Bug 21626: (follow-up) Changes based on feedback
[koha.git] / C4 / Search.pm
blob559abea4baba3a4785f26e8a7d98d570a45fe2c5
1 package C4::Search;
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>.
18 use strict;
19 #use warnings; FIXME - Bug 2505
20 require Exporter;
21 use C4::Context;
22 use C4::Biblio; # GetMarcFromKohaField, GetBiblioData
23 use C4::Koha; # getFacets
24 use Koha::DateUtils;
25 use Koha::Libraries;
26 use Lingua::Stem;
27 use C4::Search::PazPar2;
28 use XML::Simple;
29 use C4::XSLT;
30 use C4::Reserves; # GetReserveStatus
31 use C4::Debug;
32 use C4::Charset;
33 use Koha::AuthorisedValues;
34 use Koha::ItemTypes;
35 use Koha::Libraries;
36 use Koha::Patrons;
37 use YAML;
38 use URI::Escape;
39 use Business::ISBN;
40 use MARC::Record;
41 use MARC::Field;
42 use vars qw(@ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
44 BEGIN {
45 $DEBUG = ($ENV{DEBUG}) ? 1 : 0;
48 =head1 NAME
50 C4::Search - Functions for searching the Koha catalog.
52 =head1 SYNOPSIS
54 See opac/opac-search.pl or catalogue/search.pl for example of usage
56 =head1 DESCRIPTION
58 This module provides searching functions for Koha's bibliographic databases
60 =head1 FUNCTIONS
62 =cut
64 @ISA = qw(Exporter);
65 @EXPORT = qw(
66 &FindDuplicate
67 &SimpleSearch
68 &searchResults
69 &getRecords
70 &buildQuery
71 &GetDistinctValues
72 &enabled_staff_search_views
75 # make all your functions, whether exported or not;
77 =head2 FindDuplicate
79 ($biblionumber,$biblionumber,$title) = FindDuplicate($record);
81 This function attempts to find duplicate records using a hard-coded, fairly simplistic algorithm
83 =cut
85 sub FindDuplicate {
86 my ($record) = @_;
87 my $dbh = C4::Context->dbh;
88 my $result = TransformMarcToKoha( $record, '' );
89 my $sth;
90 my $query;
91 my $search;
92 my $type;
93 my ( $biblionumber, $title );
95 # search duplicate on ISBN, easy and fast..
96 # ... normalize first
97 if ( $result->{isbn} ) {
98 $result->{isbn} =~ s/\(.*$//;
99 $result->{isbn} =~ s/\s+$//;
100 $query = "isbn:$result->{isbn}";
102 else {
103 my $QParser;
104 $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser'));
105 my $titleindex;
106 my $authorindex;
107 my $op;
109 if ($QParser) {
110 $titleindex = 'title|exact';
111 $authorindex = 'author|exact';
112 $op = '&&';
113 $QParser->custom_data->{'QueryAutoTruncate'} = C4::Context->preference('QueryAutoTruncate');
114 } else {
115 $titleindex = 'ti,ext';
116 $authorindex = 'au,ext';
117 $op = 'and';
120 $result->{title} =~ s /\\//g;
121 $result->{title} =~ s /\"//g;
122 $result->{title} =~ s /\(//g;
123 $result->{title} =~ s /\)//g;
125 # FIXME: instead of removing operators, could just do
126 # quotes around the value
127 $result->{title} =~ s/(and|or|not)//g;
128 $query = "$titleindex:\"$result->{title}\"";
129 if ( $result->{author} ) {
130 $result->{author} =~ s /\\//g;
131 $result->{author} =~ s /\"//g;
132 $result->{author} =~ s /\(//g;
133 $result->{author} =~ s /\)//g;
135 # remove valid operators
136 $result->{author} =~ s/(and|or|not)//g;
137 $query .= " $op $authorindex:\"$result->{author}\"";
141 my ( $error, $searchresults, undef ) = SimpleSearch($query); # FIXME :: hardcoded !
142 my @results;
143 if (!defined $error) {
144 foreach my $possible_duplicate_record (@{$searchresults}) {
145 my $marcrecord = new_record_from_zebra(
146 'biblioserver',
147 $possible_duplicate_record
150 my $result = TransformMarcToKoha( $marcrecord, '' );
152 # FIXME :: why 2 $biblionumber ?
153 if ($result) {
154 push @results, $result->{'biblionumber'};
155 push @results, $result->{'title'};
159 return @results;
162 =head2 SimpleSearch
164 ( $error, $results, $total_hits ) = SimpleSearch( $query, $offset, $max_results, [@servers], [%options] );
166 This function provides a simple search API on the bibliographic catalog
168 =over 2
170 =item C<input arg:>
172 * $query can be a simple keyword or a complete CCL query
173 * @servers is optional. Defaults to biblioserver as found in koha-conf.xml
174 * $offset - If present, represents the number of records at the beginning to omit. Defaults to 0
175 * $max_results - if present, determines the maximum number of records to fetch. undef is All. defaults to undef.
176 * %options is optional. (e.g. "skip_normalize" allows you to skip changing : to = )
179 =item C<Return:>
181 Returns an array consisting of three elements
182 * $error is undefined unless an error is detected
183 * $results is a reference to an array of records.
184 * $total_hits is the number of hits that would have been returned with no limit
186 If an error is returned the two other return elements are undefined. If error itself is undefined
187 the other two elements are always defined
189 =item C<usage in the script:>
191 =back
193 my ( $error, $marcresults, $total_hits ) = SimpleSearch($query);
195 if (defined $error) {
196 $template->param(query_error => $error);
197 warn "error: ".$error;
198 output_html_with_http_headers $input, $cookie, $template->output;
199 exit;
202 my $hits = @{$marcresults};
203 my @results;
205 for my $r ( @{$marcresults} ) {
206 my $marcrecord = MARC::File::USMARC::decode($r);
207 my $biblio = TransformMarcToKoha($marcrecord,q{});
209 #build the iarray of hashs for the template.
210 push @results, {
211 title => $biblio->{'title'},
212 subtitle => $biblio->{'subtitle'},
213 biblionumber => $biblio->{'biblionumber'},
214 author => $biblio->{'author'},
215 publishercode => $biblio->{'publishercode'},
216 publicationyear => $biblio->{'publicationyear'},
221 $template->param(result=>\@results);
223 =cut
225 sub SimpleSearch {
226 my ( $query, $offset, $max_results, $servers, %options ) = @_;
228 return ( 'No query entered', undef, undef ) unless $query;
229 # FIXME hardcoded value. See catalog/search.pl & opac-search.pl too.
230 my @servers = defined ( $servers ) ? @$servers : ( 'biblioserver' );
231 my @zoom_queries;
232 my @tmpresults;
233 my @zconns;
234 my $results = [];
235 my $total_hits = 0;
237 my $QParser;
238 $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser') && ! ($query =~ m/\w,\w|\w=\w/));
239 if ($QParser) {
240 $QParser->custom_data->{'QueryAutoTruncate'} = C4::Context->preference('QueryAutoTruncate');
243 # Initialize & Search Zebra
244 for ( my $i = 0 ; $i < @servers ; $i++ ) {
245 eval {
246 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
247 if ($QParser) {
248 $query =~ s/=/:/g unless $options{skip_normalize};
249 $QParser->parse( $query );
250 $query = $QParser->target_syntax($servers[$i]);
251 $zoom_queries[$i] = new ZOOM::Query::PQF( $query, $zconns[$i]);
252 } else {
253 $query =~ s/:/=/g unless $options{skip_normalize};
254 $zoom_queries[$i] = new ZOOM::Query::CCL2RPN( $query, $zconns[$i]);
256 $tmpresults[$i] = $zconns[$i]->search( $zoom_queries[$i] );
258 # error handling
259 my $error =
260 $zconns[$i]->errmsg() . " ("
261 . $zconns[$i]->errcode() . ") "
262 . $zconns[$i]->addinfo() . " "
263 . $zconns[$i]->diagset();
265 return ( $error, undef, undef ) if $zconns[$i]->errcode();
267 if ($@) {
269 # caught a ZOOM::Exception
270 my $error =
271 $@->message() . " ("
272 . $@->code() . ") "
273 . $@->addinfo() . " "
274 . $@->diagset();
275 warn $error." for query: $query";
276 return ( $error, undef, undef );
280 _ZOOM_event_loop(
281 \@zconns,
282 \@tmpresults,
283 sub {
284 my ($i, $size) = @_;
285 my $first_record = defined($offset) ? $offset + 1 : 1;
286 my $hits = $tmpresults[ $i - 1 ]->size();
287 $total_hits += $hits;
288 my $last_record = $hits;
289 if ( defined $max_results && $offset + $max_results < $hits ) {
290 $last_record = $offset + $max_results;
293 for my $j ( $first_record .. $last_record ) {
294 my $record = eval {
295 $tmpresults[ $i - 1 ]->record( $j - 1 )->raw()
296 ; # 0 indexed
298 push @{$results}, $record if defined $record;
303 foreach my $zoom_query (@zoom_queries) {
304 $zoom_query->destroy();
307 return ( undef, $results, $total_hits );
310 =head2 getRecords
312 ( undef, $results_hashref, \@facets_loop ) = getRecords (
314 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
315 $results_per_page, $offset, $expanded_facet, $branches,$itemtypes,
316 $query_type, $scan
319 The all singing, all dancing, multi-server, asynchronous, scanning,
320 searching, record nabbing, facet-building
322 See verbse embedded documentation.
324 =cut
326 sub getRecords {
327 my (
328 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
329 $results_per_page, $offset, $expanded_facet, $branches,
330 $itemtypes, $query_type, $scan, $opac
331 ) = @_;
333 my @servers = @$servers_ref;
334 my @sort_by = @$sort_by_ref;
335 $offset = 0 if $offset < 0;
337 # Initialize variables for the ZOOM connection and results object
338 my $zconn;
339 my @zconns;
340 my @results;
341 my $results_hashref = ();
343 # TODO simplify this structure ( { branchcode => $branchname } is enought) and remove this parameter
344 $branches ||= { map { $_->branchcode => { branchname => $_->branchname } } Koha::Libraries->search };
346 # Initialize variables for the faceted results objects
347 my $facets_counter = {};
348 my $facets_info = {};
349 my $facets = getFacets();
351 my @facets_loop; # stores the ref to array of hashes for template facets loop
353 ### LOOP THROUGH THE SERVERS
354 for ( my $i = 0 ; $i < @servers ; $i++ ) {
355 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
357 # perform the search, create the results objects
358 # if this is a local search, use the $koha-query, if it's a federated one, use the federated-query
359 my $query_to_use = ($servers[$i] =~ /biblioserver/) ? $koha_query : $simple_query;
361 #$query_to_use = $simple_query if $scan;
362 warn $simple_query if ( $scan and $DEBUG );
364 # Check if we've got a query_type defined, if so, use it
365 eval {
366 if ($query_type) {
367 if ($query_type =~ /^ccl/) {
368 $query_to_use =~ s/\:/\=/g; # change : to = last minute (FIXME)
369 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
370 } elsif ($query_type =~ /^cql/) {
371 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CQL($query_to_use, $zconns[$i]));
372 } elsif ($query_type =~ /^pqf/) {
373 $results[$i] = $zconns[$i]->search(new ZOOM::Query::PQF($query_to_use, $zconns[$i]));
374 } else {
375 warn "Unknown query_type '$query_type'. Results undetermined.";
377 } elsif ($scan) {
378 $results[$i] = $zconns[$i]->scan( new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
379 } else {
380 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
383 if ($@) {
384 warn "WARNING: query problem with $query_to_use " . $@;
387 # Concatenate the sort_by limits and pass them to the results object
388 # Note: sort will override rank
389 my $sort_by;
390 foreach my $sort (@sort_by) {
391 if ( $sort eq "author_az" || $sort eq "author_asc" ) {
392 $sort_by .= "1=1003 <i ";
394 elsif ( $sort eq "author_za" || $sort eq "author_dsc" ) {
395 $sort_by .= "1=1003 >i ";
397 elsif ( $sort eq "popularity_asc" ) {
398 $sort_by .= "1=9003 <i ";
400 elsif ( $sort eq "popularity_dsc" ) {
401 $sort_by .= "1=9003 >i ";
403 elsif ( $sort eq "call_number_asc" ) {
404 $sort_by .= "1=8007 <i ";
406 elsif ( $sort eq "call_number_dsc" ) {
407 $sort_by .= "1=8007 >i ";
409 elsif ( $sort eq "pubdate_asc" ) {
410 $sort_by .= "1=31 <i ";
412 elsif ( $sort eq "pubdate_dsc" ) {
413 $sort_by .= "1=31 >i ";
415 elsif ( $sort eq "acqdate_asc" ) {
416 $sort_by .= "1=32 <i ";
418 elsif ( $sort eq "acqdate_dsc" ) {
419 $sort_by .= "1=32 >i ";
421 elsif ( $sort eq "title_az" || $sort eq "title_asc" ) {
422 $sort_by .= "1=4 <i ";
424 elsif ( $sort eq "title_za" || $sort eq "title_dsc" ) {
425 $sort_by .= "1=4 >i ";
427 else {
428 warn "Ignoring unrecognized sort '$sort' requested" if $sort_by;
431 if ( $sort_by && !$scan && $results[$i] ) {
432 if ( $results[$i]->sort( "yaz", $sort_by ) < 0 ) {
433 warn "WARNING sort $sort_by failed";
436 } # finished looping through servers
438 # The big moment: asynchronously retrieve results from all servers
439 _ZOOM_event_loop(
440 \@zconns,
441 \@results,
442 sub {
443 my ( $i, $size ) = @_;
444 my $results_hash;
446 # loop through the results
447 $results_hash->{'hits'} = $size;
448 my $times;
449 if ( $offset + $results_per_page <= $size ) {
450 $times = $offset + $results_per_page;
452 else {
453 $times = $size;
456 for ( my $j = $offset ; $j < $times ; $j++ ) {
457 my $records_hash;
458 my $record;
460 ## Check if it's an index scan
461 if ($scan) {
462 my ( $term, $occ ) = $results[ $i - 1 ]->display_term($j);
464 # here we create a minimal MARC record and hand it off to the
465 # template just like a normal result ... perhaps not ideal, but
466 # it works for now
467 my $tmprecord = MARC::Record->new();
468 $tmprecord->encoding('UTF-8');
469 my $tmptitle;
470 my $tmpauthor;
472 # the minimal record in author/title (depending on MARC flavour)
473 if ( C4::Context->preference("marcflavour") eq
474 "UNIMARC" )
476 $tmptitle = MARC::Field->new(
477 '200', ' ', ' ',
478 a => $term,
479 f => $occ
481 $tmprecord->append_fields($tmptitle);
483 else {
484 $tmptitle =
485 MARC::Field->new( '245', ' ', ' ', a => $term, );
486 $tmpauthor =
487 MARC::Field->new( '100', ' ', ' ', a => $occ, );
488 $tmprecord->append_fields($tmptitle);
489 $tmprecord->append_fields($tmpauthor);
491 $results_hash->{'RECORDS'}[$j] =
492 $tmprecord->as_usmarc();
495 # not an index scan
496 else {
497 $record = $results[ $i - 1 ]->record($j)->raw();
498 # warn "RECORD $j:".$record;
499 $results_hash->{'RECORDS'}[$j] = $record;
503 $results_hashref->{ $servers[ $i - 1 ] } = $results_hash;
505 # Fill the facets while we're looping, but only for the
506 # biblioserver and not for a scan
507 if ( !$scan && $servers[ $i - 1 ] =~ /biblioserver/ ) {
508 $facets_counter = GetFacets( $results[ $i - 1 ] );
509 $facets_info = _get_facets_info( $facets );
512 # BUILD FACETS
513 if ( $servers[ $i - 1 ] =~ /biblioserver/ ) {
514 for my $link_value (
515 sort { $facets_counter->{$b} <=> $facets_counter->{$a} }
516 keys %$facets_counter
519 my $expandable;
520 my $number_of_facets;
521 my @this_facets_array;
522 for my $one_facet (
523 sort {
524 $facets_counter->{$link_value}
525 ->{$b} <=> $facets_counter->{$link_value}
526 ->{$a}
527 } keys %{ $facets_counter->{$link_value} }
530 $number_of_facets++;
531 if ( ( $number_of_facets <= 5 )
532 || ( $expanded_facet eq $link_value )
533 || ( $facets_info->{$link_value}->{'expanded'} )
537 # Sanitize the link value : parenthesis, question and exclamation mark will cause errors with CCL
538 my $facet_link_value = $one_facet;
539 $facet_link_value =~ s/[()!?¡¿؟]/ /g;
541 # fix the length that will display in the label,
542 my $facet_label_value = $one_facet;
543 my $facet_max_length = C4::Context->preference(
544 'FacetLabelTruncationLength')
545 || 20;
546 $facet_label_value =
547 substr( $one_facet, 0, $facet_max_length )
548 . "..."
549 if length($facet_label_value) >
550 $facet_max_length;
552 # if it's a branch, label by the name, not the code,
553 if ( $link_value =~ /branch/ ) {
554 if ( defined $branches
555 && ref($branches) eq "HASH"
556 && defined $branches->{$one_facet}
557 && ref( $branches->{$one_facet} ) eq
558 "HASH" )
560 $facet_label_value =
561 $branches->{$one_facet}
562 ->{'branchname'};
564 else {
565 $facet_label_value = "*";
569 # if it's a itemtype, label by the name, not the code,
570 if ( $link_value =~ /itype/ ) {
571 if ( defined $itemtypes
572 && ref($itemtypes) eq "HASH"
573 && defined $itemtypes->{$one_facet}
574 && ref( $itemtypes->{$one_facet} ) eq
575 "HASH" )
577 $facet_label_value =
578 $itemtypes->{$one_facet}
579 ->{translated_description};
583 # also, if it's a location code, use the name instead of the code
584 if ( $link_value =~ /location/ ) {
585 # TODO Retrieve all authorised values at once, instead of 1 query per entry
586 my $av = Koha::AuthorisedValues->search({ category => 'LOC', authorised_value => $one_facet });
587 $facet_label_value = $av->count ? $av->next->opac_description : '';
590 # also, if it's a collection code, use the name instead of the code
591 if ( $link_value =~ /ccode/ ) {
592 # TODO Retrieve all authorised values at once, instead of 1 query per entry
593 my $av = Koha::AuthorisedValues->search({ category => 'CCODE', authorised_value => $one_facet });
594 $facet_label_value = $av->count ? $av->next->opac_description : '';
597 # but we're down with the whole label being in the link's title.
598 push @this_facets_array,
600 facet_count =>
601 $facets_counter->{$link_value}
602 ->{$one_facet},
603 facet_label_value => $facet_label_value,
604 facet_title_value => $one_facet,
605 facet_link_value => $facet_link_value,
606 type_link_value => $link_value,
608 if ($facet_label_value);
612 # handle expanded option
613 unless ( $facets_info->{$link_value}->{'expanded'} ) {
614 $expandable = 1
615 if ( ( $number_of_facets > 5 )
616 && ( $expanded_facet ne $link_value ) );
618 push @facets_loop,
620 type_link_value => $link_value,
621 type_id => $link_value . "_id",
622 "type_label_"
623 . $facets_info->{$link_value}->{'label_value'} =>
625 facets => \@this_facets_array,
626 expandable => $expandable,
627 expand => $link_value,
629 unless (
631 $facets_info->{$link_value}->{'label_value'} =~
632 /Libraries/
634 and ( Koha::Libraries->search->count == 1 )
641 # This sorts the facets into alphabetical order
642 if (@facets_loop) {
643 foreach my $f (@facets_loop) {
644 $f->{facets} = [ sort { uc($a->{facet_label_value}) cmp uc($b->{facet_label_value}) } @{ $f->{facets} } ];
646 @facets_loop = sort {$a->{expand} cmp $b->{expand}} @facets_loop;
649 return ( undef, $results_hashref, \@facets_loop );
652 sub GetFacets {
654 my $rs = shift;
655 my $facets;
657 my $use_zebra_facets = C4::Context->config('use_zebra_facets') // 0;
659 if ( $use_zebra_facets ) {
660 $facets = _get_facets_from_zebra( $rs );
661 } else {
662 $facets = _get_facets_from_records( $rs );
665 return $facets;
668 sub _get_facets_from_records {
670 my $rs = shift;
672 my $facets_maxrecs = C4::Context->preference('maxRecordsForFacets') // 20;
673 my $facets_config = getFacets();
674 my $facets = {};
675 my $size = $rs->size();
676 my $jmax = $size > $facets_maxrecs
677 ? $facets_maxrecs
678 : $size;
680 for ( my $j = 0 ; $j < $jmax ; $j++ ) {
682 my $marc_record = new_record_from_zebra (
683 'biblioserver',
684 $rs->record( $j )->raw()
687 if ( ! defined $marc_record ) {
688 warn "ERROR DECODING RECORD - $@: " .
689 $rs->record( $j )->raw();
690 next;
693 _get_facets_data_from_record( $marc_record, $facets_config, $facets );
696 return $facets;
699 =head2 _get_facets_data_from_record
701 C4::Search::_get_facets_data_from_record( $marc_record, $facets, $facets_counter );
703 Internal function that extracts facets information from a MARC::Record object
704 and populates $facets_counter for using in getRecords.
706 $facets is expected to be filled with C4::Koha::getFacets output (i.e. the configured
707 facets for Zebra).
709 =cut
711 sub _get_facets_data_from_record {
713 my ( $marc_record, $facets, $facets_counter ) = @_;
715 for my $facet (@$facets) {
717 my @used_datas = ();
719 foreach my $tag ( @{ $facet->{ tags } } ) {
721 # tag number is the first three digits
722 my $tag_num = substr( $tag, 0, 3 );
723 # subfields are the remainder
724 my $subfield_letters = substr( $tag, 3 );
726 my @fields = $marc_record->field( $tag_num );
727 foreach my $field (@fields) {
728 # If $field->indicator(1) eq 'z', it means it is a 'see from'
729 # field introduced because of IncludeSeeFromInSearches, so skip it
730 next if $field->indicator(1) eq 'z';
732 my $data = $field->as_string( $subfield_letters, $facet->{ sep } );
734 unless ( grep { /^\Q$data\E$/ } @used_datas ) {
735 push @used_datas, $data;
736 $facets_counter->{ $facet->{ idx } }->{ $data }++;
743 =head2 _get_facets_from_zebra
745 my $facets = _get_facets_from_zebra( $result_set )
747 Retrieves facets for a specified result set. It loops through the facets defined
748 in C4::Koha::getFacets and returns a hash with the following structure:
750 { facet_idx => {
751 facet_value => count
756 =cut
758 sub _get_facets_from_zebra {
760 my $rs = shift;
762 # save current elementSetName
763 my $elementSetName = $rs->option( 'elementSetName' );
765 my $facets_loop = getFacets();
766 my $facets_data = {};
767 # loop through defined facets and fill the facets hashref
768 foreach my $facet ( @$facets_loop ) {
770 my $idx = $facet->{ idx };
771 my $sep = $facet->{ sep };
772 my $facet_values = _get_facet_from_result_set( $idx, $rs, $sep );
773 if ( $facet_values ) {
774 # we've actually got a result
775 $facets_data->{ $idx } = $facet_values;
778 # set elementSetName to its previous value to avoid side effects
779 $rs->option( elementSetName => $elementSetName );
781 return $facets_data;
784 =head2 _get_facet_from_result_set
786 my $facet_values =
787 C4::Search::_get_facet_from_result_set( $facet_idx, $result_set, $sep )
789 Internal function that extracts facet information for a specific index ($facet_idx) and
790 returns a hash containing facet values and count:
793 $facet_value => $count ,
797 Warning: this function has the side effect of changing the elementSetName for the result
798 set. It is a helper function for the main loop, which takes care of backing it up for
799 restoring.
801 =cut
803 sub _get_facet_from_result_set {
805 my $facet_idx = shift;
806 my $rs = shift;
807 my $sep = shift;
809 my $internal_sep = '<*>';
810 my $facetMaxCount = C4::Context->preference('FacetMaxCount') // 20;
812 return if ( ! defined $facet_idx || ! defined $rs );
813 # zebra's facet element, untokenized index
814 my $facet_element = 'zebra::facet::' . $facet_idx . ':0:' . $facetMaxCount;
815 # configure zebra results for retrieving the desired facet
816 $rs->option( elementSetName => $facet_element );
817 # get the facet record from result set
818 my $facet = $rs->record( 0 )->raw;
819 # if the facet has no restuls...
820 return if !defined $facet;
821 # TODO: benchmark DOM vs. SAX performance
822 my $facet_dom = XML::LibXML->load_xml(
823 string => ($facet)
825 my @terms = $facet_dom->getElementsByTagName('term');
826 return if ! @terms;
828 my $facets = {};
829 foreach my $term ( @terms ) {
830 my $facet_value = $term->textContent;
831 $facet_value =~ s/\Q$internal_sep\E/$sep/ if defined $sep;
832 $facets->{ $facet_value } = $term->getAttribute( 'occur' );
835 return $facets;
838 =head2 _get_facets_info
840 my $facets_info = C4::Search::_get_facets_info( $facets )
842 Internal function that extracts facets information and properly builds
843 the data structure needed to render facet labels.
845 =cut
847 sub _get_facets_info {
849 my $facets = shift;
851 my $facets_info = {};
853 for my $facet ( @$facets ) {
854 $facets_info->{ $facet->{ idx } }->{ label_value } = $facet->{ label };
855 $facets_info->{ $facet->{ idx } }->{ expanded } = $facet->{ expanded };
858 return $facets_info;
861 sub pazGetRecords {
862 my (
863 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
864 $results_per_page, $offset, $expanded_facet, $branches,
865 $query_type, $scan
866 ) = @_;
867 #NOTE: Parameter $branches is not used here !
869 my $paz = C4::Search::PazPar2->new(C4::Context->config('pazpar2url'));
870 $paz->init();
871 $paz->search($simple_query);
872 sleep 1; # FIXME: WHY?
874 # do results
875 my $results_hashref = {};
876 my $stats = XMLin($paz->stat);
877 my $results = XMLin($paz->show($offset, $results_per_page, 'work-title:1'), forcearray => 1);
879 # for a grouped search result, the number of hits
880 # is the number of groups returned; 'bib_hits' will have
881 # the total number of bibs.
882 $results_hashref->{'biblioserver'}->{'hits'} = $results->{'merged'}->[0];
883 $results_hashref->{'biblioserver'}->{'bib_hits'} = $stats->{'hits'};
885 HIT: foreach my $hit (@{ $results->{'hit'} }) {
886 my $recid = $hit->{recid}->[0];
888 my $work_title = $hit->{'md-work-title'}->[0];
889 my $work_author;
890 if (exists $hit->{'md-work-author'}) {
891 $work_author = $hit->{'md-work-author'}->[0];
893 my $group_label = (defined $work_author) ? "$work_title / $work_author" : $work_title;
895 my $result_group = {};
896 $result_group->{'group_label'} = $group_label;
897 $result_group->{'group_merge_key'} = $recid;
899 my $count = 1;
900 if (exists $hit->{count}) {
901 $count = $hit->{count}->[0];
903 $result_group->{'group_count'} = $count;
905 for (my $i = 0; $i < $count; $i++) {
906 # FIXME -- may need to worry about diacritics here
907 my $rec = $paz->record($recid, $i);
908 push @{ $result_group->{'RECORDS'} }, $rec;
911 push @{ $results_hashref->{'biblioserver'}->{'GROUPS'} }, $result_group;
914 # pass through facets
915 my $termlist_xml = $paz->termlist('author,subject');
916 my $terms = XMLin($termlist_xml, forcearray => 1);
917 my @facets_loop = ();
918 #die Dumper($results);
919 # foreach my $list (sort keys %{ $terms->{'list'} }) {
920 # my @facets = ();
921 # foreach my $facet (sort @{ $terms->{'list'}->{$list}->{'term'} } ) {
922 # push @facets, {
923 # facet_label_value => $facet->{'name'}->[0],
924 # };
926 # push @facets_loop, ( {
927 # type_label => $list,
928 # facets => \@facets,
929 # } );
932 return ( undef, $results_hashref, \@facets_loop );
935 # TRUNCATION
936 sub _detect_truncation {
937 my ( $operand, $index ) = @_;
938 my ( @nontruncated, @righttruncated, @lefttruncated, @rightlefttruncated,
939 @regexpr );
940 $operand =~ s/^ //g;
941 my @wordlist = split( /\s/, $operand );
942 foreach my $word (@wordlist) {
943 if ( $word =~ s/^\*([^\*]+)\*$/$1/ ) {
944 push @rightlefttruncated, $word;
946 elsif ( $word =~ s/^\*([^\*]+)$/$1/ ) {
947 push @lefttruncated, $word;
949 elsif ( $word =~ s/^([^\*]+)\*$/$1/ ) {
950 push @righttruncated, $word;
952 elsif ( index( $word, "*" ) < 0 ) {
953 push @nontruncated, $word;
955 else {
956 push @regexpr, $word;
959 return (
960 \@nontruncated, \@righttruncated, \@lefttruncated,
961 \@rightlefttruncated, \@regexpr
965 # STEMMING
966 sub _build_stemmed_operand {
967 my ($operand,$lang) = @_;
968 require Lingua::Stem::Snowball ;
969 my $stemmed_operand=q{};
971 # Stemmer needs language
972 return $operand unless $lang;
974 # If operand contains a digit, it is almost certainly an identifier, and should
975 # not be stemmed. This is particularly relevant for ISBNs and ISSNs, which
976 # can contain the letter "X" - for example, _build_stemmend_operand would reduce
977 # "014100018X" to "x ", which for a MARC21 database would bring up irrelevant
978 # results (e.g., "23 x 29 cm." from the 300$c). Bug 2098.
979 return $operand if $operand =~ /\d/;
981 # FIXME: the locale should be set based on the user's language and/or search choice
982 #warn "$lang";
983 # Make sure we only use the first two letters from the language code
984 $lang = lc(substr($lang, 0, 2));
985 # The language codes for the two variants of Norwegian will now be "nb" and "nn",
986 # none of which Lingua::Stem::Snowball can use, so we need to "translate" them
987 if ($lang eq 'nb' || $lang eq 'nn') {
988 $lang = 'no';
990 my $stemmer = Lingua::Stem::Snowball->new( lang => $lang,
991 encoding => "UTF-8" );
993 my @words = split( / /, $operand );
994 my @stems = $stemmer->stem(\@words);
995 for my $stem (@stems) {
996 $stemmed_operand .= "$stem";
997 $stemmed_operand .= "?"
998 unless ( $stem =~ /(and$|or$|not$)/ ) || ( length($stem) < 3 );
999 $stemmed_operand .= " ";
1001 warn "STEMMED OPERAND: $stemmed_operand" if $DEBUG;
1002 return $stemmed_operand;
1005 # FIELD WEIGHTING
1006 sub _build_weighted_query {
1008 # FIELD WEIGHTING - This is largely experimental stuff. What I'm committing works
1009 # pretty well but could work much better if we had a smarter query parser
1010 my ( $operand, $stemmed_operand, $index ) = @_;
1011 my $stemming = C4::Context->preference("QueryStemming") || 0;
1012 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
1013 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
1014 $operand =~ s/"/ /g; # Bug 7518: searches with quotation marks don't work
1016 my $weighted_query .= "(rk=("; # Specifies that we're applying rank
1018 # Keyword, or, no index specified
1019 if ( ( $index eq 'kw' ) || ( !$index ) ) {
1020 $weighted_query .=
1021 "Title-cover,ext,r1=\"$operand\""; # exact title-cover
1022 $weighted_query .= " or ti,ext,r2=\"$operand\""; # exact title
1023 $weighted_query .= " or Title-cover,phr,r3=\"$operand\""; # phrase title
1024 $weighted_query .= " or ti,wrdl,r4=\"$operand\""; # words in title
1025 #$weighted_query .= " or any,ext,r4=$operand"; # exact any
1026 #$weighted_query .=" or kw,wrdl,r5=\"$operand\""; # word list any
1027 $weighted_query .= " or wrdl,fuzzy,r8=\"$operand\""
1028 if $fuzzy_enabled; # add fuzzy, word list
1029 $weighted_query .= " or wrdl,right-Truncation,r9=\"$stemmed_operand\""
1030 if ( $stemming and $stemmed_operand )
1031 ; # add stemming, right truncation
1032 $weighted_query .= " or wrdl,r9=\"$operand\"";
1034 # embedded sorting: 0 a-z; 1 z-a
1035 # $weighted_query .= ") or (sort1,aut=1";
1038 # Barcode searches should skip this process
1039 elsif ( $index eq 'bc' ) {
1040 $weighted_query .= "bc=\"$operand\"";
1043 # Authority-number searches should skip this process
1044 elsif ( $index eq 'an' ) {
1045 $weighted_query .= "an=\"$operand\"";
1048 # If the index is numeric, don't autoquote it.
1049 elsif ( $index =~ /,st-numeric$/ ) {
1050 $weighted_query .= " $index=$operand";
1053 # If the index already has more than one qualifier, wrap the operand
1054 # in quotes and pass it back (assumption is that the user knows what they
1055 # are doing and won't appreciate us mucking up their query
1056 elsif ( $index =~ ',' ) {
1057 $weighted_query .= " $index=\"$operand\"";
1060 #TODO: build better cases based on specific search indexes
1061 else {
1062 $weighted_query .= " $index,ext,r1=\"$operand\""; # exact index
1063 #$weighted_query .= " or (title-sort-az=0 or $index,startswithnt,st-word,r3=$operand #)";
1064 $weighted_query .= " or $index,phr,r3=\"$operand\""; # phrase index
1065 $weighted_query .= " or $index,wrdl,r6=\"$operand\""; # word list index
1066 $weighted_query .= " or $index,wrdl,fuzzy,r8=\"$operand\""
1067 if $fuzzy_enabled; # add fuzzy, word list
1068 $weighted_query .= " or $index,wrdl,rt,r9=\"$stemmed_operand\""
1069 if ( $stemming and $stemmed_operand ); # add stemming, right truncation
1072 $weighted_query .= "))"; # close rank specification
1073 return $weighted_query;
1076 =head2 getIndexes
1078 Return an array with available indexes.
1080 =cut
1082 sub getIndexes{
1083 my @indexes = (
1084 # biblio indexes
1085 'ab',
1086 'Abstract',
1087 'acqdate',
1088 'allrecords',
1089 'an',
1090 'Any',
1091 'at',
1092 'arl',
1093 'arp',
1094 'au',
1095 'aub',
1096 'aud',
1097 'audience',
1098 'auo',
1099 'aut',
1100 'Author',
1101 'Author-in-order ',
1102 'Author-personal-bibliography',
1103 'Authority-Number',
1104 'authtype',
1105 'bc',
1106 'Bib-level',
1107 'biblionumber',
1108 'bio',
1109 'biography',
1110 'callnum',
1111 'cfn',
1112 'Chronological-subdivision',
1113 'cn-bib-source',
1114 'cn-bib-sort',
1115 'cn-class',
1116 'cn-item',
1117 'cn-prefix',
1118 'cn-suffix',
1119 'cpn',
1120 'Code-institution',
1121 'Conference-name',
1122 'Conference-name-heading',
1123 'Conference-name-see',
1124 'Conference-name-seealso',
1125 'Content-type',
1126 'Control-number',
1127 'copydate',
1128 'Corporate-name',
1129 'Corporate-name-heading',
1130 'Corporate-name-see',
1131 'Corporate-name-seealso',
1132 'Country-publication',
1133 'ctype',
1134 'curriculum',
1135 'date-entered-on-file',
1136 'Date-of-acquisition',
1137 'Date-of-publication',
1138 'Date-time-last-modified',
1139 'Dewey-classification',
1140 'Dissertation-information',
1141 'diss',
1142 'dtlm',
1143 'EAN',
1144 'extent',
1145 'fic',
1146 'fiction',
1147 'Form-subdivision',
1148 'format',
1149 'Geographic-subdivision',
1150 'he',
1151 'Heading',
1152 'Heading-use-main-or-added-entry',
1153 'Heading-use-series-added-entry ',
1154 'Heading-use-subject-added-entry',
1155 'Host-item',
1156 'id-other',
1157 'ident',
1158 'Identifier-standard',
1159 'Illustration-code',
1160 'Index-term-genre',
1161 'Index-term-uncontrolled',
1162 'Interest-age-level',
1163 'Interest-grade-level',
1164 'ISBN',
1165 'isbn',
1166 'ISSN',
1167 'issn',
1168 'itemtype',
1169 'kw',
1170 'Koha-Auth-Number',
1171 'l-format',
1172 'language',
1173 'language-original',
1174 'lc-card',
1175 'LC-card-number',
1176 'lcn',
1177 'lex',
1178 'lexile-number',
1179 'llength',
1180 'ln',
1181 'ln-audio',
1182 'ln-subtitle',
1183 'Local-classification',
1184 'Local-number',
1185 'Match-heading',
1186 'Match-heading-see-from',
1187 'Material-type',
1188 'mc-itemtype',
1189 'mc-rtype',
1190 'mus',
1191 'name',
1192 'Music-number',
1193 'Name-geographic',
1194 'Name-geographic-heading',
1195 'Name-geographic-see',
1196 'Name-geographic-seealso',
1197 'nb',
1198 'Note',
1199 'notes',
1200 'ns',
1201 'nt',
1202 'Other-control-number',
1203 'pb',
1204 'Personal-name',
1205 'Personal-name-heading',
1206 'Personal-name-see',
1207 'Personal-name-seealso',
1208 'pl',
1209 'Place-publication',
1210 'pn',
1211 'popularity',
1212 'pubdate',
1213 'Publisher',
1214 'Provider',
1215 'pv',
1216 'Reading-grade-level',
1217 'Record-control-number',
1218 'rcn',
1219 'Record-type',
1220 'rtype',
1221 'se',
1222 'See',
1223 'See-also',
1224 'sn',
1225 'Stock-number',
1226 'su',
1227 'Subject',
1228 'Subject-heading-thesaurus',
1229 'Subject-name-personal',
1230 'Subject-subdivision',
1231 'Summary',
1232 'Suppress',
1233 'su-geo',
1234 'su-na',
1235 'su-to',
1236 'su-ut',
1237 'ut',
1238 'Term-genre-form',
1239 'Term-genre-form-heading',
1240 'Term-genre-form-see',
1241 'Term-genre-form-seealso',
1242 'ti',
1243 'Title',
1244 'Title-cover',
1245 'Title-series',
1246 'Title-uniform',
1247 'Title-uniform-heading',
1248 'Title-uniform-see',
1249 'Title-uniform-seealso',
1250 'totalissues',
1251 'yr',
1253 # items indexes
1254 'acqsource',
1255 'barcode',
1256 'bc',
1257 'branch',
1258 'ccode',
1259 'classification-source',
1260 'cn-sort',
1261 'coded-location-qualifier',
1262 'copynumber',
1263 'damaged',
1264 'datelastborrowed',
1265 'datelastseen',
1266 'holdingbranch',
1267 'homebranch',
1268 'issues',
1269 'item',
1270 'itemnumber',
1271 'itype',
1272 'Local-classification',
1273 'location',
1274 'lost',
1275 'materials-specified',
1276 'mc-ccode',
1277 'mc-itype',
1278 'mc-loc',
1279 'notforloan',
1280 'Number-local-acquisition',
1281 'onloan',
1282 'price',
1283 'renewals',
1284 'replacementprice',
1285 'replacementpricedate',
1286 'reserves',
1287 'restricted',
1288 'stack',
1289 'stocknumber',
1290 'inv',
1291 'uri',
1292 'withdrawn',
1294 # subject related
1297 return \@indexes;
1300 =head2 _handle_exploding_index
1302 my $query = _handle_exploding_index($index, $term)
1304 Callback routine to generate the search for "exploding" indexes (i.e.
1305 those indexes which are turned into multiple or-connected searches based
1306 on authority data).
1308 =cut
1310 sub _handle_exploding_index {
1311 my ($QParser, $filter, $params, $negate, $server) = @_;
1312 my $index = $filter;
1313 my $term = join(' ', @$params);
1315 return unless ($index =~ m/(su-br|su-na|su-rl)/ && $term);
1317 my $marcflavour = C4::Context->preference('marcflavour');
1319 my $codesubfield = $marcflavour eq 'UNIMARC' ? '5' : 'w';
1320 my $wantedcodes = '';
1321 my @subqueries = ( "\@attr 1=Subject \@attr 4=1 \"$term\"");
1322 my ($error, $results, $total_hits) = SimpleSearch( "he:$term", undef, undef, [ "authorityserver" ] );
1323 foreach my $auth (@$results) {
1324 my $record = MARC::Record->new_from_usmarc($auth);
1325 my @references = $record->field('5..');
1326 if (@references) {
1327 if ($index eq 'su-br') {
1328 $wantedcodes = 'g';
1329 } elsif ($index eq 'su-na') {
1330 $wantedcodes = 'h';
1331 } elsif ($index eq 'su-rl') {
1332 $wantedcodes = '';
1334 foreach my $reference (@references) {
1335 my $codes = $reference->subfield($codesubfield);
1336 push @subqueries, '@attr 1=Subject @attr 4=1 "' . $reference->as_string('abcdefghijlmnopqrstuvxyz') . '"' if (($codes && $codes eq $wantedcodes) || !$wantedcodes);
1340 my $query = ' @or ' x (scalar(@subqueries) - 1) . join(' ', @subqueries);
1341 return $query;
1344 =head2 parseQuery
1346 ( $operators, $operands, $indexes, $limits,
1347 $sort_by, $scan, $lang ) =
1348 parseQuery ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1350 Shim function to ease the transition from buildQuery to a new QueryParser.
1351 This function is called at the beginning of buildQuery, and modifies
1352 buildQuery's input. If it can handle the input, it returns a query that
1353 buildQuery will not try to parse.
1355 =cut
1357 sub parseQuery {
1358 my ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang) = @_;
1360 my @operators = $operators ? @$operators : ();
1361 my @indexes = $indexes ? @$indexes : ();
1362 my @operands = $operands ? @$operands : ();
1363 my @limits = $limits ? @$limits : ();
1364 my @sort_by = $sort_by ? @$sort_by : ();
1366 my $query = $operands[0];
1367 my $index;
1368 my $term;
1369 my $query_desc;
1371 my $QParser;
1372 $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser') || $query =~ s/^qp=//);
1373 undef $QParser if ($query =~ m/^(ccl=|pqf=|cql=)/ || grep (/\w,\w|\w=\w/, @operands, @indexes) );
1374 undef $QParser if (scalar @limits > 0);
1376 if ($QParser)
1378 $QParser->custom_data->{'QueryAutoTruncate'} = C4::Context->preference('QueryAutoTruncate');
1379 $query = '';
1380 for ( my $ii = 0 ; $ii <= @operands ; $ii++ ) {
1381 next unless $operands[$ii];
1382 $query .= $operators[ $ii - 1 ] eq 'or' ? ' || ' : ' && '
1383 if ($query);
1384 if ( $operands[$ii] =~ /^[^"]\W*[-|_\w]*:\w.*[^"]$/ ) {
1385 $query .= $operands[$ii];
1387 elsif ( $indexes[$ii] =~ m/su-/ ) {
1388 $query .= $indexes[$ii] . '(' . $operands[$ii] . ')';
1390 else {
1391 $query .=
1392 ( $indexes[$ii] ? "$indexes[$ii]:" : '' ) . $operands[$ii];
1395 foreach my $limit (@limits) {
1397 if ( scalar(@sort_by) > 0 ) {
1398 my $modifier_re =
1399 '#(' . join( '|', @{ $QParser->modifiers } ) . ')';
1400 $query =~ s/$modifier_re//g;
1401 foreach my $modifier (@sort_by) {
1402 $query .= " #$modifier";
1406 $query_desc = $query;
1407 $query_desc =~ s/\s+/ /g;
1408 if ( C4::Context->preference("QueryWeightFields") ) {
1410 $QParser->add_bib1_filter_map( 'su-br' => 'biblioserver' =>
1411 { 'target_syntax_callback' => \&_handle_exploding_index } );
1412 $QParser->add_bib1_filter_map( 'su-na' => 'biblioserver' =>
1413 { 'target_syntax_callback' => \&_handle_exploding_index } );
1414 $QParser->add_bib1_filter_map( 'su-rl' => 'biblioserver' =>
1415 { 'target_syntax_callback' => \&_handle_exploding_index } );
1416 $QParser->parse($query);
1417 $operands[0] = "pqf=" . $QParser->target_syntax('biblioserver');
1419 else {
1420 require Koha::QueryParser::Driver::PQF;
1421 my $modifier_re = '#(' . join( '|', @{Koha::QueryParser::Driver::PQF->modifiers}) . ')';
1422 s/$modifier_re//g for @operands;
1425 return ( $operators, \@operands, $indexes, $limits, $sort_by, $scan, $lang, $query_desc);
1428 =head2 buildQuery
1430 ( $error, $query,
1431 $simple_query, $query_cgi,
1432 $query_desc, $limit,
1433 $limit_cgi, $limit_desc,
1434 $query_type ) = buildQuery ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1436 Build queries and limits in CCL, CGI, Human,
1437 handle truncation, stemming, field weighting, fuzziness, etc.
1439 See verbose embedded documentation.
1442 =cut
1444 sub buildQuery {
1445 my ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang) = @_;
1447 warn "---------\nEnter buildQuery\n---------" if $DEBUG;
1449 my $query_desc;
1450 ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang, $query_desc) = parseQuery($operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1452 # dereference
1453 my @operators = $operators ? @$operators : ();
1454 my @indexes = $indexes ? @$indexes : ();
1455 my @operands = $operands ? @$operands : ();
1456 my @limits = $limits ? @$limits : ();
1457 my @sort_by = $sort_by ? @$sort_by : ();
1459 my $stemming = C4::Context->preference("QueryStemming") || 0;
1460 my $auto_truncation = C4::Context->preference("QueryAutoTruncate") || 0;
1461 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
1462 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
1464 my $query = $operands[0];
1465 my $simple_query = $operands[0];
1467 # initialize the variables we're passing back
1468 my $query_cgi;
1469 my $query_type;
1471 my $limit;
1472 my $limit_cgi;
1473 my $limit_desc;
1475 my $cclq = 0;
1476 my $cclindexes = getIndexes();
1477 if ( $query !~ /\s*(ccl=|pqf=|cql=)/ ) {
1478 while ( !$cclq && $query =~ /(?:^|\W)([\w-]+)(,[\w-]+)*[:=]/g ) {
1479 my $dx = lc($1);
1480 $cclq = grep { lc($_) eq $dx } @$cclindexes;
1482 $query = "ccl=$query" if $cclq;
1485 # for handling ccl, cql, pqf queries in diagnostic mode, skip the rest of the steps
1486 # DIAGNOSTIC ONLY!!
1487 if ( $query =~ /^ccl=/ ) {
1488 my $q=$';
1489 # This is needed otherwise ccl= and &limit won't work together, and
1490 # this happens when selecting a subject on the opac-detail page
1491 @limits = grep {!/^$/} @limits;
1492 my $original_q = $q; # without available part
1493 unless ( grep { /^available$/ } @limits ) {
1494 $q =~ s| and \( \( allrecords,AlwaysMatches:'' not onloan,AlwaysMatches:''\) and \(lost,st-numeric=0\) \)||;
1495 $original_q = $q;
1497 if ( @limits ) {
1498 if ( grep { /^available$/ } @limits ) {
1499 $q .= q| and ( ( allrecords,AlwaysMatches:'' not onloan,AlwaysMatches:'') and (lost,st-numeric=0) )|;
1500 delete $limits['available'];
1502 $q .= ' and '.join(' and ', @limits) if @limits;
1504 return ( undef, $q, $q, "q=ccl=".uri_escape_utf8($q), $original_q, '', '', '', 'ccl' );
1506 if ( $query =~ /^cql=/ ) {
1507 return ( undef, $', $', "q=cql=".uri_escape_utf8($'), $', '', '', '', 'cql' );
1509 if ( $query =~ /^pqf=/ ) {
1510 if ($query_desc) {
1511 $query_cgi = "q=".uri_escape_utf8($query_desc);
1512 } else {
1513 $query_desc = $';
1514 $query_cgi = "q=pqf=".uri_escape_utf8($');
1516 return ( undef, $', $', $query_cgi, $query_desc, '', '', '', 'pqf' );
1519 # pass nested queries directly
1520 # FIXME: need better handling of some of these variables in this case
1521 # Nested queries aren't handled well and this implementation is flawed and causes users to be
1522 # unable to search for anything containing () commenting out, will be rewritten for 3.4.0
1523 # if ( $query =~ /(\(|\))/ ) {
1524 # return (
1525 # undef, $query, $simple_query, $query_cgi,
1526 # $query, $limit, $limit_cgi, $limit_desc,
1527 # 'ccl'
1528 # );
1531 # Form-based queries are non-nested and fixed depth, so we can easily modify the incoming
1532 # query operands and indexes and add stemming, truncation, field weighting, etc.
1533 # Once we do so, we'll end up with a value in $query, just like if we had an
1534 # incoming $query from the user
1535 else {
1536 $query = ""
1537 ; # clear it out so we can populate properly with field-weighted, stemmed, etc. query
1538 my $previous_operand
1539 ; # a flag used to keep track if there was a previous query
1540 # if there was, we can apply the current operator
1541 # for every operand
1542 for ( my $i = 0 ; $i <= @operands ; $i++ ) {
1544 # COMBINE OPERANDS, INDEXES AND OPERATORS
1545 if ( ($operands[$i] // '') ne '' ) {
1546 $operands[$i]=~s/^\s+//;
1548 # A flag to determine whether or not to add the index to the query
1549 my $indexes_set;
1551 # If the user is sophisticated enough to specify an index, turn off field weighting, and stemming handling
1552 if ( $operands[$i] =~ /\w(:|=)/ || $scan ) {
1553 $weight_fields = 0;
1554 $stemming = 0;
1555 } else {
1556 $operands[$i] =~ s/\?/{?}/g; # need to escape question marks
1558 my $operand = $operands[$i];
1559 my $index = $indexes[$i];
1561 # Add index-specific attributes
1563 #Afaik, this 'yr' condition will only ever be met in the staff client advanced search
1564 #for "Publication date", since typing 'yr:YYYY' into the search box produces a CCL query,
1565 #which is processed higher up in this sub. Other than that, year searches are typically
1566 #handled as limits which are not processed her either.
1568 # Search ranges: Date of Publication, st-numeric
1569 if ( $index =~ /(yr|st-numeric)/ ) {
1570 #weight_fields/relevance search causes errors with date ranges
1571 #In the case of YYYY-, it will only return records with a 'yr' of YYYY (not the range)
1572 #In the case of YYYY-YYYY, it will return no results
1573 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = 0;
1576 # Date of Acquisition
1577 elsif ( $index =~ /acqdate/ ) {
1578 #stemming and auto_truncation would have zero impact since it already is YYYY-MM-DD format
1579 #Weight_fields probably SHOULD be turned OFF, otherwise you'll get records floating to the
1580 #top of the results just because they have lots of item records matching that date.
1581 #Fuzzy actually only applies during _build_weighted_query, and is reset there anyway, so
1582 #irrelevant here
1583 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = 0;
1585 # ISBN,ISSN,Standard Number, don't need special treatment
1586 elsif ( $index eq 'nb' || $index eq 'ns' || $index eq 'hi' ) {
1588 $stemming, $auto_truncation,
1589 $weight_fields, $fuzzy_enabled
1590 ) = ( 0, 0, 0, 0 );
1592 if ( $index eq 'nb' ) {
1593 if ( C4::Context->preference("SearchWithISBNVariations") ) {
1594 my @isbns = C4::Koha::GetVariationsOfISBN( $operand );
1595 $operands[$i] = $operand = '(nb=' . join(' OR nb=', @isbns) . ')';
1596 $indexes[$i] = $index = '';
1601 if(not $index){
1602 $index = 'kw';
1605 # Set default structure attribute (word list)
1606 my $struct_attr = q{};
1607 unless ( $indexes_set || !$index || $index =~ /,(st-|phr|ext|wrdl)/ || $index =~ /^(nb|ns)$/ ) {
1608 $struct_attr = ",wrdl";
1611 # Some helpful index variants
1612 my $index_plus = $index . $struct_attr . ':';
1613 my $index_plus_comma = $index . $struct_attr . ',';
1615 if ($auto_truncation){
1616 unless ( $index =~ /,(st-|phr|ext)/ ) {
1617 #FIXME only valid with LTR scripts
1618 $operand=join(" ",map{
1619 (index($_,"*")>0?"$_":"$_*")
1620 }split (/\s+/,$operand));
1621 warn $operand if $DEBUG;
1625 # Detect Truncation
1626 my $truncated_operand;
1627 my( $nontruncated, $righttruncated, $lefttruncated,
1628 $rightlefttruncated, $regexpr
1629 ) = _detect_truncation( $operand, $index );
1630 warn
1631 "TRUNCATION: NON:>@$nontruncated< RIGHT:>@$righttruncated< LEFT:>@$lefttruncated< RIGHTLEFT:>@$rightlefttruncated< REGEX:>@$regexpr<"
1632 if $DEBUG;
1634 # Apply Truncation
1635 if (
1636 scalar(@$righttruncated) + scalar(@$lefttruncated) +
1637 scalar(@$rightlefttruncated) > 0 )
1640 # Don't field weight or add the index to the query, we do it here
1641 $indexes_set = 1;
1642 undef $weight_fields;
1643 my $previous_truncation_operand;
1644 if (scalar @$nontruncated) {
1645 $truncated_operand .= "$index_plus @$nontruncated ";
1646 $previous_truncation_operand = 1;
1648 if (scalar @$righttruncated) {
1649 $truncated_operand .= "and " if $previous_truncation_operand;
1650 $truncated_operand .= $index_plus_comma . "rtrn:@$righttruncated ";
1651 $previous_truncation_operand = 1;
1653 if (scalar @$lefttruncated) {
1654 $truncated_operand .= "and " if $previous_truncation_operand;
1655 $truncated_operand .= $index_plus_comma . "ltrn:@$lefttruncated ";
1656 $previous_truncation_operand = 1;
1658 if (scalar @$rightlefttruncated) {
1659 $truncated_operand .= "and " if $previous_truncation_operand;
1660 $truncated_operand .= $index_plus_comma . "rltrn:@$rightlefttruncated ";
1661 $previous_truncation_operand = 1;
1664 $operand = $truncated_operand if $truncated_operand;
1665 warn "TRUNCATED OPERAND: >$truncated_operand<" if $DEBUG;
1667 # Handle Stemming
1668 my $stemmed_operand;
1669 $stemmed_operand = _build_stemmed_operand($operand, $lang)
1670 if $stemming;
1672 warn "STEMMED OPERAND: >$stemmed_operand<" if $DEBUG;
1674 # Handle Field Weighting
1675 my $weighted_operand;
1676 if ($weight_fields) {
1677 $weighted_operand = _build_weighted_query( $operand, $stemmed_operand, $index );
1678 $operand = $weighted_operand;
1679 $indexes_set = 1;
1682 warn "FIELD WEIGHTED OPERAND: >$weighted_operand<" if $DEBUG;
1684 ($query,$query_cgi,$query_desc,$previous_operand) = _build_initial_query({
1685 query => $query,
1686 query_cgi => $query_cgi,
1687 query_desc => $query_desc,
1688 operator => ($operators[ $i - 1 ]) ? $operators[ $i - 1 ] : '',
1689 parsed_operand => $operand,
1690 original_operand => $operands[$i] // '',
1691 index => $index,
1692 index_plus => $index_plus,
1693 indexes_set => $indexes_set,
1694 previous_operand => $previous_operand,
1697 } #/if $operands
1698 } # /for
1700 warn "QUERY BEFORE LIMITS: >$query<" if $DEBUG;
1702 # add limits
1703 my %group_OR_limits;
1704 my $availability_limit;
1705 foreach my $this_limit (@limits) {
1706 next unless $this_limit;
1707 if ( $this_limit =~ /available/ ) {
1709 ## 'available' is defined as (items.onloan is NULL) and (items.itemlost = 0)
1710 ## In English:
1711 ## all records not indexed in the onloan register (zebra) and all records with a value of lost equal to 0
1712 $availability_limit .=
1713 "( ( allrecords,AlwaysMatches='' not onloan,AlwaysMatches='') and (lost,st-numeric=0) )"; #or ( allrecords,AlwaysMatches='' not lost,AlwaysMatches='')) )";
1714 $limit_cgi .= "&limit=available";
1715 $limit_desc .= "";
1718 # group_OR_limits, prefixed by mc-
1719 # OR every member of the group
1720 elsif ( $this_limit =~ /mc/ ) {
1721 my ($k,$v) = split(/:/, $this_limit,2);
1722 if ( $k !~ /mc-i(tem)?type/ ) {
1723 # in case the mc-ccode value has complicating chars like ()'s inside it we wrap in quotes
1724 $this_limit =~ tr/"//d;
1725 $this_limit = $k.":'".$v."'";
1728 $group_OR_limits{$k} .= " or " if $group_OR_limits{$k};
1729 $limit_desc .= " or " if $group_OR_limits{$k};
1730 $group_OR_limits{$k} .= "$this_limit";
1731 $limit_cgi .= "&limit=" . uri_escape_utf8($this_limit);
1732 $limit_desc .= " $this_limit";
1735 # Regular old limits
1736 else {
1737 $limit .= " and " if $limit || $query;
1738 $limit .= "$this_limit";
1739 $limit_cgi .= "&limit=" . uri_escape_utf8($this_limit);
1740 if ($this_limit =~ /^branch:(.+)/) {
1741 my $branchcode = $1;
1742 my $library = Koha::Libraries->find( $branchcode );
1743 if (defined $library) {
1744 $limit_desc .= " branch:" . $library->branchname;
1745 } else {
1746 $limit_desc .= " $this_limit";
1748 } else {
1749 $limit_desc .= " $this_limit";
1753 foreach my $k (keys (%group_OR_limits)) {
1754 $limit .= " and " if ( $query || $limit );
1755 $limit .= "($group_OR_limits{$k})";
1757 if ($availability_limit) {
1758 $limit .= " and " if ( $query || $limit );
1759 $limit .= "($availability_limit)";
1762 # Normalize the query and limit strings
1763 # This is flawed , means we can't search anything with : in it
1764 # if user wants to do ccl or cql, start the query with that
1765 # $query =~ s/:/=/g;
1766 #NOTE: We use several several different regexps here as you can't have variable length lookback assertions
1767 $query =~ s/(?<=(ti|au|pb|su|an|kw|mc|nb|ns)):/=/g;
1768 $query =~ s/(?<=(wrdl)):/=/g;
1769 $query =~ s/(?<=(trn|phr)):/=/g;
1770 $query =~ s/(?<=(st-numeric)):/=/g;
1771 $query =~ s/(?<=(st-year)):/=/g;
1772 $query =~ s/(?<=(st-date-normalized)):/=/g;
1773 $limit =~ s/:/=/g;
1774 for ( $query, $query_desc, $limit, $limit_desc ) {
1775 s/ +/ /g; # remove extra spaces
1776 s/^ //g; # remove any beginning spaces
1777 s/ $//g; # remove any ending spaces
1778 s/==/=/g; # remove double == from query
1780 $query_cgi =~ s/^&//; # remove unnecessary & from beginning of the query cgi
1782 for ($query_cgi,$simple_query) {
1783 s/"//g;
1785 # append the limit to the query
1786 $query .= " " . $limit;
1788 # Warnings if DEBUG
1789 if ($DEBUG) {
1790 warn "QUERY:" . $query;
1791 warn "QUERY CGI:" . $query_cgi;
1792 warn "QUERY DESC:" . $query_desc;
1793 warn "LIMIT:" . $limit;
1794 warn "LIMIT CGI:" . $limit_cgi;
1795 warn "LIMIT DESC:" . $limit_desc;
1796 warn "---------\nLeave buildQuery\n---------";
1799 return (
1800 undef, $query, $simple_query, $query_cgi,
1801 $query_desc, $limit, $limit_cgi, $limit_desc,
1802 $query_type
1806 =head2 _build_initial_query
1808 ($query, $query_cgi, $query_desc, $previous_operand) = _build_initial_query($initial_query_params);
1810 Build a section of the initial query containing indexes, operators, and operands.
1812 =cut
1814 sub _build_initial_query {
1815 my ($params) = @_;
1817 my $operator = "";
1818 if ($params->{previous_operand}){
1819 #If there is a previous operand, add a supplied operator or the default 'and'
1820 $operator = ($params->{operator}) ? " ".($params->{operator})." " : ' and ';
1823 #NOTE: indexes_set is typically set when doing truncation or field weighting
1824 my $operand = ($params->{indexes_set}) ? $params->{parsed_operand} : $params->{index_plus}.$params->{parsed_operand};
1826 #e.g. "kw,wrdl:test"
1827 #e.g. " and kw,wrdl:test"
1828 $params->{query} .= $operator . $operand;
1830 $params->{query_cgi} .= "&op=".uri_escape_utf8($operator) if $operator;
1831 $params->{query_cgi} .= "&idx=".uri_escape_utf8($params->{index}) if $params->{index};
1832 $params->{query_cgi} .= "&q=".uri_escape_utf8($params->{original_operand}) if $params->{original_operand};
1834 #e.g. " and kw,wrdl: test"
1835 $params->{query_desc} .= $operator . $params->{index_plus} . " " . $params->{original_operand};
1837 $params->{previous_operand} = 1 unless $params->{previous_operand}; #If there is no previous operand, mark this as one
1839 return ($params->{query}, $params->{query_cgi}, $params->{query_desc}, $params->{previous_operand});
1842 =head2 searchResults
1844 my @search_results = searchResults($search_context, $searchdesc, $hits,
1845 $results_per_page, $offset, $scan,
1846 @marcresults);
1848 Format results in a form suitable for passing to the template
1850 =cut
1852 # IMO this subroutine is pretty messy still -- it's responsible for
1853 # building the HTML output for the template
1854 sub searchResults {
1855 my ( $search_context, $searchdesc, $hits, $results_per_page, $offset, $scan, $marcresults ) = @_;
1856 my $dbh = C4::Context->dbh;
1857 my @newresults;
1859 require C4::Items;
1861 $search_context->{'interface'} = 'opac' if !$search_context->{'interface'} || $search_context->{'interface'} ne 'intranet';
1862 my ($is_opac, $hidelostitems);
1863 if ($search_context->{'interface'} eq 'opac') {
1864 $hidelostitems = C4::Context->preference('hidelostitems');
1865 $is_opac = 1;
1868 #Build branchnames hash
1869 my %branches = map { $_->branchcode => $_->branchname } Koha::Libraries->search({}, { order_by => 'branchname' });
1871 # FIXME - We build an authorised values hash here, using the default framework
1872 # though it is possible to have different authvals for different fws.
1874 my $shelflocations =
1875 { map { $_->{authorised_value} => $_->{lib} } Koha::AuthorisedValues->get_descriptions_by_koha_field( { frameworkcode => '', kohafield => 'items.location' } ) };
1877 # get notforloan authorised value list (see $shelflocations FIXME)
1878 my $av = Koha::MarcSubfieldStructures->search({ frameworkcode => '', kohafield => 'items.notforloan', authorised_value => [ -and => {'!=' => undef }, {'!=' => ''}] });
1879 my $notforloan_authorised_value = $av->count ? $av->next->authorised_value : undef;
1881 #Get itemtype hash
1882 my $itemtypes = Koha::ItemTypes->search_with_localization;
1883 my %itemtypes = map { $_->{itemtype} => $_ } @{ $itemtypes->unblessed };
1885 #search item field code
1886 my ($itemtag, undef) = &GetMarcFromKohaField( "items.itemnumber", "" );
1888 ## find column names of items related to MARC
1889 my %subfieldstosearch;
1890 my @columns = Koha::Database->new()->schema()->resultset('Item')->result_source->columns;
1891 for my $column ( @columns ) {
1892 my ( $tagfield, $tagsubfield ) =
1893 &GetMarcFromKohaField( "items." . $column, "" );
1894 if ( defined $tagsubfield ) {
1895 $subfieldstosearch{$column} = $tagsubfield;
1899 # handle which records to actually retrieve
1900 my $times;
1901 if ( $hits && $offset + $results_per_page <= $hits ) {
1902 $times = $offset + $results_per_page;
1904 else {
1905 $times = $hits; # FIXME: if $hits is undefined, why do we want to equal it?
1908 my $marcflavour = C4::Context->preference("marcflavour");
1909 # We get the biblionumber position in MARC
1910 my ($bibliotag,$bibliosubf)=GetMarcFromKohaField('biblio.biblionumber','');
1912 # set stuff for XSLT processing here once, not later again for every record we retrieved
1913 my $xslfile;
1914 my $xslsyspref;
1915 if( $is_opac ){
1916 $xslsyspref = "OPACXSLTResultsDisplay";
1917 $xslfile = C4::Context->preference( $xslsyspref );
1918 } else {
1919 $xslsyspref = "XSLTResultsDisplay";
1920 $xslfile = C4::Context->preference( $xslsyspref ) || "default";
1922 my $lang = $xslfile ? C4::Languages::getlanguage() : undef;
1923 my $sysxml = $xslfile ? C4::XSLT::get_xslt_sysprefs() : undef;
1925 my $userenv = C4::Context->userenv;
1926 my $logged_in_user
1927 = ( defined $userenv and $userenv->{number} )
1928 ? Koha::Patrons->find( $userenv->{number} )
1929 : undef;
1930 my $patron_category_hide_lost_items = ($logged_in_user) ? $logged_in_user->category->hidelostitems : 0;
1932 # loop through all of the records we've retrieved
1933 for ( my $i = $offset ; $i <= $times - 1 ; $i++ ) {
1935 my $marcrecord;
1936 if ($scan) {
1937 # For Scan searches we built USMARC data
1938 $marcrecord = MARC::Record->new_from_usmarc( $marcresults->[$i]);
1939 } else {
1940 # Normal search, render from Zebra's output
1941 $marcrecord = new_record_from_zebra(
1942 'biblioserver',
1943 $marcresults->[$i]
1946 if ( ! defined $marcrecord ) {
1947 warn "ERROR DECODING RECORD - $@: " . $marcresults->[$i];
1948 next;
1952 my $fw = $scan
1953 ? undef
1954 : $bibliotag < 10
1955 ? GetFrameworkCode($marcrecord->field($bibliotag)->data)
1956 : GetFrameworkCode($marcrecord->subfield($bibliotag,$bibliosubf));
1958 SetUTF8Flag($marcrecord);
1959 my $oldbiblio = TransformMarcToKoha( $marcrecord, $fw );
1960 $oldbiblio->{subtitle} = GetRecordValue('subtitle', $marcrecord, $fw);
1961 $oldbiblio->{result_number} = $i + 1;
1963 # add imageurl to itemtype if there is one
1964 $oldbiblio->{imageurl} = getitemtypeimagelocation( $search_context, $itemtypes{ $oldbiblio->{itemtype} }->{imageurl} );
1966 $oldbiblio->{'authorised_value_images'} = ($search_context->{'interface'} eq 'opac' && C4::Context->preference('AuthorisedValueImages')) || ($search_context->{'interface'} eq 'intranet' && C4::Context->preference('StaffAuthorisedValueImages')) ? C4::Items::get_authorised_value_images( C4::Biblio::get_biblio_authorised_values( $oldbiblio->{'biblionumber'}, $marcrecord ) ) : [];
1967 $oldbiblio->{normalized_upc} = GetNormalizedUPC( $marcrecord,$marcflavour);
1968 $oldbiblio->{normalized_ean} = GetNormalizedEAN( $marcrecord,$marcflavour);
1969 $oldbiblio->{normalized_oclc} = GetNormalizedOCLCNumber($marcrecord,$marcflavour);
1970 $oldbiblio->{normalized_isbn} = GetNormalizedISBN(undef,$marcrecord,$marcflavour);
1971 $oldbiblio->{content_identifier_exists} = 1 if ($oldbiblio->{normalized_isbn} or $oldbiblio->{normalized_oclc} or $oldbiblio->{normalized_ean} or $oldbiblio->{normalized_upc});
1973 # edition information, if any
1974 $oldbiblio->{edition} = $oldbiblio->{editionstatement};
1975 $oldbiblio->{description} = $itemtypes{ $oldbiblio->{itemtype} }->{translated_description};
1976 # Build summary if there is one (the summary is defined in the itemtypes table)
1977 # FIXME: is this used anywhere, I think it can be commented out? -- JF
1978 if ( $itemtypes{ $oldbiblio->{itemtype} }->{summary} ) {
1979 my $summary = $itemtypes{ $oldbiblio->{itemtype} }->{summary};
1980 my @fields = $marcrecord->fields();
1982 my $newsummary;
1983 foreach my $line ( "$summary\n" =~ /(.*)\n/g ){
1984 my $tags = {};
1985 foreach my $tag ( $line =~ /\[(\d{3}[\w|\d])\]/ ) {
1986 $tag =~ /(.{3})(.)/;
1987 if($marcrecord->field($1)){
1988 my @abc = $marcrecord->field($1)->subfield($2);
1989 $tags->{$tag} = $#abc + 1 ;
1993 # We catch how many times to repeat this line
1994 my $max = 0;
1995 foreach my $tag (keys(%$tags)){
1996 $max = $tags->{$tag} if($tags->{$tag} > $max);
1999 # we replace, and repeat each line
2000 for (my $i = 0 ; $i < $max ; $i++){
2001 my $newline = $line;
2003 foreach my $tag ( $newline =~ /\[(\d{3}[\w|\d])\]/g ) {
2004 $tag =~ /(.{3})(.)/;
2006 if($marcrecord->field($1)){
2007 my @repl = $marcrecord->field($1)->subfield($2);
2008 my $subfieldvalue = $repl[$i];
2009 $newline =~ s/\[$tag\]/$subfieldvalue/g;
2012 $newsummary .= "$newline\n";
2016 $newsummary =~ s/\[(.*?)]//g;
2017 $newsummary =~ s/\n/<br\/>/g;
2018 $oldbiblio->{summary} = $newsummary;
2021 # Pull out the items fields
2022 my @fields = $marcrecord->field($itemtag);
2023 my $marcflavor = C4::Context->preference("marcflavour");
2025 # adding linked items that belong to host records
2026 if ( C4::Context->preference('EasyAnalyticalRecords') ) {
2027 my $analyticsfield = '773';
2028 if ($marcflavor eq 'MARC21' || $marcflavor eq 'NORMARC') {
2029 $analyticsfield = '773';
2030 } elsif ($marcflavor eq 'UNIMARC') {
2031 $analyticsfield = '461';
2033 foreach my $hostfield ( $marcrecord->field($analyticsfield)) {
2034 my $hostbiblionumber = $hostfield->subfield("0");
2035 my $linkeditemnumber = $hostfield->subfield("9");
2036 if( $hostbiblionumber ) {
2037 my $hostbiblio = GetMarcBiblio({
2038 biblionumber => $hostbiblionumber,
2039 embed_items => 1 });
2040 my ($itemfield, undef) = GetMarcFromKohaField( 'items.itemnumber', GetFrameworkCode($hostbiblionumber) );
2041 if( $hostbiblio ) {
2042 my @hostitems = $hostbiblio->field($itemfield);
2043 foreach my $hostitem (@hostitems){
2044 if ($hostitem->subfield("9") eq $linkeditemnumber){
2045 my $linkeditem =$hostitem;
2046 # append linked items if they exist
2047 push @fields, $linkeditem if $linkeditem;
2055 # Setting item statuses for display
2056 my @available_items_loop;
2057 my @onloan_items_loop;
2058 my @other_items_loop;
2060 my $available_items;
2061 my $onloan_items;
2062 my $other_items;
2064 my $ordered_count = 0;
2065 my $available_count = 0;
2066 my $onloan_count = 0;
2067 my $longoverdue_count = 0;
2068 my $other_count = 0;
2069 my $withdrawn_count = 0;
2070 my $itemlost_count = 0;
2071 my $hideatopac_count = 0;
2072 my $itembinding_count = 0;
2073 my $itemdamaged_count = 0;
2074 my $item_in_transit_count = 0;
2075 my $can_place_holds = 0;
2076 my $item_onhold_count = 0;
2077 my $notforloan_count = 0;
2078 my $items_count = scalar(@fields);
2079 my $maxitems_pref = C4::Context->preference('maxItemsinSearchResults');
2080 my $maxitems = $maxitems_pref ? $maxitems_pref - 1 : 1;
2081 my @hiddenitems; # hidden itemnumbers based on OpacHiddenItems syspref
2083 # loop through every item
2084 foreach my $field (@fields) {
2085 my $item;
2087 # populate the items hash
2088 foreach my $code ( keys %subfieldstosearch ) {
2089 $item->{$code} = $field->subfield( $subfieldstosearch{$code} );
2091 $item->{description} = $itemtypes{ $item->{itype} }{translated_description};
2093 # OPAC hidden items
2094 if ($is_opac) {
2095 # hidden because lost
2096 if ($hidelostitems && $item->{itemlost}) {
2097 $hideatopac_count++;
2098 next;
2100 # hidden based on OpacHiddenItems syspref
2101 my @hi = C4::Items::GetHiddenItemnumbers({ items=> [ $item ], borcat => $search_context->{category} });
2102 if (scalar @hi) {
2103 push @hiddenitems, @hi;
2104 $hideatopac_count++;
2105 next;
2109 my $hbranch = C4::Context->preference('StaffSearchResultsDisplayBranch');
2110 my $otherbranch = $hbranch eq 'homebranch' ? 'holdingbranch' : 'homebranch';
2112 # set item's branch name, use HomeOrHoldingBranch syspref first, fall back to the other one
2113 if ($item->{$hbranch}) {
2114 $item->{'branchname'} = $branches{$item->{$hbranch}};
2116 elsif ($item->{$otherbranch}) { # Last resort
2117 $item->{'branchname'} = $branches{$item->{$otherbranch}};
2120 my $prefix = $item->{$hbranch} . '--' . $item->{location} . $item->{itype} . $item->{itemcallnumber};
2121 # For each grouping of items (onloan, available, unavailable), we build a key to store relevant info about that item
2122 if ( $item->{onloan}
2123 and $logged_in_user
2124 and !( $patron_category_hide_lost_items and $item->{itemlost} ) )
2126 $onloan_count++;
2127 my $key = $prefix . $item->{onloan} . $item->{barcode};
2128 $onloan_items->{$key}->{due_date} = $item->{onloan};
2129 $onloan_items->{$key}->{count}++ if $item->{$hbranch};
2130 $onloan_items->{$key}->{branchname} = $item->{branchname};
2131 $onloan_items->{$key}->{location} = $shelflocations->{ $item->{location} };
2132 $onloan_items->{$key}->{itemcallnumber} = $item->{itemcallnumber};
2133 $onloan_items->{$key}->{description} = $item->{description};
2134 $onloan_items->{$key}->{imageurl} =
2135 getitemtypeimagelocation( $search_context->{'interface'}, $itemtypes{ $item->{itype} }->{imageurl} );
2137 # if something's checked out and lost, mark it as 'long overdue'
2138 if ( $item->{itemlost} ) {
2139 $onloan_items->{$key}->{longoverdue}++;
2140 $longoverdue_count++;
2142 else { # can place holds as long as item isn't lost
2143 $can_place_holds = 1;
2147 # items not on loan, but still unavailable ( lost, withdrawn, damaged )
2148 else {
2150 $item->{notforloan}=1 if !$item->{notforloan} && $itemtypes{ C4::Context->preference("item-level_itypes")? $item->{itype}: $oldbiblio->{itemtype} }->{notforloan};
2152 # item is on order
2153 if ( $item->{notforloan} < 0 ) {
2154 $ordered_count++;
2155 } elsif ( $item->{notforloan} > 0 ) {
2156 $notforloan_count++;
2159 # is item in transit?
2160 my $transfertwhen = '';
2161 my ($transfertfrom, $transfertto);
2163 # is item on the reserve shelf?
2164 my $reservestatus = '';
2166 unless ($item->{withdrawn}
2167 || $item->{itemlost}
2168 || $item->{damaged}
2169 || $item->{notforloan}
2170 || ( C4::Context->preference('MaxSearchResultsItemsPerRecordStatusCheck')
2171 && $items_count > C4::Context->preference('MaxSearchResultsItemsPerRecordStatusCheck') ) ) {
2173 # A couple heuristics to limit how many times
2174 # we query the database for item transfer information, sacrificing
2175 # accuracy in some cases for speed;
2177 # 1. don't query if item has one of the other statuses
2178 # 2. don't check transit status if the bib has
2179 # more than 20 items
2181 # FIXME: to avoid having the query the database like this, and to make
2182 # the in transit status count as unavailable for search limiting,
2183 # should map transit status to record indexed in Zebra.
2185 ($transfertwhen, $transfertfrom, $transfertto) = C4::Circulation::GetTransfers($item->{itemnumber});
2186 $reservestatus = C4::Reserves::GetReserveStatus( $item->{itemnumber} );
2189 # item is withdrawn, lost, damaged, not for loan, reserved or in transit
2190 if ( $item->{withdrawn}
2191 || $item->{itemlost}
2192 || $item->{damaged}
2193 || $item->{notforloan}
2194 || $reservestatus eq 'Waiting'
2195 || ($transfertwhen ne ''))
2197 $withdrawn_count++ if $item->{withdrawn};
2198 $itemlost_count++ if $item->{itemlost};
2199 $itemdamaged_count++ if $item->{damaged};
2200 $item_in_transit_count++ if $transfertwhen ne '';
2201 $item_onhold_count++ if $reservestatus eq 'Waiting';
2202 $item->{status} = $item->{withdrawn} . "-" . $item->{itemlost} . "-" . $item->{damaged} . "-" . $item->{notforloan};
2204 # can place a hold on a item if
2205 # not lost nor withdrawn
2206 # not damaged unless AllowHoldsOnDamagedItems is true
2207 # item is either for loan or on order (notforloan < 0)
2208 $can_place_holds = 1
2209 if (
2210 !$item->{itemlost}
2211 && !$item->{withdrawn}
2212 && ( !$item->{damaged} || C4::Context->preference('AllowHoldsOnDamagedItems') )
2213 && ( !$item->{notforloan} || $item->{notforloan} < 0 )
2216 $other_count++;
2218 my $key = $prefix . $item->{status};
2219 foreach (qw(withdrawn itemlost damaged branchname itemcallnumber)) {
2220 $other_items->{$key}->{$_} = $item->{$_};
2222 $other_items->{$key}->{intransit} = ( $transfertwhen ne '' ) ? 1 : 0;
2223 $other_items->{$key}->{onhold} = ($reservestatus) ? 1 : 0;
2224 $other_items->{$key}->{notforloan} = GetAuthorisedValueDesc('','',$item->{notforloan},'','',$notforloan_authorised_value) if $notforloan_authorised_value and $item->{notforloan};
2225 $other_items->{$key}->{count}++ if $item->{$hbranch};
2226 $other_items->{$key}->{location} = $shelflocations->{ $item->{location} };
2227 $other_items->{$key}->{description} = $item->{description};
2228 $other_items->{$key}->{imageurl} = getitemtypeimagelocation( $search_context->{'interface'}, $itemtypes{ $item->{itype} }->{imageurl} );
2230 # item is available
2231 else {
2232 $can_place_holds = 1;
2233 $available_count++;
2234 $available_items->{$prefix}->{count}++ if $item->{$hbranch};
2235 foreach (qw(branchname itemcallnumber description)) {
2236 $available_items->{$prefix}->{$_} = $item->{$_};
2238 $available_items->{$prefix}->{location} = $shelflocations->{ $item->{location} };
2239 $available_items->{$prefix}->{imageurl} = getitemtypeimagelocation( $search_context->{'interface'}, $itemtypes{ $item->{itype} }->{imageurl} );
2242 } # notforloan, item level and biblioitem level
2244 # if all items are hidden, do not show the record
2245 if ($items_count > 0 && $hideatopac_count == $items_count) {
2246 next;
2249 my ( $availableitemscount, $onloanitemscount, $otheritemscount );
2250 for my $key ( sort keys %$onloan_items ) {
2251 (++$onloanitemscount > $maxitems) and last;
2252 push @onloan_items_loop, $onloan_items->{$key};
2254 for my $key ( sort keys %$other_items ) {
2255 (++$otheritemscount > $maxitems) and last;
2256 push @other_items_loop, $other_items->{$key};
2258 for my $key ( sort keys %$available_items ) {
2259 (++$availableitemscount > $maxitems) and last;
2260 push @available_items_loop, $available_items->{$key}
2263 # XSLT processing of some stuff
2264 # we fetched the sysprefs already before the loop through all retrieved record!
2265 if (!$scan && $xslfile) {
2266 $oldbiblio->{XSLTResultsRecord} = XSLTParse4Display($oldbiblio->{biblionumber}, $marcrecord, $xslsyspref, 1, \@hiddenitems, $sysxml, $xslfile, $lang);
2269 # if biblio level itypes are used and itemtype is notforloan, it can't be reserved either
2270 if (!C4::Context->preference("item-level_itypes")) {
2271 if ($itemtypes{ $oldbiblio->{itemtype} }->{notforloan}) {
2272 $can_place_holds = 0;
2275 $oldbiblio->{norequests} = 1 unless $can_place_holds;
2276 $oldbiblio->{items_count} = $items_count;
2277 $oldbiblio->{available_items_loop} = \@available_items_loop;
2278 $oldbiblio->{onloan_items_loop} = \@onloan_items_loop;
2279 $oldbiblio->{other_items_loop} = \@other_items_loop;
2280 $oldbiblio->{availablecount} = $available_count;
2281 $oldbiblio->{availableplural} = 1 if $available_count > 1;
2282 $oldbiblio->{onloancount} = $onloan_count;
2283 $oldbiblio->{onloanplural} = 1 if $onloan_count > 1;
2284 $oldbiblio->{othercount} = $other_count;
2285 $oldbiblio->{otherplural} = 1 if $other_count > 1;
2286 $oldbiblio->{withdrawncount} = $withdrawn_count;
2287 $oldbiblio->{itemlostcount} = $itemlost_count;
2288 $oldbiblio->{damagedcount} = $itemdamaged_count;
2289 $oldbiblio->{intransitcount} = $item_in_transit_count;
2290 $oldbiblio->{onholdcount} = $item_onhold_count;
2291 $oldbiblio->{orderedcount} = $ordered_count;
2292 $oldbiblio->{notforloancount} = $notforloan_count;
2294 if (C4::Context->preference("AlternateHoldingsField") && $items_count == 0) {
2295 my $fieldspec = C4::Context->preference("AlternateHoldingsField");
2296 my $subfields = substr $fieldspec, 3;
2297 my $holdingsep = C4::Context->preference("AlternateHoldingsSeparator") || ' ';
2298 my @alternateholdingsinfo = ();
2299 my @holdingsfields = $marcrecord->field(substr $fieldspec, 0, 3);
2300 my $alternateholdingscount = 0;
2302 for my $field (@holdingsfields) {
2303 my %holding = ( holding => '' );
2304 my $havesubfield = 0;
2305 for my $subfield ($field->subfields()) {
2306 if ((index $subfields, $$subfield[0]) >= 0) {
2307 $holding{'holding'} .= $holdingsep if (length $holding{'holding'} > 0);
2308 $holding{'holding'} .= $$subfield[1];
2309 $havesubfield++;
2312 if ($havesubfield) {
2313 push(@alternateholdingsinfo, \%holding);
2314 $alternateholdingscount++;
2318 $oldbiblio->{'ALTERNATEHOLDINGS'} = \@alternateholdingsinfo;
2319 $oldbiblio->{'alternateholdings_count'} = $alternateholdingscount;
2322 push( @newresults, $oldbiblio );
2325 return @newresults;
2328 =head2 enabled_staff_search_views
2330 %hash = enabled_staff_search_views()
2332 This function returns a hash that contains three flags obtained from the system
2333 preferences, used to determine whether a particular staff search results view
2334 is enabled.
2336 =over 2
2338 =item C<Output arg:>
2340 * $hash{can_view_MARC} is true only if the MARC view is enabled
2341 * $hash{can_view_ISBD} is true only if the ISBD view is enabled
2342 * $hash{can_view_labeledMARC} is true only if the Labeled MARC view is enabled
2344 =item C<usage in the script:>
2346 =back
2348 $template->param ( C4::Search::enabled_staff_search_views );
2350 =cut
2352 sub enabled_staff_search_views
2354 return (
2355 can_view_MARC => C4::Context->preference('viewMARC'), # 1 if the staff search allows the MARC view
2356 can_view_ISBD => C4::Context->preference('viewISBD'), # 1 if the staff search allows the ISBD view
2357 can_view_labeledMARC => C4::Context->preference('viewLabeledMARC'), # 1 if the staff search allows the Labeled MARC view
2361 =head2 z3950_search_args
2363 $arrayref = z3950_search_args($matchpoints)
2365 This function returns an array reference that contains the search parameters to be
2366 passed to the Z39.50 search script (z3950_search.pl). The array elements
2367 are hash refs whose keys are name and value, and whose values are the
2368 name of a search parameter, the value of that search parameter and the URL encoded
2369 value of that parameter.
2371 The search parameter names are lccn, isbn, issn, title, author, dewey and subject.
2373 The search parameter values are obtained from the bibliographic record whose
2374 data is in a hash reference in $matchpoints, as returned by Biblio::GetBiblioData().
2376 If $matchpoints is a scalar, it is assumed to be an unnamed query descriptor, e.g.
2377 a general purpose search argument. In this case, the returned array contains only
2378 entry: the key is 'title' and the value is derived from $matchpoints.
2380 If a search parameter value is undefined or empty, it is not included in the returned
2381 array.
2383 The returned array reference may be passed directly to the template parameters.
2385 =over 2
2387 =item C<Output arg:>
2389 * $array containing hash refs as described above
2391 =item C<usage in the script:>
2393 =back
2395 $data = Biblio::GetBiblioData($bibno);
2396 $template->param ( MYLOOP => C4::Search::z3950_search_args($data) )
2398 *OR*
2400 $template->param ( MYLOOP => C4::Search::z3950_search_args($searchscalar) )
2402 =cut
2404 sub z3950_search_args {
2405 my $bibrec = shift;
2407 my $isbn_string = ref( $bibrec ) ? $bibrec->{title} : $bibrec;
2408 my $isbn = Business::ISBN->new( $isbn_string );
2410 if (defined $isbn && $isbn->is_valid)
2412 if ( ref($bibrec) ) {
2413 $bibrec->{isbn} = $isbn_string;
2414 $bibrec->{title} = undef;
2415 } else {
2416 $bibrec = { isbn => $isbn_string };
2419 else {
2420 $bibrec = { title => $bibrec } if !ref $bibrec;
2422 my $array = [];
2423 for my $field (qw/ lccn isbn issn title author dewey subject /)
2425 push @$array, { name => $field, value => $bibrec->{$field} }
2426 if defined $bibrec->{$field};
2428 return $array;
2431 =head2 GetDistinctValues($field);
2433 C<$field> is a reference to the fields array
2435 =cut
2437 sub GetDistinctValues {
2438 my ($fieldname,$string)=@_;
2439 # returns a reference to a hash of references to branches...
2440 if ($fieldname=~/\./){
2441 my ($table,$column)=split /\./, $fieldname;
2442 my $dbh = C4::Context->dbh;
2443 warn "select DISTINCT($column) as value, count(*) as cnt from $table group by lib order by $column " if $DEBUG;
2444 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 ");
2445 $sth->execute;
2446 my $elements=$sth->fetchall_arrayref({});
2447 return $elements;
2449 else {
2450 $string||= qq("");
2451 my @servers=qw<biblioserver authorityserver>;
2452 my (@zconns,@results);
2453 for ( my $i = 0 ; $i < @servers ; $i++ ) {
2454 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
2455 $results[$i] =
2456 $zconns[$i]->scan(
2457 ZOOM::Query::CCL2RPN->new( qq"$fieldname $string", $zconns[$i])
2460 # The big moment: asynchronously retrieve results from all servers
2461 my @elements;
2462 _ZOOM_event_loop(
2463 \@zconns,
2464 \@results,
2465 sub {
2466 my ( $i, $size ) = @_;
2467 for ( my $j = 0 ; $j < $size ; $j++ ) {
2468 my %hashscan;
2469 @hashscan{qw(value cnt)} =
2470 $results[ $i - 1 ]->display_term($j);
2471 push @elements, \%hashscan;
2475 return \@elements;
2479 =head2 _ZOOM_event_loop
2481 _ZOOM_event_loop(\@zconns, \@results, sub {
2482 my ( $i, $size ) = @_;
2483 ....
2484 } );
2486 Processes a ZOOM event loop and passes control to a closure for
2487 processing the results, and destroying the resultsets.
2489 =cut
2491 sub _ZOOM_event_loop {
2492 my ($zconns, $results, $callback) = @_;
2493 while ( ( my $i = ZOOM::event( $zconns ) ) != 0 ) {
2494 my $ev = $zconns->[ $i - 1 ]->last_event();
2495 if ( $ev == ZOOM::Event::ZEND ) {
2496 next unless $results->[ $i - 1 ];
2497 my $size = $results->[ $i - 1 ]->size();
2498 if ( $size > 0 ) {
2499 $callback->($i, $size);
2504 foreach my $result (@$results) {
2505 $result->destroy();
2509 =head2 new_record_from_zebra
2511 Given raw data from a searchengine result set, return a MARC::Record object
2513 This helper function is needed to take into account all the involved
2514 system preferences and configuration variables to properly create the
2515 MARC::Record object.
2517 If we are using GRS-1, then the raw data we get from Zebra should be USMARC
2518 data. If we are using DOM, then it has to be MARCXML.
2520 If we are using elasticsearch, it'll already be a MARC::Record and this
2521 function needs a new name.
2523 =cut
2525 sub new_record_from_zebra {
2527 my $server = shift;
2528 my $raw_data = shift;
2529 # Set the default indexing modes
2530 my $search_engine = C4::Context->preference("SearchEngine");
2531 if ($search_engine eq 'Elasticsearch') {
2532 return ref $raw_data eq 'MARC::Record' ? $raw_data : MARC::Record->new_from_xml( $raw_data, 'UTF-8' );
2534 my $index_mode = ( $server eq 'biblioserver' )
2535 ? C4::Context->config('zebra_bib_index_mode') // 'dom'
2536 : C4::Context->config('zebra_auth_index_mode') // 'dom';
2538 my $marc_record = eval {
2539 if ( $index_mode eq 'dom' ) {
2540 MARC::Record->new_from_xml( $raw_data, 'UTF-8' );
2541 } else {
2542 MARC::Record->new_from_usmarc( $raw_data );
2546 if ($@) {
2547 return;
2548 } else {
2549 return $marc_record;
2554 END { } # module clean-up code here (global destructor)
2557 __END__
2559 =head1 AUTHOR
2561 Koha Development Team <http://koha-community.org/>
2563 =cut