $weighted_query .= " $index,ext,r1=\"$operand\""; # exact index
935 #$weighted_query .= " or (title-sort-az=0 or $index,startswithnt,st-word,r3=$operand #)";
936 $weighted_query .= " or $index,phr,r3=\"$operand\""; # phrase index
937 $weighted_query .= " or $index,wrdl,r6=\"$operand\""; # word list index
938 $weighted_query .= " or $index,wrdl,fuzzy,r8=\"$operand\""
939 if $fuzzy_enabled; # add fuzzy, word list
940 $weighted_query .= " or $index,wrdl,rt,r9=\"$stemmed_operand\""
941 if ( $stemming and $stemmed_operand ); # add stemming, right truncation
944 $weighted_query .= "))"; # close rank specification
945 return $weighted_query;
950 Return an array with available indexes.
974 'Author-personal-bibliography',
984 'Chronological-subdivision',
994 'Conference-name-heading',
995 'Conference-name-see',
996 'Conference-name-seealso',
999 'Control-number-identifier',
1003 'Corporate-name-heading',
1004 'Corporate-name-see',
1005 'Corporate-name-seealso',
1006 'Country-publication',
1009 'date-entered-on-file',
1010 'Date-of-acquisition',
1011 'Date-of-publication',
1012 'Date-time-last-modified',
1013 'Dewey-classification',
1014 'Dissertation-information',
1023 'Geographic-subdivision',
1026 'Heading-use-main-or-added-entry',
1027 'Heading-use-series-added-entry ',
1028 'Heading-use-subject-added-entry',
1032 'Identifier-standard',
1033 'Illustration-code',
1035 'Index-term-uncontrolled',
1036 'Interest-age-level',
1037 'Interest-grade-level',
1047 'language-original',
1057 'Local-classification',
1060 'Match-heading-see-from',
1068 'Name-geographic-heading',
1069 'Name-geographic-see',
1070 'Name-geographic-seealso',
1076 'Other-control-number',
1079 'Personal-name-heading',
1080 'Personal-name-see',
1081 'Personal-name-seealso',
1083 'Place-publication',
1090 'Reading-grade-level',
1091 'Record-control-number',
1102 'Subject-heading-thesaurus',
1103 'Subject-name-personal',
1104 'Subject-subdivision',
1113 'Term-genre-form-heading',
1114 'Term-genre-form-see',
1115 'Term-genre-form-seealso',
1121 'Title-uniform-heading',
1122 'Title-uniform-see',
1123 'Title-uniform-seealso',
1133 'classification-source',
1135 'coded-location-qualifier',
1146 'Local-classification',
1149 'materials-specified',
1154 'Number-local-acquisition',
1159 'replacementpricedate',
1177 $simple_query, $query_cgi,
1178 $query_desc, $limit,
1179 $limit_cgi, $limit_desc,
1180 $query_type ) = buildQuery ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1182 Build queries and limits in CCL, CGI, Human,
1183 handle truncation, stemming, field weighting, fuzziness, etc.
1185 See verbose embedded documentation.
1191 my ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang) = @_;
1196 my @operators = $operators ? @$operators : ();
1197 my @indexes = $indexes ? @$indexes : ();
1198 my @operands = $operands ? @$operands : ();
1199 my @limits = $limits ? @$limits : ();
1200 my @sort_by = $sort_by ? @$sort_by : ();
1202 my $stemming = C4::Context->preference("QueryStemming") || 0;
1203 my $auto_truncation = C4::Context->preference("QueryAutoTruncate") || 0;
1204 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
1205 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
1207 my $query = $operands[0] // "";
1208 my $simple_query = $operands[0];
1210 # initialize the variables we're passing back
1219 my $cclindexes = getIndexes();
1220 if ( $query !~ /\s*(ccl=|pqf=|cql=)/ ) {
1221 while ( !$cclq && $query =~ /(?:^|\W)([\w-]+)(,[\w-]+)*[:=]/g ) {
1223 $cclq = grep { lc($_) eq $dx } @$cclindexes;
1225 $query = "ccl=$query" if $cclq;
1228 # for handling ccl, cql, pqf queries in diagnostic mode, skip the rest of the steps
1230 if ( $query =~ /^ccl=/ ) {
1232 # This is needed otherwise ccl= and &limit won't work together, and
1233 # this happens when selecting a subject on the opac-detail page
1234 @limits = grep {!/^$/} @limits;
1235 my $original_q = $q; # without available part
1236 unless ( grep { $_ eq 'available' } @limits ) {
1237 $q =~ s| and \( \(allrecords,AlwaysMatches=''\) and \(not-onloan-count,st-numeric >= 1\) and \(lost,st-numeric=0\) \)||;
1241 if ( grep { $_ eq 'available' } @limits ) {
1242 $q .= q| and ( (allrecords,AlwaysMatches='') and (not-onloan-count,st-numeric >= 1) and (lost,st-numeric=0) )|;
1243 @limits = grep {!/^available$/} @limits;
1245 $q .= ' and '.join(' and ', @limits) if @limits;
1247 return ( undef, $q, $q, "q=ccl=".uri_escape_utf8($q), $original_q, '', '', '', 'ccl' );
1249 if ( $query =~ /^cql=/ ) {
1250 return ( undef, $', $', "q=cql=".uri_escape_utf8($'), $', '', '', '', 'cql' );
1252 if ( $query =~ /^pqf=/ ) {
1254 $query_cgi = "q=pqf=".uri_escape_utf8($');
1255 return ( undef, $', $', $query_cgi, $query_desc, '', '', '', 'pqf' );
1258 # pass nested queries directly
1259 # FIXME: need better handling of some of these variables in this case
1260 # Nested queries aren't handled well and this implementation is flawed and causes users to be
1261 # unable to search for anything containing () commenting out, will be rewritten for 3.4.0
1262 # if ( $query =~ /(\(|\))/ ) {
1264 # undef, $query, $simple_query, $query_cgi,
1265 # $query, $limit, $limit_cgi, $limit_desc,
1270 # Form-based queries are non-nested and fixed depth, so we can easily modify the incoming
1271 # query operands and indexes and add stemming, truncation, field weighting, etc.
1272 # Once we do so, we'll end up with a value in $query, just like if we had an
1273 # incoming $query from the user
1276 ; # clear it out so we can populate properly with field-weighted, stemmed, etc. query
1277 my $previous_operand
1278 ; # a flag used to keep track if there was a previous query
1279 # if there was, we can apply the current operator
1281 for ( my $i = 0 ; $i <= @operands ; $i++ ) {
1283 # COMBINE OPERANDS, INDEXES AND OPERATORS
1284 if ( ($operands[$i] // '') ne '' ) {
1285 $operands[$i]=~s/^\s+//;
1287 # A flag to determine whether or not to add the index to the query
1290 # If the user is sophisticated enough to specify an index, turn off field weighting, and stemming handling
1291 if ( $operands[$i] =~ /\w(:|=)/ || $scan ) {
1295 $operands[$i] =~ s/\?/{?}/g; # need to escape question marks
1297 my $operand = $operands[$i];
1298 my $index = $indexes[$i] || 'kw';
1300 # Add index-specific attributes
1302 #Afaik, this 'yr' condition will only ever be met in the staff interface advanced search
1303 #for "Publication date", since typing 'yr:YYYY' into the search box produces a CCL query,
1304 #which is processed higher up in this sub. Other than that, year searches are typically
1305 #handled as limits which are not processed her either.
1307 # Search ranges: Date of Publication, st-numeric
1308 if ( $index =~ /(yr|st-numeric)/ ) {
1309 #weight_fields/relevance search causes errors with date ranges
1310 #In the case of YYYY-, it will only return records with a 'yr' of YYYY (not the range)
1311 #In the case of YYYY-YYYY, it will return no results
1312 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = 0;
1315 # Date of Acquisition
1316 elsif ( $index =~ /acqdate/ ) {
1317 #stemming and auto_truncation would have zero impact since it already is YYYY-MM-DD format
1318 #Weight_fields probably SHOULD be turned OFF, otherwise you'll get records floating to the
1319 #top of the results just because they have lots of item records matching that date.
1320 #Fuzzy actually only applies during _build_weighted_query, and is reset there anyway, so
1322 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = 0;
1324 # ISBN,ISSN,Standard Number, don't need special treatment
1325 elsif ( $index eq 'nb' || $index eq 'ns' || $index eq 'hi' ) {
1327 $stemming, $auto_truncation,
1328 $weight_fields, $fuzzy_enabled
1331 if ( $index eq 'nb' ) {
1332 if ( C4::Context->preference("SearchWithISBNVariations") ) {
1333 my @isbns = C4::Koha::GetVariationsOfISBN( $operand );
1334 $operands[$i] = $operand = '(nb=' . join(' OR nb=', @isbns) . ')';
1335 $indexes[$i] = $index = 'kw';
1340 # Set default structure attribute (word list)
1341 my $struct_attr = q{};
1342 unless ( $indexes_set || $index =~ /,(st-|phr|ext|wrdl)/ || $index =~ /^(nb|ns)$/ ) {
1343 $struct_attr = ",wrdl";
1346 # Some helpful index variants
1347 my $index_plus = $index . $struct_attr . ':';
1348 my $index_plus_comma = $index . $struct_attr . ',';
1350 if ($auto_truncation){
1351 unless ( $index =~ /,(st-|phr|ext)/ ) {
1352 #FIXME only valid with LTR scripts
1353 $operand=join(" ",map{
1354 (index($_,"*")>0?"$_":"$_*")
1355 }split (/\s+/,$operand));
1360 my $truncated_operand = q{};
1361 my( $nontruncated, $righttruncated, $lefttruncated,
1362 $rightlefttruncated, $regexpr
1363 ) = _detect_truncation( $operand, $index );
1365 Koha::Logger->get->debug(
1366 "TRUNCATION: NON:>@$nontruncated< RIGHT:>@$righttruncated< LEFT:>@$lefttruncated< RIGHTLEFT:>@$rightlefttruncated< REGEX:>@$regexpr<");
1370 scalar(@$righttruncated) + scalar(@$lefttruncated) +
1371 scalar(@$rightlefttruncated) > 0 )
1374 # Don't field weight or add the index to the query, we do it here
1376 undef $weight_fields;
1377 my $previous_truncation_operand;
1378 if (scalar @$nontruncated) {
1379 $truncated_operand .= "$index_plus @$nontruncated ";
1380 $previous_truncation_operand = 1;
1382 if (scalar @$righttruncated) {
1383 $truncated_operand .= "and " if $previous_truncation_operand;
1384 $truncated_operand .= $index_plus_comma . "rtrn:@$righttruncated ";
1385 $previous_truncation_operand = 1;
1387 if (scalar @$lefttruncated) {
1388 $truncated_operand .= "and " if $previous_truncation_operand;
1389 $truncated_operand .= $index_plus_comma . "ltrn:@$lefttruncated ";
1390 $previous_truncation_operand = 1;
1392 if (scalar @$rightlefttruncated) {
1393 $truncated_operand .= "and " if $previous_truncation_operand;
1394 $truncated_operand .= $index_plus_comma . "rltrn:@$rightlefttruncated ";
1395 $previous_truncation_operand = 1;
1398 $operand = $truncated_operand if $truncated_operand;
1399 Koha::Logger->get->debug("TRUNCATED OPERAND: >$truncated_operand<");
1402 my $stemmed_operand = q{};
1403 $stemmed_operand = _build_stemmed_operand($operand, $lang)
1406 Koha::Logger->get->debug("STEMMED OPERAND: >$stemmed_operand<");
1408 # Handle Field Weighting
1409 my $weighted_operand = q{};
1410 if ($weight_fields) {
1411 $weighted_operand = _build_weighted_query( $operand, $stemmed_operand, $index );
1412 $operand = $weighted_operand;
1416 Koha::Logger->get->debug("FIELD WEIGHTED OPERAND: >$weighted_operand<");
1418 #Use relevance ranking when not using a weighted query (which adds relevance ranking of its own)
1420 #N.B. Truncation is mutually exclusive with Weighted Queries,
1421 #so even if QueryWeightFields is turned on, QueryAutoTruncate will turn it off, thus
1422 #the need for this relevance wrapper.
1423 $operand = "(rk=($operand))" unless $weight_fields;
1425 ($query,$query_cgi,$query_desc,$previous_operand) = _build_initial_query({
1427 query_cgi => $query_cgi,
1428 query_desc => $query_desc,
1429 operator => ($operators[ $i - 1 ]) ? $operators[ $i - 1 ] : '',
1430 parsed_operand => $operand,
1431 original_operand => $operands[$i] // '',
1433 index_plus => $index_plus,
1434 indexes_set => $indexes_set,
1435 previous_operand => $previous_operand,
1441 Koha::Logger->get->debug("QUERY BEFORE LIMITS: >$query<");
1444 my %group_OR_limits;
1445 my $availability_limit;
1446 foreach my $this_limit (@limits) {
1447 next unless $this_limit;
1448 if ( $this_limit =~ /available/ ) {
1450 ## 'available' is defined as (items.onloan is NULL) and (items.itemlost = 0)
1452 ## all records not indexed in the onloan register (zebra) and all records with a value of lost equal to 0
1453 $availability_limit .=
1454 "( (allrecords,AlwaysMatches='') and (not-onloan-count,st-numeric >= 1) and (lost,st-numeric=0) )";
1455 $limit_cgi .= "&limit=available";
1459 # group_OR_limits, prefixed by mc-
1460 # OR every member of the group
1461 elsif ( $this_limit =~ /mc/ ) {
1462 my ($k,$v) = split(/:/, $this_limit,2);
1463 if ( $k !~ /mc-i(tem)?type/ ) {
1464 # in case the mc-ccode value has complicating chars like ()'s inside it we wrap in quotes
1465 $this_limit =~ tr/"//d;
1466 $this_limit = $k.':"'.$v.'"';
1469 $group_OR_limits{$k} .= " or " if $group_OR_limits{$k};
1470 $limit_desc .= " or " if $group_OR_limits{$k};
1471 $group_OR_limits{$k} .= "$this_limit";
1472 $limit_cgi .= "&limit=" . uri_escape_utf8($this_limit);
1473 $limit_desc .= " $this_limit";
1476 # Regular old limits
1478 $limit .= " and " if $limit || $query;
1479 $limit .= "$this_limit";
1480 $limit_cgi .= "&limit=" . uri_escape_utf8($this_limit);
1481 if ($this_limit =~ /^branch:(.+)/) {
1482 my $branchcode = $1;
1483 my $library = Koha::Libraries->find( $branchcode );
1484 if (defined $library) {
1485 $limit_desc .= " branch:" . $library->branchname;
1487 $limit_desc .= " $this_limit";
1490 $limit_desc .= " $this_limit";
1494 foreach my $k (keys (%group_OR_limits)) {
1495 $limit .= " and " if ( $query || $limit );
1496 $limit .= "($group_OR_limits{$k})";
1498 if ($availability_limit) {
1499 $limit .= " and " if ( $query || $limit );
1500 $limit .= "($availability_limit)";
1503 # Normalize the query and limit strings
1504 # This is flawed , means we can't search anything with : in it
1505 # if user wants to do ccl or cql, start the query with that
1506 # $query =~ s/:/=/g;
1507 #NOTE: We use several several different regexps here as you can't have variable length lookback assertions
1508 $query =~ s/(?<=(ti|au|pb|su|an|kw|mc|nb|ns)):/=/g;
1509 $query =~ s/(?<=(wrdl)):/=/g;
1510 $query =~ s/(?<=(trn|phr)):/=/g;
1511 $query =~ s/(?<=(st-numeric)):/=/g;
1512 $query =~ s/(?<=(st-year)):/=/g;
1513 $query =~ s/(?<=(st-date-normalized)):/=/g;
1515 # Removing warnings for later substitutions
1517 $query_desc //= q{};
1520 $limit_desc //= q{};
1522 $simple_query //= q{};
1524 for ( $query, $query_desc, $limit, $limit_desc ) {
1525 s/ +/ /g; # remove extra spaces
1526 s/^ //g; # remove any beginning spaces
1527 s/ $//g; # remove any ending spaces
1528 s/==/=/g; # remove double == from query
1530 $query_cgi =~ s/^&//; # remove unnecessary & from beginning of the query cgi
1532 for ($query_cgi,$simple_query) {
1535 # append the limit to the query
1536 $query .= " " . $limit;
1538 Koha::Logger->get->debug(
1539 sprintf "buildQuery returns\nQUERY:%s\nQUERY CGI:%s\nQUERY DESC:%s\nLIMIT:%s\nLIMIT CGI:%s\nLIMIT DESC:%s",
1540 $query, $query_cgi, $query_desc, $limit, $limit_cgi, $limit_desc );
1543 undef, $query, $simple_query, $query_cgi,
1544 $query_desc, $limit, $limit_cgi, $limit_desc,
1549 =head2 _build_initial_query
1551 ($query, $query_cgi, $query_desc, $previous_operand) = _build_initial_query($initial_query_params);
1553 Build a section of the initial query containing indexes, operators, and operands.
1557 sub _build_initial_query {
1561 if ($params->{previous_operand}){
1562 #If there is a previous operand, add a supplied operator or the default 'and'
1563 $operator = ($params->{operator}) ? " ".($params->{operator})." " : ' and ';
1566 #NOTE: indexes_set is typically set when doing truncation or field weighting
1567 my $operand = ($params->{indexes_set}) ? $params->{parsed_operand} : $params->{index_plus}.$params->{parsed_operand};
1569 #e.g. "kw,wrdl:test"
1570 #e.g. " and kw,wrdl:test"
1571 $params->{query} .= $operator . $operand;
1573 $params->{query_cgi} .= "&op=".uri_escape_utf8($operator) if $operator;
1574 $params->{query_cgi} .= "&idx=".uri_escape_utf8($params->{index}) if $params->{index};
1575 $params->{query_cgi} .= "&q=".uri_escape_utf8($params->{original_operand}) if $params->{original_operand};
1577 #e.g. " and kw,wrdl: test"
1578 $params->{query_desc} .= $operator . ( $params->{index_plus} // q{} ) . " " . ( $params->{original_operand} // q{} );
1580 $params->{previous_operand} = 1 unless $params->{previous_operand}; #If there is no previous operand, mark this as one
1582 return ($params->{query}, $params->{query_cgi}, $params->{query_desc}, $params->{previous_operand});
1585 =head2 searchResults
1587 my @search_results = searchResults($search_context, $searchdesc, $hits,
1588 $results_per_page, $offset, $scan,
1591 Format results in a form suitable for passing to the template
1595 # IMO this subroutine is pretty messy still -- it's responsible for
1596 # building the HTML output for the template
1598 my ( $search_context, $searchdesc, $hits, $results_per_page, $offset, $scan, $marcresults, $xslt_variables ) = @_;
1599 my $dbh = C4::Context->dbh;
1604 $search_context->{'interface'} = 'opac' if !$search_context->{'interface'} || $search_context->{'interface'} ne 'intranet';
1605 my ($is_opac, $hidelostitems);
1606 if ($search_context->{'interface'} eq 'opac') {
1607 $hidelostitems = C4::Context->preference('hidelostitems');
1611 my $record_processor = Koha::RecordProcessor->new({
1612 filters => 'ViewPolicy'
1615 #Build branchnames hash
1616 my %branches = map { $_->branchcode => $_->branchname } Koha::Libraries->search({}, { order_by => 'branchname' });
1618 # FIXME - We build an authorised values hash here, using the default framework
1619 # though it is possible to have different authvals for different fws.
1621 my $shelflocations =
1622 { map { $_->{authorised_value} => $_->{lib} } Koha::AuthorisedValues->get_descriptions_by_koha_field( { frameworkcode => '', kohafield => 'items.location' } ) };
1624 # get notforloan authorised value list (see $shelflocations FIXME)
1625 my $av = Koha::MarcSubfieldStructures->search({ frameworkcode => '', kohafield => 'items.notforloan', authorised_value => [ -and => {'!=' => undef }, {'!=' => ''}] });
1626 my $notforloan_authorised_value = $av->count ? $av->next->authorised_value : undef;
1629 my $itemtypes = Koha::ItemTypes->search_with_localization;
1630 my %itemtypes = map { $_->{itemtype} => $_ } @{ $itemtypes->unblessed };
1632 #search item field code
1633 my ($itemtag, undef) = &GetMarcFromKohaField( "items.itemnumber" );
1635 ## find column names of items related to MARC
1636 my %subfieldstosearch;
1637 my @columns = Koha::Database->new()->schema()->resultset('Item')->result_source->columns;
1638 for my $column ( @columns ) {
1639 my ( $tagfield, $tagsubfield ) =
1640 &GetMarcFromKohaField( "items." . $column );
1641 if ( defined $tagsubfield ) {
1642 $subfieldstosearch{$column} = $tagsubfield;
1646 # handle which records to actually retrieve
1647 my $times; # Times is which record to process up to
1648 if ( $hits && $offset + $results_per_page <= $hits ) {
1649 $times = $offset + $results_per_page;
1652 $times = $hits; # If less hits than results_per_page+offset we go to the end
1655 my $marcflavour = C4::Context->preference("marcflavour");
1656 # We get the biblionumber position in MARC
1657 my ($bibliotag,$bibliosubf)=GetMarcFromKohaField( 'biblio.biblionumber' );
1659 # set stuff for XSLT processing here once, not later again for every record we retrieved
1661 my $userenv = C4::Context->userenv;
1663 = ( defined $userenv and $userenv->{number} )
1664 ? Koha::Patrons->find( $userenv->{number} )
1666 my $patron_category_hide_lost_items = ($logged_in_user) ? $logged_in_user->category->hidelostitems : 0;
1668 # loop through all of the records we've retrieved
1669 for ( my $i = $offset ; $i <= $times - 1 ; $i++ ) {
1673 # For Scan searches we built USMARC data
1674 $marcrecord = MARC::Record->new_from_usmarc( $marcresults->[$i]);
1676 # Normal search, render from Zebra's output
1677 $marcrecord = new_record_from_zebra(
1682 if ( ! defined $marcrecord ) {
1683 warn "ERROR DECODING RECORD - $@: " . $marcresults->[$i];
1691 ? GetFrameworkCode($marcrecord->field($bibliotag)->data)
1692 : GetFrameworkCode($marcrecord->subfield($bibliotag,$bibliosubf));
1694 SetUTF8Flag($marcrecord);
1695 my $oldbiblio = TransformMarcToKoha( $marcrecord, $fw, 'no_items' );
1696 $oldbiblio->{result_number} = $i + 1;
1698 $oldbiblio->{normalized_upc} = GetNormalizedUPC( $marcrecord,$marcflavour);
1699 $oldbiblio->{normalized_ean} = GetNormalizedEAN( $marcrecord,$marcflavour);
1700 $oldbiblio->{normalized_oclc} = GetNormalizedOCLCNumber($marcrecord,$marcflavour);
1701 $oldbiblio->{normalized_isbn} = GetNormalizedISBN($oldbiblio->{isbn},$marcrecord,$marcflavour); # Use existing ISBN from record if we got one
1702 $oldbiblio->{content_identifier_exists} = 1 if ($oldbiblio->{normalized_isbn} or $oldbiblio->{normalized_oclc} or $oldbiblio->{normalized_ean} or $oldbiblio->{normalized_upc});
1704 # edition information, if any
1705 $oldbiblio->{edition} = $oldbiblio->{editionstatement};
1707 my $itemtype = $oldbiblio->{itemtype} ? $itemtypes{$oldbiblio->{itemtype}} : undef;
1708 # add imageurl to itemtype if there is one
1709 $oldbiblio->{imageurl} = $itemtype ? getitemtypeimagelocation( $search_context->{'interface'}, $itemtype->{imageurl} ) : q{};
1710 # Build summary if there is one (the summary is defined in the itemtypes table)
1711 $oldbiblio->{description} = $itemtype ? $itemtype->{translated_description} : q{};
1713 # Pull out the items fields
1714 my @fields = $marcrecord->field($itemtag);
1715 my $marcflavor = C4::Context->preference("marcflavour");
1717 # adding linked items that belong to host records
1718 if ( C4::Context->preference('EasyAnalyticalRecords') ) {
1719 my $analyticsfield = '773';
1720 if ($marcflavor eq 'MARC21' || $marcflavor eq 'NORMARC') {
1721 $analyticsfield = '773';
1722 } elsif ($marcflavor eq 'UNIMARC') {
1723 $analyticsfield = '461';
1725 foreach my $hostfield ( $marcrecord->field($analyticsfield)) {
1726 my $hostbiblionumber = $hostfield->subfield("0");
1727 my $linkeditemnumber = $hostfield->subfield("9");
1728 if( $hostbiblionumber ) {
1729 my $linkeditemmarc = C4::Items::GetMarcItem( $hostbiblionumber, $linkeditemnumber );
1730 if ($linkeditemmarc) {
1731 my $linkeditemfield = $linkeditemmarc->field($itemtag);
1732 if ($linkeditemfield) {
1733 push( @fields, $linkeditemfield );
1740 # Setting item statuses for display
1741 my @available_items_loop;
1742 my @onloan_items_loop;
1743 my @other_items_loop;
1745 my $available_items;
1749 my $ordered_count = 0;
1750 my $available_count = 0;
1751 my $onloan_count = 0;
1752 my $longoverdue_count = 0;
1753 my $other_count = 0;
1754 my $withdrawn_count = 0;
1755 my $itemlost_count = 0;
1756 my $hideatopac_count = 0;
1757 my $itembinding_count = 0;
1758 my $itemdamaged_count = 0;
1759 my $item_in_transit_count = 0;
1760 my $can_place_holds = 0;
1761 my $item_onhold_count = 0;
1762 my $notforloan_count = 0;
1763 my $items_count = scalar(@fields);
1764 my $maxitems_pref = C4::Context->preference('maxItemsinSearchResults');
1765 my $maxitems = $maxitems_pref ? $maxitems_pref - 1 : 1;
1766 my @hiddenitems; # hidden itemnumbers based on OpacHiddenItems syspref
1768 # loop through every item
1769 foreach my $field (@fields) {
1772 # populate the items hash
1773 foreach my $code ( keys %subfieldstosearch ) {
1774 $item->{$code} = $field->subfield( $subfieldstosearch{$code} );
1776 $item->{description} = $itemtypes{ $item->{itype} }{translated_description} if $item->{itype};
1780 # hidden because lost
1781 if ($hidelostitems && $item->{itemlost}) {
1782 $hideatopac_count++;
1785 # hidden based on OpacHiddenItems syspref
1786 my @hi = C4::Items::GetHiddenItemnumbers({ items=> [ $item ], borcat => $search_context->{category} });
1788 push @hiddenitems, @hi;
1789 $hideatopac_count++;
1794 my $hbranch = C4::Context->preference('StaffSearchResultsDisplayBranch');
1795 my $otherbranch = $hbranch eq 'homebranch' ? 'holdingbranch' : 'homebranch';
1797 # set item's branch name, use HomeOrHoldingBranch syspref first, fall back to the other one
1798 if ($item->{$hbranch}) {
1799 $item->{'branchname'} = $branches{$item->{$hbranch}};
1801 elsif ($item->{$otherbranch}) { # Last resort
1802 $item->{'branchname'} = $branches{$item->{$otherbranch}};
1806 ( $item->{$hbranch} ? $item->{$hbranch} . '--' : q{} )
1807 . ( $item->{location} ? $item->{location} : q{} )
1808 . ( $item->{itype} ? $item->{itype} : q{} )
1809 . ( $item->{itemcallnumber} ? $item->{itemcallnumber} : q{} );
1810 # For each grouping of items (onloan, available, unavailable), we build a key to store relevant info about that item
1811 if ( $item->{onloan}
1813 and !( $patron_category_hide_lost_items and $item->{itemlost} ) )
1816 my $key = $prefix . $item->{onloan} . $item->{barcode};
1817 $onloan_items->{$key}->{due_date} = $item->{onloan};
1818 $onloan_items->{$key}->{count}++ if $item->{$hbranch};
1819 $onloan_items->{$key}->{branchname} = $item->{branchname};
1820 $onloan_items->{$key}->{location} = $shelflocations->{ $item->{location} } if $item->{location};
1821 $onloan_items->{$key}->{itemcallnumber} = $item->{itemcallnumber};
1822 $onloan_items->{$key}->{description} = $item->{description};
1823 $onloan_items->{$key}->{imageurl} =
1824 getitemtypeimagelocation( $search_context->{'interface'}, $itemtypes{ $item->{itype} }->{imageurl} );
1826 # if something's checked out and lost, mark it as 'long overdue'
1827 if ( $item->{itemlost} ) {
1828 $onloan_items->{$key}->{longoverdue}++;
1829 $longoverdue_count++;
1831 else { # can place holds as long as item isn't lost
1832 $can_place_holds = 1;
1836 # items not on loan, but still unavailable ( lost, withdrawn, damaged )
1839 my $itemtype = C4::Context->preference("item-level_itypes")? $item->{itype}: $oldbiblio->{itemtype};
1840 $item->{notforloan} = 1 if !$item->{notforloan} &&
1841 $itemtype && $itemtypes{ $itemtype }->{notforloan};
1844 if ( $item->{notforloan} < 0 ) {
1846 } elsif ( $item->{notforloan} > 0 ) {
1847 $notforloan_count++;
1850 # is item in transit?
1851 my $transfertwhen = '';
1852 my ($transfertfrom, $transfertto);
1854 # is item on the reserve shelf?
1855 my $reservestatus = '';
1857 unless ($item->{withdrawn}
1858 || $item->{itemlost}
1860 || $item->{notforloan}
1861 || ( C4::Context->preference('MaxSearchResultsItemsPerRecordStatusCheck')
1862 && $items_count > C4::Context->preference('MaxSearchResultsItemsPerRecordStatusCheck') ) ) {
1864 # A couple heuristics to limit how many times
1865 # we query the database for item transfer information, sacrificing
1866 # accuracy in some cases for speed;
1868 # 1. don't query if item has one of the other statuses
1869 # 2. don't check transit status if the bib has
1870 # more than 20 items
1872 # FIXME: to avoid having the query the database like this, and to make
1873 # the in transit status count as unavailable for search limiting,
1874 # should map transit status to record indexed in Zebra.
1876 ($transfertwhen, $transfertfrom, $transfertto) = C4::Circulation::GetTransfers($item->{itemnumber});
1877 $reservestatus = C4::Reserves::GetReserveStatus( $item->{itemnumber} );
1880 # item is withdrawn, lost, damaged, not for loan, reserved or in transit
1881 if ( $item->{withdrawn}
1882 || $item->{itemlost}
1884 || $item->{notforloan}
1885 || $reservestatus eq 'Waiting'
1886 || ($transfertwhen && $transfertwhen ne ''))
1888 $withdrawn_count++ if $item->{withdrawn};
1889 $itemlost_count++ if $item->{itemlost};
1890 $itemdamaged_count++ if $item->{damaged};
1891 $item_in_transit_count++ if $transfertwhen && $transfertwhen ne '';
1892 $item_onhold_count++ if $reservestatus eq 'Waiting';
1893 $item->{status} = ($item->{withdrawn}//q{}) . "-" . ($item->{itemlost}//q{}) . "-" . ($item->{damaged}//q{}) . "-" . ($item->{notforloan}//q{});
1895 # can place a hold on a item if
1896 # not lost nor withdrawn
1897 # not damaged unless AllowHoldsOnDamagedItems is true
1898 # item is either for loan or on order (notforloan < 0)
1899 $can_place_holds = 1
1902 && !$item->{withdrawn}
1903 && ( !$item->{damaged} || C4::Context->preference('AllowHoldsOnDamagedItems') )
1904 && ( !$item->{notforloan} || $item->{notforloan} < 0 )
1909 my $key = $prefix . $item->{status};
1910 foreach (qw(withdrawn itemlost damaged branchname itemcallnumber)) {
1911 $other_items->{$key}->{$_} = $item->{$_};
1913 $other_items->{$key}->{intransit} = ( $transfertwhen ne '' ) ? 1 : 0;
1914 $other_items->{$key}->{onhold} = ($reservestatus) ? 1 : 0;
1915 $other_items->{$key}->{notforloan} = GetAuthorisedValueDesc('','',$item->{notforloan},'','',$notforloan_authorised_value) if $notforloan_authorised_value and $item->{notforloan};
1916 $other_items->{$key}->{count}++ if $item->{$hbranch};
1917 $other_items->{$key}->{location} = $shelflocations->{ $item->{location} } if $item->{location};
1918 $other_items->{$key}->{description} = $item->{description};
1919 $other_items->{$key}->{imageurl} = getitemtypeimagelocation( $search_context->{'interface'}, $itemtypes{ $item->{itype}//q{} }->{imageurl} );
1923 $can_place_holds = 1;
1925 $available_items->{$prefix}->{count}++ if $item->{$hbranch};
1926 foreach (qw(branchname itemcallnumber description)) {
1927 $available_items->{$prefix}->{$_} = $item->{$_};
1929 $available_items->{$prefix}->{location} = $shelflocations->{ $item->{location} } if $item->{location};
1930 $available_items->{$prefix}->{imageurl} = getitemtypeimagelocation( $search_context->{'interface'}, $itemtypes{ $item->{itype}//q{} }->{imageurl} );
1933 } # notforloan, item level and biblioitem level
1935 # if all items are hidden, do not show the record
1936 if ( C4::Context->preference('OpacHiddenItemsHidesRecord') && $items_count > 0 && $hideatopac_count == $items_count) {
1940 my ( $availableitemscount, $onloanitemscount, $otheritemscount );
1941 for my $key ( sort keys %$onloan_items ) {
1942 (++$onloanitemscount > $maxitems) and last;
1943 push @onloan_items_loop, $onloan_items->{$key};
1945 for my $key ( sort keys %$other_items ) {
1946 (++$otheritemscount > $maxitems) and last;
1947 push @other_items_loop, $other_items->{$key};
1949 for my $key ( sort keys %$available_items ) {
1950 (++$availableitemscount > $maxitems) and last;
1951 push @available_items_loop, $available_items->{$key}
1954 # XSLT processing of some stuff
1955 # we fetched the sysprefs already before the loop through all retrieved record!
1957 $record_processor->options({
1958 frameworkcode => $fw,
1959 interface => $search_context->{'interface'}
1962 $record_processor->process($marcrecord);
1964 $oldbiblio->{XSLTResultsRecord} = XSLTParse4Display(
1966 biblionumber => $oldbiblio->{biblionumber},
1967 record => $marcrecord,
1970 ? 'OPACXSLTResultsDisplay'
1971 : 'XSLTResultsDisplay'
1974 hidden_items => \@hiddenitems,
1975 xslt_variables => $xslt_variables
1980 # if biblio level itypes are used and itemtype is notforloan, it can't be reserved either
1981 if (!C4::Context->preference("item-level_itypes")) {
1982 if ($itemtype && $itemtype->{notforloan}) {
1983 $can_place_holds = 0;
1986 $oldbiblio->{norequests} = 1 unless $can_place_holds;
1987 $oldbiblio->{items_count} = $items_count;
1988 $oldbiblio->{available_items_loop} = \@available_items_loop;
1989 $oldbiblio->{onloan_items_loop} = \@onloan_items_loop;
1990 $oldbiblio->{other_items_loop} = \@other_items_loop;
1991 $oldbiblio->{availablecount} = $available_count;
1992 $oldbiblio->{availableplural} = 1 if $available_count > 1;
1993 $oldbiblio->{onloancount} = $onloan_count;
1994 $oldbiblio->{onloanplural} = 1 if $onloan_count > 1;
1995 $oldbiblio->{othercount} = $other_count;
1996 $oldbiblio->{otherplural} = 1 if $other_count > 1;
1997 $oldbiblio->{withdrawncount} = $withdrawn_count;
1998 $oldbiblio->{itemlostcount} = $itemlost_count;
1999 $oldbiblio->{damagedcount} = $itemdamaged_count;
2000 $oldbiblio->{intransitcount} = $item_in_transit_count;
2001 $oldbiblio->{onholdcount} = $item_onhold_count;
2002 $oldbiblio->{orderedcount} = $ordered_count;
2003 $oldbiblio->{notforloancount} = $notforloan_count;
2005 if (C4::Context->preference("AlternateHoldingsField") && $items_count == 0) {
2006 my $fieldspec = C4::Context->preference("AlternateHoldingsField");
2007 my $subfields = substr $fieldspec, 3;
2008 my $holdingsep = C4::Context->preference("AlternateHoldingsSeparator") || ' ';
2009 my @alternateholdingsinfo = ();
2010 my @holdingsfields = $marcrecord->field(substr $fieldspec, 0, 3);
2011 my $alternateholdingscount = 0;
2013 for my $field (@holdingsfields) {
2014 my %holding = ( holding => '' );
2015 my $havesubfield = 0;
2016 for my $subfield ($field->subfields()) {
2017 if ((index $subfields, $$subfield[0]) >= 0) {
2018 $holding{'holding'} .= $holdingsep if (length $holding{'holding'} > 0);
2019 $holding{'holding'} .= $$subfield[1];
2023 if ($havesubfield) {
2024 push(@alternateholdingsinfo, \%holding);
2025 $alternateholdingscount++;
2029 $oldbiblio->{'ALTERNATEHOLDINGS'} = \@alternateholdingsinfo;
2030 $oldbiblio->{'alternateholdings_count'} = $alternateholdingscount;
2033 $oldbiblio->{biblio_object} = Koha::Biblios->find( $oldbiblio->{biblionumber} );
2035 push( @newresults, $oldbiblio );
2041 =head2 enabled_staff_search_views
2043 %hash = enabled_staff_search_views()
2045 This function returns a hash that contains three flags obtained from the system
2046 preferences, used to determine whether a particular staff search results view
2051 =item C<Output arg:>
2053 * $hash{can_view_MARC} is true only if the MARC view is enabled
2054 * $hash{can_view_ISBD} is true only if the ISBD view is enabled
2055 * $hash{can_view_labeledMARC} is true only if the Labeled MARC view is enabled
2057 =item C<usage in the script:>
2061 $template->param ( C4::Search::enabled_staff_search_views );
2065 sub enabled_staff_search_views
2068 can_view_MARC => C4::Context->preference('viewMARC'), # 1 if the staff search allows the MARC view
2069 can_view_ISBD => C4::Context->preference('viewISBD'), # 1 if the staff search allows the ISBD view
2070 can_view_labeledMARC => C4::Context->preference('viewLabeledMARC'), # 1 if the staff search allows the Labeled MARC view
2074 =head2 z3950_search_args
2076 $arrayref = z3950_search_args($matchpoints)
2078 This function returns an array reference that contains the search parameters to be
2079 passed to the Z39.50 search script (z3950_search.pl). The array elements
2080 are hash refs whose keys are name and value, and whose values are the
2081 name of a search parameter, the value of that search parameter and the URL encoded
2082 value of that parameter.
2084 The search parameter names are lccn, isbn, issn, title, author, dewey and subject.
2086 The search parameter values are obtained from the bibliographic record whose
2087 data is in a hash reference in $matchpoints, as returned by Biblio::GetBiblioData().
2089 If $matchpoints is a scalar, it is assumed to be an unnamed query descriptor, e.g.
2090 a general purpose search argument. In this case, the returned array contains only
2091 entry: the key is 'title' and the value is derived from $matchpoints.
2093 If a search parameter value is undefined or empty, it is not included in the returned
2096 The returned array reference may be passed directly to the template parameters.
2100 =item C<Output arg:>
2102 * $array containing hash refs as described above
2104 =item C<usage in the script:>
2108 $data = Biblio::GetBiblioData($bibno);
2109 $template->param ( MYLOOP => C4::Search::z3950_search_args($data) )
2113 $template->param ( MYLOOP => C4::Search::z3950_search_args($searchscalar) )
2117 sub z3950_search_args {
2120 my $isbn_string = ref( $bibrec ) ? $bibrec->{title} : $bibrec;
2121 my $isbn = Business::ISBN->new( $isbn_string );
2123 if (defined $isbn && $isbn->is_valid)
2125 if ( ref($bibrec) ) {
2126 $bibrec->{isbn} = $isbn_string;
2127 $bibrec->{title} = undef;
2129 $bibrec = { isbn => $isbn_string };
2133 $bibrec = { title => $bibrec } if !ref $bibrec;
2136 for my $field (qw/ lccn isbn issn title author dewey subject /)
2138 push @$array, { name => $field, value => $bibrec->{$field} }
2139 if defined $bibrec->{$field};
2144 =head2 GetDistinctValues($field);
2146 C<$field> is a reference to the fields array
2150 sub GetDistinctValues {
2151 my ($fieldname,$string)=@_;
2152 # returns a reference to a hash of references to branches...
2153 if ($fieldname=~/\./){
2154 my ($table,$column)=split /\./, $fieldname;
2155 my $dbh = C4::Context->dbh;
2156 my $sth = $dbh->prepare("select DISTINCT($column) as value, count(*) as cnt from $table ".($string?" where $column like \"$string%\"":"")."group by value order by $column ");
2158 my $elements=$sth->fetchall_arrayref({});
2163 my @servers=qw<biblioserver authorityserver>;
2164 my (@zconns,@results);
2165 for ( my $i = 0 ; $i < @servers ; $i++ ) {
2166 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
2169 ZOOM::Query::CCL2RPN->new( qq"$fieldname $string", $zconns[$i])
2172 # The big moment: asynchronously retrieve results from all servers
2178 my ( $i, $size ) = @_;
2179 for ( my $j = 0 ; $j < $size ; $j++ ) {
2181 @hashscan{qw(value cnt)} =
2182 $results[ $i - 1 ]->display_term($j);
2183 push @elements, \%hashscan;
2191 =head2 _ZOOM_event_loop
2193 _ZOOM_event_loop(\@zconns, \@results, sub {
2194 my ( $i, $size ) = @_;
2198 Processes a ZOOM event loop and passes control to a closure for
2199 processing the results, and destroying the resultsets.
2203 sub _ZOOM_event_loop {
2204 my ($zconns, $results, $callback) = @_;
2205 while ( ( my $i = ZOOM::event( $zconns ) ) != 0 ) {
2206 my $ev = $zconns->[ $i - 1 ]->last_event();
2207 if ( $ev == ZOOM::Event::ZEND ) {
2208 next unless $results->[ $i - 1 ];
2209 my $size = $results->[ $i - 1 ]->size();
2211 $callback->($i, $size);
2216 foreach my $result (@$results) {
2221 =head2 new_record_from_zebra
2223 Given raw data from a searchengine result set, return a MARC::Record object
2225 This helper function is needed to take into account all the involved
2226 system preferences and configuration variables to properly create the
2227 MARC::Record object.
2229 If we are using GRS-1, then the raw data we get from Zebra should be USMARC
2230 data. If we are using DOM, then it has to be MARCXML.
2232 If we are using elasticsearch, it'll already be a MARC::Record and this
2233 function needs a new name.
2237 sub new_record_from_zebra {
2240 my $raw_data = shift;
2241 # Set the default indexing modes
2242 my $search_engine = C4::Context->preference("SearchEngine");
2243 if ($search_engine eq 'Elasticsearch') {
2244 return ref $raw_data eq 'MARC::Record' ? $raw_data : MARC::Record->new_from_xml( $raw_data, 'UTF-8' );
2246 my $index_mode = ( $server eq 'biblioserver' )
2247 ? C4::Context->config('zebra_bib_index_mode') // 'dom'
2248 : C4::Context->config('zebra_auth_index_mode') // 'dom';
2250 my $marc_record = eval {
2251 if ( $index_mode eq 'dom' ) {
2252 MARC::Record->new_from_xml( $raw_data, 'UTF-8' );
2254 MARC::Record->new_from_usmarc( $raw_data );
2261 return $marc_record;
2266 END { } # module clean-up code here (global destructor)
2273 Koha Development Team <http://koha-community.org/>