?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Bug 25642: Fix description of new feature in update_dbix_class_files.pl
[koha.git] / C4 / Search.pm
blob8997dba0648dc714d07e5e7b1115f6ce10274b94
1 package C4::Search;
2
3 # This file is part of Koha.
4 #
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.
9 #
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.
14 #
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>.
17
18 use Modern::Perl;
19 require Exporter;
20 use C4::Context;
21 use C4::Biblio; # GetMarcFromKohaField, GetBiblioData
22 use C4::Koha; # getFacets
23 use Koha::DateUtils;
24 use Koha::Libraries;
25 use Lingua::Stem;
26 use C4::Search::PazPar2;
27 use XML::Simple;
28 use C4::XSLT;
29 use C4::Reserves; # GetReserveStatus
30 use C4::Debug;
31 use C4::Charset;
32 use Koha::AuthorisedValues;
33 use Koha::ItemTypes;
34 use Koha::Libraries;
35 use Koha::Patrons;
36 use Koha::RecordProcessor;
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);
43
44 BEGIN {
45 $DEBUG = ($ENV{DEBUG}) ? 1 : 0;
46 }
47
48 =head1 NAME
49
50 C4::Search - Functions for searching the Koha catalog.
51
52 =head1 SYNOPSIS
53
54 See opac/opac-search.pl or catalogue/search.pl for example of usage
55
56 =head1 DESCRIPTION
57
58 This module provides searching functions for Koha's bibliographic databases
59
60 =head1 FUNCTIONS
61
62 =cut
63
64 @ISA = qw(Exporter);
65 @EXPORT = qw(
66 &FindDuplicate
67 &SimpleSearch
68 &searchResults
69 &getRecords
70 &buildQuery
71 &GetDistinctValues
72 &enabled_staff_search_views
73 );
74
75 # make all your functions, whether exported or not;
76
77 =head2 FindDuplicate
78
79 ($biblionumber,$biblionumber,$title) = FindDuplicate($record);
80
81 This function attempts to find duplicate records using a hard-coded, fairly simplistic algorithm
82
83 =cut
84
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 );
94
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}";
101 }
102 else {
103
104 my $titleindex = 'ti,ext';
105 my $authorindex = 'au,ext';
106 my $op = 'and';
107
108 $result->{title} =~ s /\\//g;
109 $result->{title} =~ s /\"//g;
110 $result->{title} =~ s /\(//g;
111 $result->{title} =~ s /\)//g;
112
113 # FIXME: instead of removing operators, could just do
114 # quotes around the value
115 $result->{title} =~ s/(and|or|not)//g;
116 $query = "$titleindex:\"$result->{title}\"";
117 if ( $result->{author} ) {
118 $result->{author} =~ s /\\//g;
119 $result->{author} =~ s /\"//g;
120 $result->{author} =~ s /\(//g;
121 $result->{author} =~ s /\)//g;
122
123 # remove valid operators
124 $result->{author} =~ s/(and|or|not)//g;
125 $query .= " $op $authorindex:\"$result->{author}\"";
126 }
127 }
128
129 my ( $error, $searchresults, undef ) = SimpleSearch($query); # FIXME :: hardcoded !
130 my @results;
131 if (!defined $error) {
132 foreach my $possible_duplicate_record (@{$searchresults}) {
133 my $marcrecord = new_record_from_zebra(
134 'biblioserver',
135 $possible_duplicate_record
136 );
137
138 my $result = TransformMarcToKoha( $marcrecord, '' );
139
140 # FIXME :: why 2 $biblionumber ?
141 if ($result) {
142 push @results, $result->{'biblionumber'};
143 push @results, $result->{'title'};
144 }
145 }
146 }
147 return @results;
148 }
149
150 =head2 SimpleSearch
151
152 ( $error, $results, $total_hits ) = SimpleSearch( $query, $offset, $max_results, [@servers], [%options] );
153
154 This function provides a simple search API on the bibliographic catalog
155
156 =over 2
157
158 =item C<input arg:>
159
160 * $query can be a simple keyword or a complete CCL query
161 * @servers is optional. Defaults to biblioserver as found in koha-conf.xml
162 * $offset - If present, represents the number of records at the beginning to omit. Defaults to 0
163 * $max_results - if present, determines the maximum number of records to fetch. undef is All. defaults to undef.
164 * %options is optional. (e.g. "skip_normalize" allows you to skip changing : to = )
165
166
167 =item C<Return:>
168
169 Returns an array consisting of three elements
170 * $error is undefined unless an error is detected
171 * $results is a reference to an array of records.
172 * $total_hits is the number of hits that would have been returned with no limit
173
174 If an error is returned the two other return elements are undefined. If error itself is undefined
175 the other two elements are always defined
176
177 =item C<usage in the script:>
178
179 =back
180
181 my ( $error, $marcresults, $total_hits ) = SimpleSearch($query);
182
183 if (defined $error) {
184 $template->param(query_error => $error);
185 warn "error: ".$error;
186 output_html_with_http_headers $input, $cookie, $template->output;
187 exit;
188 }
189
190 my $hits = @{$marcresults};
191 my @results;
192
193 for my $r ( @{$marcresults} ) {
194 my $marcrecord = MARC::File::USMARC::decode($r);
195 my $biblio = TransformMarcToKoha($marcrecord,q{});
196
197 #build the iarray of hashs for the template.
198 push @results, {
199 title => $biblio->{'title'},
200 subtitle => $biblio->{'subtitle'},
201 biblionumber => $biblio->{'biblionumber'},
202 author => $biblio->{'author'},
203 publishercode => $biblio->{'publishercode'},
204 publicationyear => $biblio->{'publicationyear'},
205 };
206
207 }
208
209 $template->param(result=>\@results);
210
211 =cut
212
213 sub SimpleSearch {
214 my ( $query, $offset, $max_results, $servers, %options ) = @_;
215
216 return ( 'No query entered', undef, undef ) unless $query;
217 # FIXME hardcoded value. See catalog/search.pl & opac-search.pl too.
218 my @servers = defined ( $servers ) ? @$servers : ( 'biblioserver' );
219 my @zoom_queries;
220 my @tmpresults;
221 my @zconns;
222 my $results = [];
223 my $total_hits = 0;
224
225 # Initialize & Search Zebra
226 for ( my $i = 0 ; $i < @servers ; $i++ ) {
227 eval {
228 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
229 $query =~ s/:/=/g unless $options{skip_normalize};
230 $zoom_queries[$i] = new ZOOM::Query::CCL2RPN( $query, $zconns[$i]);
231 $tmpresults[$i] = $zconns[$i]->search( $zoom_queries[$i] );
232
233 # error handling
234 my $error =
235 $zconns[$i]->errmsg() . " ("
236 . $zconns[$i]->errcode() . ") "
237 . $zconns[$i]->addinfo() . " "
238 . $zconns[$i]->diagset();
239
240 return ( $error, undef, undef ) if $zconns[$i]->errcode();
241 };
242 if ($@) {
243
244 # caught a ZOOM::Exception
245 my $error =
246 $@->message() . " ("
247 . $@->code() . ") "
248 . $@->addinfo() . " "
249 . $@->diagset();
250 warn $error." for query: $query";
251 return ( $error, undef, undef );
252 }
253 }
254
255 _ZOOM_event_loop(
256 \@zconns,
257 \@tmpresults,
258 sub {
259 my ($i, $size) = @_;
260 my $first_record = defined($offset) ? $offset + 1 : 1;
261 my $hits = $tmpresults[ $i - 1 ]->size();
262 $total_hits += $hits;
263 my $last_record = $hits;
264 if ( defined $max_results && $offset + $max_results < $hits ) {
265 $last_record = $offset + $max_results;
266 }
267
268 for my $j ( $first_record .. $last_record ) {
269 my $record = eval {
270 $tmpresults[ $i - 1 ]->record( $j - 1 )->raw()
271 ; # 0 indexed
272 };
273 push @{$results}, $record if defined $record;
274 }
275 }
276 );
277
278 foreach my $zoom_query (@zoom_queries) {
279 $zoom_query->destroy();
280 }
281
282 return ( undef, $results, $total_hits );
283 }
284
285 =head2 getRecords
286
287 ( undef, $results_hashref, \@facets_loop ) = getRecords (
288
289 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
290 $results_per_page, $offset, $branches, $itemtypes,
291 $query_type, $scan, $opac
292 );
293
294 The all singing, all dancing, multi-server, asynchronous, scanning,
295 searching, record nabbing, facet-building
296
297 See verbose embedded documentation.
298
299 =cut
300
301 sub getRecords {
302 my (
303 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
304 $results_per_page, $offset, $branches, $itemtypes,
305 $query_type, $scan, $opac
306 ) = @_;
307
308 my @servers = @$servers_ref;
309 my @sort_by = @$sort_by_ref;
310 $offset = 0 if $offset < 0;
311
312 # Initialize variables for the ZOOM connection and results object
313 my $zconn;
314 my @zconns;
315 my @results;
316 my $results_hashref = ();
317
318 # TODO simplify this structure ( { branchcode => $branchname } is enought) and remove this parameter
319 $branches ||= { map { $_->branchcode => { branchname => $_->branchname } } Koha::Libraries->search };
320
321 # Initialize variables for the faceted results objects
322 my $facets_counter = {};
323 my $facets_info = {};
324 my $facets = getFacets();
325
326 my @facets_loop; # stores the ref to array of hashes for template facets loop
327
328 ### LOOP THROUGH THE SERVERS
329 for ( my $i = 0 ; $i < @servers ; $i++ ) {
330 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
331
332 # perform the search, create the results objects
333 # if this is a local search, use the $koha-query, if it's a federated one, use the federated-query
334 my $query_to_use = ($servers[$i] =~ /biblioserver/) ? $koha_query : $simple_query;
335
336 #$query_to_use = $simple_query if $scan;
337 warn $simple_query if ( $scan and $DEBUG );
338
339 # Check if we've got a query_type defined, if so, use it
340 eval {
341 if ($query_type) {
342 if ($query_type =~ /^ccl/) {
343 $query_to_use =~ s/\:/\=/g; # change : to = last minute (FIXME)
344 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
345 } elsif ($query_type =~ /^cql/) {
346 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CQL($query_to_use, $zconns[$i]));
347 } elsif ($query_type =~ /^pqf/) {
348 $results[$i] = $zconns[$i]->search(new ZOOM::Query::PQF($query_to_use, $zconns[$i]));
349 } else {
350 warn "Unknown query_type '$query_type'. Results undetermined.";
351 }
352 } elsif ($scan) {
353 $results[$i] = $zconns[$i]->scan( new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
354 } else {
355 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
356 }
357 };
358 if ($@) {
359 warn "WARNING: query problem with $query_to_use " . $@;
360 }
361
362 # Concatenate the sort_by limits and pass them to the results object
363 # Note: sort will override rank
364 my $sort_by;
365 foreach my $sort (@sort_by) {
366 if ( $sort eq "author_az" || $sort eq "author_asc" ) {
367 $sort_by .= "1=1003 <i ";
368 }
369 elsif ( $sort eq "author_za" || $sort eq "author_dsc" ) {
370 $sort_by .= "1=1003 >i ";
371 }
372 elsif ( $sort eq "popularity_asc" ) {
373 $sort_by .= "1=9003 <i ";
374 }
375 elsif ( $sort eq "popularity_dsc" ) {
376 $sort_by .= "1=9003 >i ";
377 }
378 elsif ( $sort eq "call_number_asc" ) {
379 $sort_by .= "1=8007 <i ";
380 }
381 elsif ( $sort eq "call_number_dsc" ) {
382 $sort_by .= "1=8007 >i ";
383 }
384 elsif ( $sort eq "pubdate_asc" ) {
385 $sort_by .= "1=31 <i ";
386 }
387 elsif ( $sort eq "pubdate_dsc" ) {
388 $sort_by .= "1=31 >i ";
389 }
390 elsif ( $sort eq "acqdate_asc" ) {
391 $sort_by .= "1=32 <i ";
392 }
393 elsif ( $sort eq "acqdate_dsc" ) {
394 $sort_by .= "1=32 >i ";
395 }
396 elsif ( $sort eq "title_az" || $sort eq "title_asc" ) {
397 $sort_by .= "1=4 <i ";
398 }
399 elsif ( $sort eq "title_za" || $sort eq "title_dsc" ) {
400 $sort_by .= "1=4 >i ";
401 }
402 else {
403 warn "Ignoring unrecognized sort '$sort' requested" if $sort_by;
404 }
405 }
406 if ( $sort_by && !$scan && $results[$i] ) {
407 if ( $results[$i]->sort( "yaz", $sort_by ) < 0 ) {
408 warn "WARNING sort $sort_by failed";
409 }
410 }
411 } # finished looping through servers
412
413 # The big moment: asynchronously retrieve results from all servers
414 _ZOOM_event_loop(
415 \@zconns,
416 \@results,
417 sub {
418 my ( $i, $size ) = @_;
419 my $results_hash;
420
421 # loop through the results
422 $results_hash->{'hits'} = $size;
423 my $times;
424 if ( $offset + $results_per_page <= $size ) {
425 $times = $offset + $results_per_page;
426 }
427 else {
428 $times = $size;
429 }
430
431 for ( my $j = $offset ; $j < $times ; $j++ ) {
432 my $records_hash;
433 my $record;
434
435 ## Check if it's an index scan
436 if ($scan) {
437 my ( $term, $occ ) = $results[ $i - 1 ]->display_term($j);
438
439 # here we create a minimal MARC record and hand it off to the
440 # template just like a normal result ... perhaps not ideal, but
441 # it works for now
442 my $tmprecord = MARC::Record->new();
443 $tmprecord->encoding('UTF-8');
444 my $tmptitle;
445 my $tmpauthor;
446
447 # the minimal record in author/title (depending on MARC flavour)
448 if ( C4::Context->preference("marcflavour") eq
449 "UNIMARC" )
450 {
451 $tmptitle = MARC::Field->new(
452 '200', ' ', ' ',
453 a => $term,
454 f => $occ
455 );
456 $tmprecord->append_fields($tmptitle);
457 }
458 else {
459 $tmptitle =
460 MARC::Field->new( '245', ' ', ' ', a => $term, );
461 $tmpauthor =
462 MARC::Field->new( '100', ' ', ' ', a => $occ, );
463 $tmprecord->append_fields($tmptitle);
464 $tmprecord->append_fields($tmpauthor);
465 }
466 $results_hash->{'RECORDS'}[$j] =
467 $tmprecord->as_usmarc();
468 }
469
470 # not an index scan
471 else {
472 $record = $results[ $i - 1 ]->record($j)->raw();
473 # warn "RECORD $j:".$record;
474 $results_hash->{'RECORDS'}[$j] = $record;
475 }
476
477 }
478 $results_hashref->{ $servers[ $i - 1 ] } = $results_hash;
479
480 # Fill the facets while we're looping, but only for the
481 # biblioserver and not for a scan
482 if ( !$scan && $servers[ $i - 1 ] =~ /biblioserver/ ) {
483 $facets_counter = GetFacets( $results[ $i - 1 ] );
484 $facets_info = _get_facets_info( $facets );
485 }
486
487 # BUILD FACETS
488 if ( $servers[ $i - 1 ] =~ /biblioserver/ ) {
489 for my $link_value (
490 sort { $a cmp $b } keys %$facets_counter
491 )
492 {
493 my @this_facets_array;
494 for my $one_facet (
495 sort {
496 $facets_counter->{$link_value}
497 ->{$b} <=> $facets_counter->{$link_value}
498 ->{$a}
499 } keys %{ $facets_counter->{$link_value} }
500 )
501 {
502 # Sanitize the link value : parenthesis, question and exclamation mark will cause errors with CCL
503 my $facet_link_value = $one_facet;
504 $facet_link_value =~ s/[()!?¡¿؟]/ /g;
505
506 # fix the length that will display in the label,
507 my $facet_label_value = $one_facet;
508 my $facet_max_length = C4::Context->preference(
509 'FacetLabelTruncationLength')
510 || 20;
511 $facet_label_value =
512 substr( $one_facet, 0, $facet_max_length )
513 . "..."
514 if length($facet_label_value) >
515 $facet_max_length;
516
517 # if it's a branch, label by the name, not the code,
518 if ( $link_value =~ /branch/ ) {
519 if ( defined $branches
520 && ref($branches) eq "HASH"
521 && defined $branches->{$one_facet}
522 && ref( $branches->{$one_facet} ) eq
523 "HASH" )
524 {
525 $facet_label_value =
526 $branches->{$one_facet}
527 ->{'branchname'};
528 }
529 else {
530 $facet_label_value = "*";
531 }
532 }
533
534 # if it's a itemtype, label by the name, not the code,
535 if ( $link_value =~ /itype/ ) {
536 if ( defined $itemtypes
537 && ref($itemtypes) eq "HASH"
538 && defined $itemtypes->{$one_facet}
539 && ref( $itemtypes->{$one_facet} ) eq
540 "HASH" )
541 {
542 $facet_label_value =
543 $itemtypes->{$one_facet}
544 ->{translated_description};
545 }
546 }
547
548 # also, if it's a location code, use the name instead of the code
549 if ( $link_value =~ /location/ ) {
550 # TODO Retrieve all authorised values at once, instead of 1 query per entry
551 my $av = Koha::AuthorisedValues->search({ category => 'LOC', authorised_value => $one_facet });
552 $facet_label_value = $av->count ? $av->next->opac_description : '';
553 }
554
555 # also, if it's a collection code, use the name instead of the code
556 if ( $link_value =~ /ccode/ ) {
557 # TODO Retrieve all authorised values at once, instead of 1 query per entry
558 my $av = Koha::AuthorisedValues->search({ category => 'CCODE', authorised_value => $one_facet });
559 $facet_label_value = $av->count ? $av->next->opac_description : '';
560 }
561
562 # but we're down with the whole label being in the link's title.
563 push @this_facets_array,
564 {
565 facet_count =>
566 $facets_counter->{$link_value}
567 ->{$one_facet},
568 facet_label_value => $facet_label_value,
569 facet_title_value => $one_facet,
570 facet_link_value => $facet_link_value,
571 type_link_value => $link_value,
572 }
573 if ($facet_label_value);
574 }
575
576 push @facets_loop,
577 {
578 type_link_value => $link_value,
579 type_id => $link_value . "_id",
580 "type_label_"
581 . $facets_info->{$link_value}->{'label_value'} =>
582 1,
583 facets => \@this_facets_array,
584 }
585 unless (
586 (
587 $facets_info->{$link_value}->{'label_value'} =~
588 /Libraries/
589 )
590 and ( Koha::Libraries->search->count == 1 )
591 );
592 }
593 }
594 }
595 );
596
597 # This sorts the facets into alphabetical order
598 if (@facets_loop) {
599 foreach my $f (@facets_loop) {
600 $f->{facets} = [ sort { uc($a->{facet_label_value}) cmp uc($b->{facet_label_value}) } @{ $f->{facets} } ];
601 }
602 }
603
604 return ( undef, $results_hashref, \@facets_loop );
605 }
606
607 sub GetFacets {
608
609 my $rs = shift;
610 my $facets;
611
612 my $use_zebra_facets = C4::Context->config('use_zebra_facets') // 0;
613
614 if ( $use_zebra_facets ) {
615 $facets = _get_facets_from_zebra( $rs );
616 } else {
617 $facets = _get_facets_from_records( $rs );
618 }
619
620 return $facets;
621 }
622
623 sub _get_facets_from_records {
624
625 my $rs = shift;
626
627 my $facets_maxrecs = C4::Context->preference('maxRecordsForFacets') // 20;
628 my $facets_config = getFacets();
629 my $facets = {};
630 my $size = $rs->size();
631 my $jmax = $size > $facets_maxrecs
632 ? $facets_maxrecs
633 : $size;
634
635 for ( my $j = 0 ; $j < $jmax ; $j++ ) {
636
637 my $marc_record = new_record_from_zebra (
638 'biblioserver',
639 $rs->record( $j )->raw()
640 );
641
642 if ( ! defined $marc_record ) {
643 warn "ERROR DECODING RECORD - $@: " .
644 $rs->record( $j )->raw();
645 next;
646 }
647
648 _get_facets_data_from_record( $marc_record, $facets_config, $facets );
649 }
650
651 return $facets;
652 }
653
654 =head2 _get_facets_data_from_record
655
656 C4::Search::_get_facets_data_from_record( $marc_record, $facets, $facets_counter );
657
658 Internal function that extracts facets information from a MARC::Record object
659 and populates $facets_counter for using in getRecords.
660
661 $facets is expected to be filled with C4::Koha::getFacets output (i.e. the configured
662 facets for Zebra).
663
664 =cut
665
666 sub _get_facets_data_from_record {
667
668 my ( $marc_record, $facets, $facets_counter ) = @_;
669
670 for my $facet (@$facets) {
671
672 my @used_datas = ();
673
674 foreach my $tag ( @{ $facet->{ tags } } ) {
675
676 # tag number is the first three digits
677 my $tag_num = substr( $tag, 0, 3 );
678 # subfields are the remainder
679 my $subfield_letters = substr( $tag, 3 );
680
681 my @fields = $marc_record->field( $tag_num );
682 foreach my $field (@fields) {
683 # If $field->indicator(1) eq 'z', it means it is a 'see from'
684 # field introduced because of IncludeSeeFromInSearches, so skip it
685 next if $field->indicator(1) eq 'z';
686
687 my $data = $field->as_string( $subfield_letters, $facet->{ sep } );
688
689 unless ( grep { $_ eq $data } @used_datas ) {
690 push @used_datas, $data;
691 $facets_counter->{ $facet->{ idx } }->{ $data }++;
692 }
693 }
694 }
695 }
696 }
697
698 =head2 _get_facets_from_zebra
699
700 my $facets = _get_facets_from_zebra( $result_set )
701
702 Retrieves facets for a specified result set. It loops through the facets defined
703 in C4::Koha::getFacets and returns a hash with the following structure:
704
705 { facet_idx => {
706 facet_value => count
707 },
708 ...
709 }
710
711 =cut
712
713 sub _get_facets_from_zebra {
714
715 my $rs = shift;
716
717 # save current elementSetName
718 my $elementSetName = $rs->option( 'elementSetName' );
719
720 my $facets_loop = getFacets();
721 my $facets_data = {};
722 # loop through defined facets and fill the facets hashref
723 foreach my $facet ( @$facets_loop ) {
724
725 my $idx = $facet->{ idx };
726 my $sep = $facet->{ sep };
727 my $facet_values = _get_facet_from_result_set( $idx, $rs, $sep );
728 if ( $facet_values ) {
729 # we've actually got a result
730 $facets_data->{ $idx } = $facet_values;
731 }
732 }
733 # set elementSetName to its previous value to avoid side effects
734 $rs->option( elementSetName => $elementSetName );
735
736 return $facets_data;
737 }
738
739 =head2 _get_facet_from_result_set
740
741 my $facet_values =
742 C4::Search::_get_facet_from_result_set( $facet_idx, $result_set, $sep )
743
744 Internal function that extracts facet information for a specific index ($facet_idx) and
745 returns a hash containing facet values and count:
746
747 {
748 $facet_value => $count ,
749 ...
750 }
751
752 Warning: this function has the side effect of changing the elementSetName for the result
753 set. It is a helper function for the main loop, which takes care of backing it up for
754 restoring.
755
756 =cut
757
758 sub _get_facet_from_result_set {
759
760 my $facet_idx = shift;
761 my $rs = shift;
762 my $sep = shift;
763
764 my $internal_sep = '<*>';
765 my $facetMaxCount = C4::Context->preference('FacetMaxCount') // 20;
766
767 return if ( ! defined $facet_idx || ! defined $rs );
768 # zebra's facet element, untokenized index
769 my $facet_element = 'zebra::facet::' . $facet_idx . ':0:' . $facetMaxCount;
770 # configure zebra results for retrieving the desired facet
771 $rs->option( elementSetName => $facet_element );
772 # get the facet record from result set
773 my $facet = $rs->record( 0 )->raw;
774 # if the facet has no restuls...
775 return if !defined $facet;
776 # TODO: benchmark DOM vs. SAX performance
777 my $facet_dom = XML::LibXML->load_xml(
778 string => ($facet)
779 );
780 my @terms = $facet_dom->getElementsByTagName('term');
781 return if ! @terms;
782
783 my $facets = {};
784 foreach my $term ( @terms ) {
785 my $facet_value = $term->textContent;
786 $facet_value =~ s/\Q$internal_sep\E/$sep/ if defined $sep;
787 $facets->{ $facet_value } = $term->getAttribute( 'occur' );
788 }
789
790 return $facets;
791 }
792
793 =head2 _get_facets_info
794
795 my $facets_info = C4::Search::_get_facets_info( $facets )
796
797 Internal function that extracts facets information and properly builds
798 the data structure needed to render facet labels.
799
800 =cut
801
802 sub _get_facets_info {
803
804 my $facets = shift;
805
806 my $facets_info = {};
807
808 for my $facet ( @$facets ) {
809 $facets_info->{ $facet->{ idx } }->{ label_value } = $facet->{ label };
810 }
811
812 return $facets_info;
813 }
814
815 sub pazGetRecords {
816 my (
817 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
818 $results_per_page, $offset, $branches, $query_type,
819 $scan
820 ) = @_;
821 #NOTE: Parameter $branches is not used here !
822
823 my $paz = C4::Search::PazPar2->new(C4::Context->config('pazpar2url'));
824 $paz->init();
825 $paz->search($simple_query);
826 sleep 1; # FIXME: WHY?
827
828 # do results
829 my $results_hashref = {};
830 my $stats = XMLin($paz->stat);
831 my $results = XMLin($paz->show($offset, $results_per_page, 'work-title:1'), forcearray => 1);
832
833 # for a grouped search result, the number of hits
834 # is the number of groups returned; 'bib_hits' will have
835 # the total number of bibs.
836 $results_hashref->{'biblioserver'}->{'hits'} = $results->{'merged'}->[0];
837 $results_hashref->{'biblioserver'}->{'bib_hits'} = $stats->{'hits'};
838
839 HIT: foreach my $hit (@{ $results->{'hit'} }) {
840 my $recid = $hit->{recid}->[0];
841
842 my $work_title = $hit->{'md-work-title'}->[0];
843 my $work_author;
844 if (exists $hit->{'md-work-author'}) {
845 $work_author = $hit->{'md-work-author'}->[0];
846 }
847 my $group_label = (defined $work_author) ? "$work_title / $work_author" : $work_title;
848
849 my $result_group = {};
850 $result_group->{'group_label'} = $group_label;
851 $result_group->{'group_merge_key'} = $recid;
852
853 my $count = 1;
854 if (exists $hit->{count}) {
855 $count = $hit->{count}->[0];
856 }
857 $result_group->{'group_count'} = $count;
858
859 for (my $i = 0; $i < $count; $i++) {
860 # FIXME -- may need to worry about diacritics here
861 my $rec = $paz->record($recid, $i);
862 push @{ $result_group->{'RECORDS'} }, $rec;
863 }
864
865 push @{ $results_hashref->{'biblioserver'}->{'GROUPS'} }, $result_group;
866 }
867
868 # pass through facets
869 my $termlist_xml = $paz->termlist('author,subject');
870 my $terms = XMLin($termlist_xml, forcearray => 1);
871 my @facets_loop = ();
872 #die Dumper($results);
873 # foreach my $list (sort keys %{ $terms->{'list'} }) {
874 # my @facets = ();
875 # foreach my $facet (sort @{ $terms->{'list'}->{$list}->{'term'} } ) {
876 # push @facets, {
877 # facet_label_value => $facet->{'name'}->[0],
878 # };
879 # }
880 # push @facets_loop, ( {
881 # type_label => $list,
882 # facets => \@facets,
883 # } );
884 # }
885
886 return ( undef, $results_hashref, \@facets_loop );
887 }
888
889 # TRUNCATION
890 sub _detect_truncation {
891 my ( $operand, $index ) = @_;
892 my ( @nontruncated, @righttruncated, @lefttruncated, @rightlefttruncated,
893 @regexpr );
894 $operand =~ s/^ //g;
895 my @wordlist = split( /\s/, $operand );
896 foreach my $word (@wordlist) {
897 if ( $word =~ s/^\*([^\*]+)\*$/$1/ ) {
898 push @rightlefttruncated, $word;
899 }
900 elsif ( $word =~ s/^\*([^\*]+)$/$1/ ) {
901 push @lefttruncated, $word;
902 }
903 elsif ( $word =~ s/^([^\*]+)\*$/$1/ ) {
904 push @righttruncated, $word;
905 }
906 elsif ( index( $word, "*" ) < 0 ) {
907 push @nontruncated, $word;
908 }
909 else {
910 push @regexpr, $word;
911 }
912 }
913 return (
914 \@nontruncated, \@righttruncated, \@lefttruncated,
915 \@rightlefttruncated, \@regexpr
916 );
917 }
918
919 # STEMMING
920 sub _build_stemmed_operand {
921 my ($operand,$lang) = @_;
922 require Lingua::Stem::Snowball ;
923 my $stemmed_operand=q{};
924
925 # Stemmer needs language
926 return $operand unless $lang;
927
928 # If operand contains a digit, it is almost certainly an identifier, and should
929 # not be stemmed. This is particularly relevant for ISBNs and ISSNs, which
930 # can contain the letter "X" - for example, _build_stemmend_operand would reduce
931 # "014100018X" to "x ", which for a MARC21 database would bring up irrelevant
932 # results (e.g., "23 x 29 cm." from the 300$c). Bug 2098.
933 return $operand if $operand =~ /\d/;
934
935 # FIXME: the locale should be set based on the user's language and/or search choice
936 #warn "$lang";
937 # Make sure we only use the first two letters from the language code
938 $lang = lc(substr($lang, 0, 2));
939 # The language codes for the two variants of Norwegian will now be "nb" and "nn",
940 # none of which Lingua::Stem::Snowball can use, so we need to "translate" them
941 if ($lang eq 'nb' || $lang eq 'nn') {
942 $lang = 'no';
943 }
944 my $stemmer = Lingua::Stem::Snowball->new( lang => $lang,
945 encoding => "UTF-8" );
946
947 my @words = split( / /, $operand );
948 my @stems = $stemmer->stem(\@words);
949 for my $stem (@stems) {
950 $stemmed_operand .= "$stem";
951 $stemmed_operand .= "?"
952 unless ( $stem =~ /(and$|or$|not$)/ ) || ( length($stem) < 3 );
953 $stemmed_operand .= " ";
954 }
955 warn "STEMMED OPERAND: $stemmed_operand" if $DEBUG;
956 return $stemmed_operand;
957 }
958
959 # FIELD WEIGHTING
960 sub _build_weighted_query {
961
962 # FIELD WEIGHTING - This is largely experimental stuff. What I'm committing works
963 # pretty well but could work much better if we had a smarter query parser
964 my ( $operand, $stemmed_operand, $index ) = @_;
965 my $stemming = C4::Context->preference("QueryStemming") || 0;
966 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
967 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
968 $operand =~ s/"/ /g; # Bug 7518: searches with quotation marks don't work
969
970 my $weighted_query .= "(rk=("; # Specifies that we're applying rank
971
972 # Keyword, or, no index specified
973 if ( ( $index eq 'kw' ) || ( !$index ) ) {
974 $weighted_query .=
975 "Title-cover,ext,r1=\"$operand\""; # exact title-cover
976 $weighted_query .= " or ti,ext,r2=\"$operand\""; # exact title
977 $weighted_query .= " or Title-cover,phr,r3=\"$operand\""; # phrase title
978 $weighted_query .= " or ti,wrdl,r4=\"$operand\""; # words in title
979 #$weighted_query .= " or any,ext,r4=$operand"; # exact any
980 #$weighted_query .=" or kw,wrdl,r5=\"$operand\""; # word list any
981 $weighted_query .= " or wrdl,fuzzy,r8=\"$operand\""
982 if $fuzzy_enabled; # add fuzzy, word list
983 $weighted_query .= " or wrdl,right-Truncation,r9=\"$stemmed_operand\""
984 if ( $stemming and $stemmed_operand )
985 ; # add stemming, right truncation
986 $weighted_query .= " or wrdl,r9=\"$operand\"";
987
988 # embedded sorting: 0 a-z; 1 z-a
989 # $weighted_query .= ") or (sort1,aut=1";
990 }
991
992 # Barcode searches should skip this process
993 elsif ( $index eq 'bc' ) {
994 $weighted_query .= "bc=\"$operand\"";
995 }
996
997 # Authority-number searches should skip this process
998 elsif ( $index eq 'an' ) {
999 $weighted_query .= "an=\"$operand\"";
1000 }
1001
1002 # If the index is numeric, don't autoquote it.
1003 elsif ( $index =~ /,st-numeric$/ ) {
1004 $weighted_query .= " $index=$operand";
1005 }
1006
1007 # If the index already has more than one qualifier, wrap the operand
1008 # in quotes and pass it back (assumption is that the user knows what they
1009 # are doing and won't appreciate us mucking up their query
1010 elsif ( $index =~ ',' ) {
1011 $weighted_query .= " $index=\"$operand\"";
1012 }
1013
1014 #TODO: build better cases based on specific search indexes
1015 else {
1016 $weighted_query .= " $index,ext,r1=\"$operand\""; # exact index
1017 #$weighted_query .= " or (title-sort-az=0 or $index,startswithnt,st-word,r3=$operand #)";
1018 $weighted_query .= " or $index,phr,r3=\"$operand\""; # phrase index
1019 $weighted_query .= " or $index,wrdl,r6=\"$operand\""; # word list index
1020 $weighted_query .= " or $index,wrdl,fuzzy,r8=\"$operand\""
1021 if $fuzzy_enabled; # add fuzzy, word list
1022 $weighted_query .= " or $index,wrdl,rt,r9=\"$stemmed_operand\""
1023 if ( $stemming and $stemmed_operand ); # add stemming, right truncation
1024 }
1025
1026 $weighted_query .= "))"; # close rank specification
1027 return $weighted_query;
1028 }
1029
1030 =head2 getIndexes
1031
1032 Return an array with available indexes.
1033
1034 =cut
1035
1036 sub getIndexes{
1037 my @indexes = (
1038 # biblio indexes
1039 'ab',
1040 'Abstract',
1041 'acqdate',
1042 'allrecords',
1043 'an',
1044 'Any',
1045 'at',
1046 'arl',
1047 'arp',
1048 'au',
1049 'aub',
1050 'aud',
1051 'audience',
1052 'auo',
1053 'aut',
1054 'Author',
1055 'Author-in-order ',
1056 'Author-personal-bibliography',
1057 'Authority-Number',
1058 'authtype',
1059 'bc',
1060 'Bib-level',
1061 'biblionumber',
1062 'bio',
1063 'biography',
1064 'callnum',
1065 'cfn',
1066 'Chronological-subdivision',
1067 'cn-bib-source',
1068 'cn-bib-sort',
1069 'cn-class',
1070 'cn-item',
1071 'cn-prefix',
1072 'cn-suffix',
1073 'cpn',
1074 'Code-institution',
1075 'Conference-name',
1076 'Conference-name-heading',
1077 'Conference-name-see',
1078 'Conference-name-seealso',
1079 'Content-type',
1080 'Control-number',
1081 'copydate',
1082 'Corporate-name',
1083 'Corporate-name-heading',
1084 'Corporate-name-see',
1085 'Corporate-name-seealso',
1086 'Country-publication',
1087 'ctype',
1088 'curriculum',
1089 'date-entered-on-file',
1090 'Date-of-acquisition',
1091 'Date-of-publication',
1092 'Date-time-last-modified',
1093 'Dewey-classification',
1094 'Dissertation-information',
1095 'diss',
1096 'dtlm',
1097 'EAN',
1098 'extent',
1099 'fic',
1100 'fiction',
1101 'Form-subdivision',
1102 'format',
1103 'Geographic-subdivision',
1104 'he',
1105 'Heading',
1106 'Heading-use-main-or-added-entry',
1107 'Heading-use-series-added-entry ',
1108 'Heading-use-subject-added-entry',
1109 'Host-item',
1110 'id-other',
1111 'ident',
1112 'Identifier-standard',
1113 'Illustration-code',
1114 'Index-term-genre',
1115 'Index-term-uncontrolled',
1116 'Interest-age-level',
1117 'Interest-grade-level',
1118 'ISBN',
1119 'isbn',
1120 'ISSN',
1121 'issn',
1122 'itemtype',
1123 'kw',
1124 'Koha-Auth-Number',
1125 'l-format',
1126 'language',
1127 'language-original',
1128 'lc-card',
1129 'LC-card-number',
1130 'lcn',
1131 'lex',
1132 'lexile-number',
1133 'llength',
1134 'ln',
1135 'ln-audio',
1136 'ln-subtitle',
1137 'Local-classification',
1138 'Local-number',
1139 'Match-heading',
1140 'Match-heading-see-from',
1141 'Material-type',
1142 'mc-itemtype',
1143 'mc-rtype',
1144 'mus',
1145 'name',
1146 'Music-number',
1147 'Name-geographic',
1148 'Name-geographic-heading',
1149 'Name-geographic-see',
1150 'Name-geographic-seealso',
1151 'nb',
1152 'Note',
1153 'notes',
1154 'ns',
1155 'nt',
1156 'Other-control-number',
1157 'pb',
1158 'Personal-name',
1159 'Personal-name-heading',
1160 'Personal-name-see',
1161 'Personal-name-seealso',
1162 'pl',
1163 'Place-publication',
1164 'pn',
1165 'popularity',
1166 'pubdate',
1167 'Publisher',
1168 'Provider',
1169 'pv',
1170 'Reading-grade-level',
1171 'Record-control-number',
1172 'rcn',
1173 'Record-type',
1174 'rtype',
1175 'se',
1176 'See',
1177 'See-also',
1178 'sn',
1179 'Stock-number',
1180 'su',
1181 'Subject',
1182 'Subject-heading-thesaurus',
1183 'Subject-name-personal',
1184 'Subject-subdivision',
1185 'Summary',
1186 'Suppress',
1187 'su-geo',
1188 'su-na',
1189 'su-to',
1190 'su-ut',
1191 'ut',
1192 'Term-genre-form',
1193 'Term-genre-form-heading',
1194 'Term-genre-form-see',
1195 'Term-genre-form-seealso',
1196 'ti',
1197 'Title',
1198 'Title-cover',
1199 'Title-series',
1200 'Title-uniform',
1201 'Title-uniform-heading',
1202 'Title-uniform-see',
1203 'Title-uniform-seealso',
1204 'totalissues',
1205 'yr',
1206
1207 # items indexes
1208 'acqsource',
1209 'barcode',
1210 'bc',
1211 'branch',
1212 'ccode',
1213 'classification-source',
1214 'cn-sort',
1215 'coded-location-qualifier',
1216 'copynumber',
1217 'damaged',
1218 'datelastborrowed',
1219 'datelastseen',
1220 'holdingbranch',
1221 'homebranch',
1222 'issues',
1223 'item',
1224 'itemnumber',
1225 'itype',
1226 'Local-classification',
1227 'location',
1228 'lost',
1229 'materials-specified',
1230 'mc-ccode',
1231 'mc-itype',
1232 'mc-loc',
1233 'notforloan',
1234 'Number-local-acquisition',
1235 'onloan',
1236 'price',
1237 'renewals',
1238 'replacementprice',
1239 'replacementpricedate',
1240 'reserves',
1241 'restricted',
1242 'stack',
1243 'stocknumber',
1244 'inv',
1245 'uri',
1246 'withdrawn',
1247
1248 # subject related
1249 );
1250
1251 return \@indexes;
1252 }
1253
1254 =head2 buildQuery
1255
1256 ( $error, $query,
1257 $simple_query, $query_cgi,
1258 $query_desc, $limit,
1259 $limit_cgi, $limit_desc,
1260 $query_type ) = buildQuery ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1261
1262 Build queries and limits in CCL, CGI, Human,
1263 handle truncation, stemming, field weighting, fuzziness, etc.
1264
1265 See verbose embedded documentation.
1266
1267
1268 =cut
1269
1270 sub buildQuery {
1271 my ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang) = @_;
1272
1273 warn "---------\nEnter buildQuery\n---------" if $DEBUG;
1274
1275 my $query_desc;
1276
1277 # dereference
1278 my @operators = $operators ? @$operators : ();
1279 my @indexes = $indexes ? @$indexes : ();
1280 my @operands = $operands ? @$operands : ();
1281 my @limits = $limits ? @$limits : ();
1282 my @sort_by = $sort_by ? @$sort_by : ();
1283
1284 my $stemming = C4::Context->preference("QueryStemming") || 0;
1285 my $auto_truncation = C4::Context->preference("QueryAutoTruncate") || 0;
1286 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
1287 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
1288
1289 my $query = $operands[0];
1290 my $simple_query = $operands[0];
1291
1292 # initialize the variables we're passing back
1293 my $query_cgi;
1294 my $query_type;
1295
1296 my $limit;
1297 my $limit_cgi;
1298 my $limit_desc;
1299
1300 my $cclq = 0;
1301 my $cclindexes = getIndexes();
1302 if ( $query !~ /\s*(ccl=|pqf=|cql=)/ ) {
1303 while ( !$cclq && $query =~ /(?:^|\W)([\w-]+)(,[\w-]+)*[:=]/g ) {
1304 my $dx = lc($1);
1305 $cclq = grep { lc($_) eq $dx } @$cclindexes;
1306 }
1307 $query = "ccl=$query" if $cclq;
1308 }
1309
1310 # for handling ccl, cql, pqf queries in diagnostic mode, skip the rest of the steps
1311 # DIAGNOSTIC ONLY!!
1312 if ( $query =~ /^ccl=/ ) {
1313 my $q=$';
1314 # This is needed otherwise ccl= and &limit won't work together, and
1315 # this happens when selecting a subject on the opac-detail page
1316 @limits = grep {!/^$/} @limits;
1317 my $original_q = $q; # without available part
1318 unless ( grep { $_ eq 'available' } @limits ) {
1319 $q =~ s| and \( \(allrecords,AlwaysMatches=''\) and \(not-onloan-count,st-numeric >= 1\) and \(lost,st-numeric=0\) \)||;
1320 $original_q = $q;
1321 }
1322 if ( @limits ) {
1323 if ( grep { $_ eq 'available' } @limits ) {
1324 $q .= q| and ( (allrecords,AlwaysMatches='') and (not-onloan-count,st-numeric >= 1) and (lost,st-numeric=0) )|;
1325 @limits = grep {!/^available$/} @limits;
1326 }
1327 $q .= ' and '.join(' and ', @limits) if @limits;
1328 }
1329 return ( undef, $q, $q, "q=ccl=".uri_escape_utf8($q), $original_q, '', '', '', 'ccl' );
1330 }
1331 if ( $query =~ /^cql=/ ) {
1332 return ( undef, $', $', "q=cql=".uri_escape_utf8($'), $', '', '', '', 'cql' );
1333 }
1334 if ( $query =~ /^pqf=/ ) {
1335 $query_desc = $';
1336 $query_cgi = "q=pqf=".uri_escape_utf8($');
1337 return ( undef, $', $', $query_cgi, $query_desc, '', '', '', 'pqf' );
1338 }
1339
1340 # pass nested queries directly
1341 # FIXME: need better handling of some of these variables in this case
1342 # Nested queries aren't handled well and this implementation is flawed and causes users to be
1343 # unable to search for anything containing () commenting out, will be rewritten for 3.4.0
1344 # if ( $query =~ /(\(|\))/ ) {
1345 # return (
1346 # undef, $query, $simple_query, $query_cgi,
1347 # $query, $limit, $limit_cgi, $limit_desc,
1348 # 'ccl'
1349 # );
1350 # }
1351
1352 # Form-based queries are non-nested and fixed depth, so we can easily modify the incoming
1353 # query operands and indexes and add stemming, truncation, field weighting, etc.
1354 # Once we do so, we'll end up with a value in $query, just like if we had an
1355 # incoming $query from the user
1356 else {
1357 $query = ""
1358 ; # clear it out so we can populate properly with field-weighted, stemmed, etc. query
1359 my $previous_operand
1360 ; # a flag used to keep track if there was a previous query
1361 # if there was, we can apply the current operator
1362 # for every operand
1363 for ( my $i = 0 ; $i <= @operands ; $i++ ) {
1364
1365 # COMBINE OPERANDS, INDEXES AND OPERATORS
1366 if ( ($operands[$i] // '') ne '' ) {
1367 $operands[$i]=~s/^\s+//;
1368
1369 # A flag to determine whether or not to add the index to the query
1370 my $indexes_set;
1371
1372 # If the user is sophisticated enough to specify an index, turn off field weighting, and stemming handling
1373 if ( $operands[$i] =~ /\w(:|=)/ || $scan ) {
1374 $weight_fields = 0;
1375 $stemming = 0;
1376 } else {
1377 $operands[$i] =~ s/\?/{?}/g; # need to escape question marks
1378 }
1379 my $operand = $operands[$i];
1380 my $index = $indexes[$i] || 'kw';
1381
1382 # Add index-specific attributes
1383
1384 #Afaik, this 'yr' condition will only ever be met in the staff client advanced search
1385 #for "Publication date", since typing 'yr:YYYY' into the search box produces a CCL query,
1386 #which is processed higher up in this sub. Other than that, year searches are typically
1387 #handled as limits which are not processed her either.
1388
1389 # Search ranges: Date of Publication, st-numeric
1390 if ( $index =~ /(yr|st-numeric)/ ) {
1391 #weight_fields/relevance search causes errors with date ranges
1392 #In the case of YYYY-, it will only return records with a 'yr' of YYYY (not the range)
1393 #In the case of YYYY-YYYY, it will return no results
1394 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = 0;
1395 }
1396
1397 # Date of Acquisition
1398 elsif ( $index =~ /acqdate/ ) {
1399 #stemming and auto_truncation would have zero impact since it already is YYYY-MM-DD format
1400 #Weight_fields probably SHOULD be turned OFF, otherwise you'll get records floating to the
1401 #top of the results just because they have lots of item records matching that date.
1402 #Fuzzy actually only applies during _build_weighted_query, and is reset there anyway, so
1403 #irrelevant here
1404 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = 0;
1405 }
1406 # ISBN,ISSN,Standard Number, don't need special treatment
1407 elsif ( $index eq 'nb' || $index eq 'ns' || $index eq 'hi' ) {
1408 (
1409 $stemming, $auto_truncation,
1410 $weight_fields, $fuzzy_enabled
1411 ) = ( 0, 0, 0, 0 );
1412
1413 if ( $index eq 'nb' ) {
1414 if ( C4::Context->preference("SearchWithISBNVariations") ) {
1415 my @isbns = C4::Koha::GetVariationsOfISBN( $operand );
1416 $operands[$i] = $operand = '(nb=' . join(' OR nb=', @isbns) . ')';
1417 $indexes[$i] = $index = 'kw';
1418 }
1419 }
1420 }
1421
1422 # Set default structure attribute (word list)
1423 my $struct_attr = q{};
1424 unless ( $indexes_set || $index =~ /,(st-|phr|ext|wrdl)/ || $index =~ /^(nb|ns)$/ ) {
1425 $struct_attr = ",wrdl";
1426 }
1427
1428 # Some helpful index variants
1429 my $index_plus = $index . $struct_attr . ':';
1430 my $index_plus_comma = $index . $struct_attr . ',';
1431
1432 if ($auto_truncation){
1433 unless ( $index =~ /,(st-|phr|ext)/ ) {
1434 #FIXME only valid with LTR scripts
1435 $operand=join(" ",map{
1436 (index($_,"*")>0?"$_":"$_*")
1437 }split (/\s+/,$operand));
1438 warn $operand if $DEBUG;
1439 }
1440 }
1441
1442 # Detect Truncation
1443 my $truncated_operand;
1444 my( $nontruncated, $righttruncated, $lefttruncated,
1445 $rightlefttruncated, $regexpr
1446 ) = _detect_truncation( $operand, $index );
1447 warn
1448 "TRUNCATION: NON:>@$nontruncated< RIGHT:>@$righttruncated< LEFT:>@$lefttruncated< RIGHTLEFT:>@$rightlefttruncated< REGEX:>@$regexpr<"
1449 if $DEBUG;
1450
1451 # Apply Truncation
1452 if (
1453 scalar(@$righttruncated) + scalar(@$lefttruncated) +
1454 scalar(@$rightlefttruncated) > 0 )
1455 {
1456
1457 # Don't field weight or add the index to the query, we do it here
1458 $indexes_set = 1;
1459 undef $weight_fields;
1460 my $previous_truncation_operand;
1461 if (scalar @$nontruncated) {
1462 $truncated_operand .= "$index_plus @$nontruncated ";
1463 $previous_truncation_operand = 1;
1464 }
1465 if (scalar @$righttruncated) {
1466 $truncated_operand .= "and " if $previous_truncation_operand;
1467 $truncated_operand .= $index_plus_comma . "rtrn:@$righttruncated ";
1468 $previous_truncation_operand = 1;
1469 }
1470 if (scalar @$lefttruncated) {
1471 $truncated_operand .= "and " if $previous_truncation_operand;
1472 $truncated_operand .= $index_plus_comma . "ltrn:@$lefttruncated ";
1473 $previous_truncation_operand = 1;
1474 }
1475 if (scalar @$rightlefttruncated) {
1476 $truncated_operand .= "and " if $previous_truncation_operand;
1477 $truncated_operand .= $index_plus_comma . "rltrn:@$rightlefttruncated ";
1478 $previous_truncation_operand = 1;
1479 }
1480 }
1481 $operand = $truncated_operand if $truncated_operand;
1482 warn "TRUNCATED OPERAND: >$truncated_operand<" if $DEBUG;
1483
1484 # Handle Stemming
1485 my $stemmed_operand;
1486 $stemmed_operand = _build_stemmed_operand($operand, $lang)
1487 if $stemming;
1488
1489 warn "STEMMED OPERAND: >$stemmed_operand<" if $DEBUG;
1490
1491 # Handle Field Weighting
1492 my $weighted_operand;
1493 if ($weight_fields) {
1494 $weighted_operand = _build_weighted_query( $operand, $stemmed_operand, $index );
1495 $operand = $weighted_operand;
1496 $indexes_set = 1;
1497 }
1498
1499 warn "FIELD WEIGHTED OPERAND: >$weighted_operand<" if $DEBUG;
1500
1501 ($query,$query_cgi,$query_desc,$previous_operand) = _build_initial_query({
1502 query => $query,
1503 query_cgi => $query_cgi,
1504 query_desc => $query_desc,
1505 operator => ($operators[ $i - 1 ]) ? $operators[ $i - 1 ] : '',
1506 parsed_operand => $operand,
1507 original_operand => $operands[$i] // '',
1508 index => $index,
1509 index_plus => $index_plus,
1510 indexes_set => $indexes_set,
1511 previous_operand => $previous_operand,
1512 });
1513
1514 } #/if $operands
1515 } # /for
1516 }
1517 warn "QUERY BEFORE LIMITS: >$query<" if $DEBUG;
1518
1519 # add limits
1520 my %group_OR_limits;
1521 my $availability_limit;
1522 foreach my $this_limit (@limits) {
1523 next unless $this_limit;
1524 if ( $this_limit =~ /available/ ) {
1525 #
1526 ## 'available' is defined as (items.onloan is NULL) and (items.itemlost = 0)
1527 ## In English:
1528 ## all records not indexed in the onloan register (zebra) and all records with a value of lost equal to 0
1529 $availability_limit .=
1530 "( (allrecords,AlwaysMatches='') and (not-onloan-count,st-numeric >= 1) and (lost,st-numeric=0) )";
1531 $limit_cgi .= "&limit=available";
1532 $limit_desc .= "";
1533 }
1534
1535 # group_OR_limits, prefixed by mc-
1536 # OR every member of the group
1537 elsif ( $this_limit =~ /mc/ ) {
1538 my ($k,$v) = split(/:/, $this_limit,2);
1539 if ( $k !~ /mc-i(tem)?type/ ) {
1540 # in case the mc-ccode value has complicating chars like ()'s inside it we wrap in quotes
1541 $this_limit =~ tr/"//d;
1542 $this_limit = $k.":'".$v."'";
1543 }
1544
1545 $group_OR_limits{$k} .= " or " if $group_OR_limits{$k};
1546 $limit_desc .= " or " if $group_OR_limits{$k};
1547 $group_OR_limits{$k} .= "$this_limit";
1548 $limit_cgi .= "&limit=" . uri_escape_utf8($this_limit);
1549 $limit_desc .= " $this_limit";
1550 }
1551
1552 # Regular old limits
1553 else {
1554 $limit .= " and " if $limit || $query;
1555 $limit .= "$this_limit";
1556 $limit_cgi .= "&limit=" . uri_escape_utf8($this_limit);
1557 if ($this_limit =~ /^branch:(.+)/) {
1558 my $branchcode = $1;
1559 my $library = Koha::Libraries->find( $branchcode );
1560 if (defined $library) {
1561 $limit_desc .= " branch:" . $library->branchname;
1562 } else {
1563 $limit_desc .= " $this_limit";
1564 }
1565 } else {
1566 $limit_desc .= " $this_limit";
1567 }
1568 }
1569 }
1570 foreach my $k (keys (%group_OR_limits)) {
1571 $limit .= " and " if ( $query || $limit );
1572 $limit .= "($group_OR_limits{$k})";
1573 }
1574 if ($availability_limit) {
1575 $limit .= " and " if ( $query || $limit );
1576 $limit .= "($availability_limit)";
1577 }
1578
1579 # Normalize the query and limit strings
1580 # This is flawed , means we can't search anything with : in it
1581 # if user wants to do ccl or cql, start the query with that
1582 # $query =~ s/:/=/g;
1583 #NOTE: We use several several different regexps here as you can't have variable length lookback assertions
1584 $query =~ s/(?<=(ti|au|pb|su|an|kw|mc|nb|ns)):/=/g;
1585 $query =~ s/(?<=(wrdl)):/=/g;
1586 $query =~ s/(?<=(trn|phr)):/=/g;
1587 $query =~ s/(?<=(st-numeric)):/=/g;
1588 $query =~ s/(?<=(st-year)):/=/g;
1589 $query =~ s/(?<=(st-date-normalized)):/=/g;
1590
1591 # Removing warnings for later substitutions
1592 $query //= q{};
1593 $query_desc //= q{};
1594 $query_cgi //= q{};
1595 $limit //= q{};
1596 $limit_desc //= q{};
1597 $limit =~ s/:/=/g;
1598 for ( $query, $query_desc, $limit, $limit_desc ) {
1599 s/ +/ /g; # remove extra spaces
1600 s/^ //g; # remove any beginning spaces
1601 s/ $//g; # remove any ending spaces
1602 s/==/=/g; # remove double == from query
1603 }
1604 $query_cgi =~ s/^&//; # remove unnecessary & from beginning of the query cgi
1605
1606 for ($query_cgi,$simple_query) {
1607 s/"//g;
1608 }
1609 # append the limit to the query
1610 $query .= " " . $limit;
1611
1612 # Warnings if DEBUG
1613 if ($DEBUG) {
1614 warn "QUERY:" . $query;
1615 warn "QUERY CGI:" . $query_cgi;
1616 warn "QUERY DESC:" . $query_desc;
1617 warn "LIMIT:" . $limit;
1618 warn "LIMIT CGI:" . $limit_cgi;
1619 warn "LIMIT DESC:" . $limit_desc;
1620 warn "---------\nLeave buildQuery\n---------";
1621 }
1622
1623 return (
1624 undef, $query, $simple_query, $query_cgi,
1625 $query_desc, $limit, $limit_cgi, $limit_desc,
1626 $query_type
1627 );
1628 }
1629
1630 =head2 _build_initial_query
1631
1632 ($query, $query_cgi, $query_desc, $previous_operand) = _build_initial_query($initial_query_params);
1633
1634 Build a section of the initial query containing indexes, operators, and operands.
1635
1636 =cut
1637
1638 sub _build_initial_query {
1639 my ($params) = @_;
1640
1641 my $operator = "";
1642 if ($params->{previous_operand}){
1643 #If there is a previous operand, add a supplied operator or the default 'and'
1644 $operator = ($params->{operator}) ? " ".($params->{operator})." " : ' and ';
1645 }
1646
1647 #NOTE: indexes_set is typically set when doing truncation or field weighting
1648 my $operand = ($params->{indexes_set}) ? $params->{parsed_operand} : $params->{index_plus}.$params->{parsed_operand};
1649
1650 #e.g. "kw,wrdl:test"
1651 #e.g. " and kw,wrdl:test"
1652 $params->{query} .= $operator . $operand;
1653
1654 $params->{query_cgi} .= "&op=".uri_escape_utf8($operator) if $operator;
1655 $params->{query_cgi} .= "&idx=".uri_escape_utf8($params->{index}) if $params->{index};
1656 $params->{query_cgi} .= "&q=".uri_escape_utf8($params->{original_operand}) if $params->{original_operand};
1657
1658 #e.g. " and kw,wrdl: test"
1659 $params->{query_desc} .= $operator . ( $params->{index_plus} // q{} ) . " " . ( $params->{original_operand} // q{} );
1660
1661 $params->{previous_operand} = 1 unless $params->{previous_operand}; #If there is no previous operand, mark this as one
1662
1663 return ($params->{query}, $params->{query_cgi}, $params->{query_desc}, $params->{previous_operand});
1664 }
1665
1666 =head2 searchResults
1667
1668 my @search_results = searchResults($search_context, $searchdesc, $hits,
1669 $results_per_page, $offset, $scan,
1670 @marcresults);
1671
1672 Format results in a form suitable for passing to the template
1673
1674 =cut
1675
1676 # IMO this subroutine is pretty messy still -- it's responsible for
1677 # building the HTML output for the template
1678 sub searchResults {
1679 my ( $search_context, $searchdesc, $hits, $results_per_page, $offset, $scan, $marcresults, $xslt_variables ) = @_;
1680 my $dbh = C4::Context->dbh;
1681 my @newresults;
1682
1683 require C4::Items;
1684
1685 $search_context->{'interface'} = 'opac' if !$search_context->{'interface'} || $search_context->{'interface'} ne 'intranet';
1686 my ($is_opac, $hidelostitems);
1687 if ($search_context->{'interface'} eq 'opac') {
1688 $hidelostitems = C4::Context->preference('hidelostitems');
1689 $is_opac = 1;
1690 }
1691
1692 my $record_processor = Koha::RecordProcessor->new({
1693 filters => 'ViewPolicy'
1694 });
1695
1696 #Build branchnames hash
1697 my %branches = map { $_->branchcode => $_->branchname } Koha::Libraries->search({}, { order_by => 'branchname' });
1698
1699 # FIXME - We build an authorised values hash here, using the default framework
1700 # though it is possible to have different authvals for different fws.
1701
1702 my $shelflocations =
1703 { map { $_->{authorised_value} => $_->{lib} } Koha::AuthorisedValues->get_descriptions_by_koha_field( { frameworkcode => '', kohafield => 'items.location' } ) };
1704
1705 # get notforloan authorised value list (see $shelflocations FIXME)
1706 my $av = Koha::MarcSubfieldStructures->search({ frameworkcode => '', kohafield => 'items.notforloan', authorised_value => [ -and => {'!=' => undef }, {'!=' => ''}] });
1707 my $notforloan_authorised_value = $av->count ? $av->next->authorised_value : undef;
1708
1709 #Get itemtype hash
1710 my $itemtypes = Koha::ItemTypes->search_with_localization;
1711 my %itemtypes = map { $_->{itemtype} => $_ } @{ $itemtypes->unblessed };
1712
1713 #search item field code
1714 my ($itemtag, undef) = &GetMarcFromKohaField( "items.itemnumber" );
1715
1716 ## find column names of items related to MARC
1717 my %subfieldstosearch;
1718 my @columns = Koha::Database->new()->schema()->resultset('Item')->result_source->columns;
1719 for my $column ( @columns ) {
1720 my ( $tagfield, $tagsubfield ) =
1721 &GetMarcFromKohaField( "items." . $column );
1722 if ( defined $tagsubfield ) {
1723 $subfieldstosearch{$column} = $tagsubfield;
1724 }
1725 }
1726
1727 # handle which records to actually retrieve
1728 my $times;
1729 if ( $hits && $offset + $results_per_page <= $hits ) {
1730 $times = $offset + $results_per_page;
1731 }
1732 else {
1733 $times = $hits; # FIXME: if $hits is undefined, why do we want to equal it?
1734 }
1735
1736 my $marcflavour = C4::Context->preference("marcflavour");
1737 # We get the biblionumber position in MARC
1738 my ($bibliotag,$bibliosubf)=GetMarcFromKohaField( 'biblio.biblionumber' );
1739
1740 # set stuff for XSLT processing here once, not later again for every record we retrieved
1741 my $xslfile;
1742 my $xslsyspref;
1743 if( $is_opac ){
1744 $xslsyspref = "OPACXSLTResultsDisplay";
1745 $xslfile = C4::Context->preference( $xslsyspref );
1746 } else {
1747 $xslsyspref = "XSLTResultsDisplay";
1748 $xslfile = C4::Context->preference( $xslsyspref ) || "default";
1749 }
1750 my $lang = $xslfile ? C4::Languages::getlanguage() : undef;
1751 my $sysxml = $xslfile ? C4::XSLT::get_xslt_sysprefs() : undef;
1752
1753 my $userenv = C4::Context->userenv;
1754 my $logged_in_user
1755 = ( defined $userenv and $userenv->{number} )
1756 ? Koha::Patrons->find( $userenv->{number} )
1757 : undef;
1758 my $patron_category_hide_lost_items = ($logged_in_user) ? $logged_in_user->category->hidelostitems : 0;
1759
1760 # loop through all of the records we've retrieved
1761 for ( my $i = $offset ; $i <= $times - 1 ; $i++ ) {
1762
1763 my $marcrecord;
1764 if ($scan) {
1765 # For Scan searches we built USMARC data
1766 $marcrecord = MARC::Record->new_from_usmarc( $marcresults->[$i]);
1767 } else {
1768 # Normal search, render from Zebra's output
1769 $marcrecord = new_record_from_zebra(
1770 'biblioserver',
1771 $marcresults->[$i]
1772 );
1773
1774 if ( ! defined $marcrecord ) {
1775 warn "ERROR DECODING RECORD - $@: " . $marcresults->[$i];
1776 next;
1777 }
1778 }
1779
1780 my $fw = $scan
1781 ? undef
1782 : $bibliotag < 10
1783 ? GetFrameworkCode($marcrecord->field($bibliotag)->data)
1784 : GetFrameworkCode($marcrecord->subfield($bibliotag,$bibliosubf));
1785
1786 SetUTF8Flag($marcrecord);
1787 my $oldbiblio = TransformMarcToKoha( $marcrecord, $fw );
1788 $oldbiblio->{result_number} = $i + 1;
1789
1790 $oldbiblio->{normalized_upc} = GetNormalizedUPC( $marcrecord,$marcflavour);
1791 $oldbiblio->{normalized_ean} = GetNormalizedEAN( $marcrecord,$marcflavour);
1792 $oldbiblio->{normalized_oclc} = GetNormalizedOCLCNumber($marcrecord,$marcflavour);
1793 $oldbiblio->{normalized_isbn} = GetNormalizedISBN(undef,$marcrecord,$marcflavour);
1794 $oldbiblio->{content_identifier_exists} = 1 if ($oldbiblio->{normalized_isbn} or $oldbiblio->{normalized_oclc} or $oldbiblio->{normalized_ean} or $oldbiblio->{normalized_upc});
1795
1796 # edition information, if any
1797 $oldbiblio->{edition} = $oldbiblio->{editionstatement};
1798
1799 my $itemtype = $oldbiblio->{itemtype} ? $itemtypes{$oldbiblio->{itemtype}} : undef;
1800 # add imageurl to itemtype if there is one
1801 $oldbiblio->{imageurl} = $itemtype ? getitemtypeimagelocation( $search_context->{'interface'}, $itemtype->{imageurl} ) : q{};
1802 # Build summary if there is one (the summary is defined in the itemtypes table)
1803 $oldbiblio->{description} = $itemtype ? $itemtype->{translated_description} : q{};
1804
1805 # FIXME: this is only used in the deprecated non-XLST opac results
1806 if ( !$xslfile && $is_opac && $itemtype && $itemtype->{summary} ) {
1807 my $summary = $itemtypes{ $oldbiblio->{itemtype} }->{summary};
1808 my @fields = $marcrecord->fields();
1809
1810 my $newsummary;
1811 foreach my $line ( "$summary\n" =~ /(.*)\n/g ){
1812 my $tags = {};
1813 foreach my $tag ( $line =~ /\[(\d{3}[\w|\d])\]/ ) {
1814 $tag =~ /(.{3})(.)/;
1815 if($marcrecord->field($1)){
1816 my @abc = $marcrecord->field($1)->subfield($2);
1817 $tags->{$tag} = $#abc + 1 ;
1818 }
1819 }
1820
1821 # We catch how many times to repeat this line
1822 my $max = 0;
1823 foreach my $tag (keys(%$tags)){
1824 $max = $tags->{$tag} if($tags->{$tag} > $max);
1825 }
1826
1827 # we replace, and repeat each line
1828 for (my $i = 0 ; $i < $max ; $i++){
1829 my $newline = $line;
1830
1831 foreach my $tag ( $newline =~ /\[(\d{3}[\w|\d])\]/g ) {
1832 $tag =~ /(.{3})(.)/;
1833
1834 if($marcrecord->field($1)){
1835 my @repl = $marcrecord->field($1)->subfield($2);
1836 my $subfieldvalue = $repl[$i];
1837 $newline =~ s/\[$tag\]/$subfieldvalue/g;
1838 }
1839 }
1840 $newsummary .= "$newline\n";
1841 }
1842 }
1843
1844 $newsummary =~ s/\[(.*?)]//g;
1845 $newsummary =~ s/\n/<br\/>/g;
1846 $oldbiblio->{summary} = $newsummary;
1847 }
1848
1849 # Pull out the items fields
1850 my @fields = $marcrecord->field($itemtag);
1851 my $marcflavor = C4::Context->preference("marcflavour");
1852
1853 # adding linked items that belong to host records
1854 if ( C4::Context->preference('EasyAnalyticalRecords') ) {
1855 my $analyticsfield = '773';
1856 if ($marcflavor eq 'MARC21' || $marcflavor eq 'NORMARC') {
1857 $analyticsfield = '773';
1858 } elsif ($marcflavor eq 'UNIMARC') {
1859 $analyticsfield = '461';
1860 }
1861 foreach my $hostfield ( $marcrecord->field($analyticsfield)) {
1862 my $hostbiblionumber = $hostfield->subfield("0");
1863 my $linkeditemnumber = $hostfield->subfield("9");
1864 if( $hostbiblionumber ) {
1865 my $linkeditemmarc = C4::Items::GetMarcItem( $hostbiblionumber, $linkeditemnumber );
1866 if ($linkeditemmarc) {
1867 my $linkeditemfield = $linkeditemmarc->field($itemtag);
1868 if ($linkeditemfield) {
1869 push( @fields, $linkeditemfield );
1870 }
1871 }
1872 }
1873 }
1874 }
1875
1876 # Setting item statuses for display
1877 my @available_items_loop;
1878 my @onloan_items_loop;
1879 my @other_items_loop;
1880
1881 my $available_items;
1882 my $onloan_items;
1883 my $other_items;
1884
1885 my $ordered_count = 0;
1886 my $available_count = 0;
1887 my $onloan_count = 0;
1888 my $longoverdue_count = 0;
1889 my $other_count = 0;
1890 my $withdrawn_count = 0;
1891 my $itemlost_count = 0;
1892 my $hideatopac_count = 0;
1893 my $itembinding_count = 0;
1894 my $itemdamaged_count = 0;
1895 my $item_in_transit_count = 0;
1896 my $can_place_holds = 0;
1897 my $item_onhold_count = 0;
1898 my $notforloan_count = 0;
1899 my $items_count = scalar(@fields);
1900 my $maxitems_pref = C4::Context->preference('maxItemsinSearchResults');
1901 my $maxitems = $maxitems_pref ? $maxitems_pref - 1 : 1;
1902 my @hiddenitems; # hidden itemnumbers based on OpacHiddenItems syspref
1903
1904 # loop through every item
1905 foreach my $field (@fields) {
1906 my $item;
1907
1908 # populate the items hash
1909 foreach my $code ( keys %subfieldstosearch ) {
1910 $item->{$code} = $field->subfield( $subfieldstosearch{$code} );
1911 }
1912 $item->{description} = $itemtypes{ $item->{itype} }{translated_description} if $item->{itype};
1913
1914 # OPAC hidden items
1915 if ($is_opac) {
1916 # hidden because lost
1917 if ($hidelostitems && $item->{itemlost}) {
1918 $hideatopac_count++;
1919 next;
1920 }
1921 # hidden based on OpacHiddenItems syspref
1922 my @hi = C4::Items::GetHiddenItemnumbers({ items=> [ $item ], borcat => $search_context->{category} });
1923 if (scalar @hi) {
1924 push @hiddenitems, @hi;
1925 $hideatopac_count++;
1926 next;
1927 }
1928 }
1929
1930 my $hbranch = C4::Context->preference('StaffSearchResultsDisplayBranch');
1931 my $otherbranch = $hbranch eq 'homebranch' ? 'holdingbranch' : 'homebranch';
1932
1933 # set item's branch name, use HomeOrHoldingBranch syspref first, fall back to the other one
1934 if ($item->{$hbranch}) {
1935 $item->{'branchname'} = $branches{$item->{$hbranch}};
1936 }
1937 elsif ($item->{$otherbranch}) { # Last resort
1938 $item->{'branchname'} = $branches{$item->{$otherbranch}};
1939 }
1940
1941 my $prefix =
1942 ( $item->{$hbranch} ? $item->{$hbranch} . '--' : q{} )
1943 . ( $item->{location} ? $item->{location} : q{} )
1944 . ( $item->{itype} ? $item->{itype} : q{} )
1945 . ( $item->{itemcallnumber} ? $item->{itemcallnumber} : q{} );
1946 # For each grouping of items (onloan, available, unavailable), we build a key to store relevant info about that item
1947 if ( $item->{onloan}
1948 and $logged_in_user
1949 and !( $patron_category_hide_lost_items and $item->{itemlost} ) )
1950 {
1951 $onloan_count++;
1952 my $key = $prefix . $item->{onloan} . $item->{barcode};
1953 $onloan_items->{$key}->{due_date} = $item->{onloan};
1954 $onloan_items->{$key}->{count}++ if $item->{$hbranch};
1955 $onloan_items->{$key}->{branchname} = $item->{branchname};
1956 $onloan_items->{$key}->{location} = $shelflocations->{ $item->{location} };
1957 $onloan_items->{$key}->{itemcallnumber} = $item->{itemcallnumber};
1958 $onloan_items->{$key}->{description} = $item->{description};
1959 $onloan_items->{$key}->{imageurl} =
1960 getitemtypeimagelocation( $search_context->{'interface'}, $itemtypes{ $item->{itype} }->{imageurl} );
1961
1962 # if something's checked out and lost, mark it as 'long overdue'
1963 if ( $item->{itemlost} ) {
1964 $onloan_items->{$key}->{longoverdue}++;
1965 $longoverdue_count++;
1966 }
1967 else { # can place holds as long as item isn't lost
1968 $can_place_holds = 1;
1969 }
1970 }
1971
1972 # items not on loan, but still unavailable ( lost, withdrawn, damaged )
1973 else {
1974
1975 my $itemtype = C4::Context->preference("item-level_itypes")? $item->{itype}: $oldbiblio->{itemtype};
1976 $item->{notforloan} = 1 if !$item->{notforloan} &&
1977 $itemtype && $itemtypes{ $itemtype }->{notforloan};
1978
1979 # item is on order
1980 if ( $item->{notforloan} < 0 ) {
1981 $ordered_count++;
1982 } elsif ( $item->{notforloan} > 0 ) {
1983 $notforloan_count++;
1984 }
1985
1986 # is item in transit?
1987 my $transfertwhen = '';
1988 my ($transfertfrom, $transfertto);
1989
1990 # is item on the reserve shelf?
1991 my $reservestatus = '';
1992
1993 unless ($item->{withdrawn}
1994 || $item->{itemlost}
1995 || $item->{damaged}
1996 || $item->{notforloan}
1997 || ( C4::Context->preference('MaxSearchResultsItemsPerRecordStatusCheck')
1998 && $items_count > C4::Context->preference('MaxSearchResultsItemsPerRecordStatusCheck') ) ) {
1999
2000 # A couple heuristics to limit how many times
2001 # we query the database for item transfer information, sacrificing
2002 # accuracy in some cases for speed;
2003 #
2004 # 1. don't query if item has one of the other statuses
2005 # 2. don't check transit status if the bib has
2006 # more than 20 items
2007 #
2008 # FIXME: to avoid having the query the database like this, and to make
2009 # the in transit status count as unavailable for search limiting,
2010 # should map transit status to record indexed in Zebra.
2011 #
2012 ($transfertwhen, $transfertfrom, $transfertto) = C4::Circulation::GetTransfers($item->{itemnumber});
2013 $reservestatus = C4::Reserves::GetReserveStatus( $item->{itemnumber} );
2014 }
2015
2016 # item is withdrawn, lost, damaged, not for loan, reserved or in transit
2017 if ( $item->{withdrawn}
2018 || $item->{itemlost}
2019 || $item->{damaged}
2020 || $item->{notforloan}
2021 || $reservestatus eq 'Waiting'
2022 || ($transfertwhen && $transfertwhen ne ''))
2023 {
2024 $withdrawn_count++ if $item->{withdrawn};
2025 $itemlost_count++ if $item->{itemlost};
2026 $itemdamaged_count++ if $item->{damaged};
2027 $item_in_transit_count++ if $transfertwhen && $transfertwhen ne '';
2028 $item_onhold_count++ if $reservestatus eq 'Waiting';
2029 $item->{status} = ($item->{withdrawn}//q{}) . "-" . ($item->{itemlost}//q{}) . "-" . ($item->{damaged}//q{}) . "-" . ($item->{notforloan}//q{});
2030
2031 # can place a hold on a item if
2032 # not lost nor withdrawn
2033 # not damaged unless AllowHoldsOnDamagedItems is true
2034 # item is either for loan or on order (notforloan < 0)
2035 $can_place_holds = 1
2036 if (
2037 !$item->{itemlost}
2038 && !$item->{withdrawn}
2039 && ( !$item->{damaged} || C4::Context->preference('AllowHoldsOnDamagedItems') )
2040 && ( !$item->{notforloan} || $item->{notforloan} < 0 )
2041 );
2042
2043 $other_count++;
2044
2045 my $key = $prefix . $item->{status};
2046 foreach (qw(withdrawn itemlost damaged branchname itemcallnumber)) {
2047 $other_items->{$key}->{$_} = $item->{$_};
2048 }
2049 $other_items->{$key}->{intransit} = ( $transfertwhen ne '' ) ? 1 : 0;
2050 $other_items->{$key}->{onhold} = ($reservestatus) ? 1 : 0;
2051 $other_items->{$key}->{notforloan} = GetAuthorisedValueDesc('','',$item->{notforloan},'','',$notforloan_authorised_value) if $notforloan_authorised_value and $item->{notforloan};
2052 $other_items->{$key}->{count}++ if $item->{$hbranch};
2053 $other_items->{$key}->{location} = $shelflocations->{ $item->{location} } if $item->{location};
2054 $other_items->{$key}->{description} = $item->{description};
2055 $other_items->{$key}->{imageurl} = getitemtypeimagelocation( $search_context->{'interface'}, $itemtypes{ $item->{itype}//q{} }->{imageurl} );
2056 }
2057 # item is available
2058 else {
2059 $can_place_holds = 1;
2060 $available_count++;
2061 $available_items->{$prefix}->{count}++ if $item->{$hbranch};
2062 foreach (qw(branchname itemcallnumber description)) {
2063 $available_items->{$prefix}->{$_} = $item->{$_};
2064 }
2065 $available_items->{$prefix}->{location} = $shelflocations->{ $item->{location} } if $item->{location};
2066 $available_items->{$prefix}->{imageurl} = getitemtypeimagelocation( $search_context->{'interface'}, $itemtypes{ $item->{itype}//q{} }->{imageurl} );
2067 }
2068 }
2069 } # notforloan, item level and biblioitem level
2070
2071 # if all items are hidden, do not show the record
2072 if ($items_count > 0 && $hideatopac_count == $items_count) {
2073 next;
2074 }
2075
2076 my ( $availableitemscount, $onloanitemscount, $otheritemscount );
2077 for my $key ( sort keys %$onloan_items ) {
2078 (++$onloanitemscount > $maxitems) and last;
2079 push @onloan_items_loop, $onloan_items->{$key};
2080 }
2081 for my $key ( sort keys %$other_items ) {
2082 (++$otheritemscount > $maxitems) and last;
2083 push @other_items_loop, $other_items->{$key};
2084 }
2085 for my $key ( sort keys %$available_items ) {
2086 (++$availableitemscount > $maxitems) and last;
2087 push @available_items_loop, $available_items->{$key}
2088 }
2089
2090 # XSLT processing of some stuff
2091 # we fetched the sysprefs already before the loop through all retrieved record!
2092 if (!$scan && $xslfile) {
2093 $record_processor->options({
2094 frameworkcode => $fw,
2095 interface => $search_context->{'interface'}
2096 });
2097
2098 $record_processor->process($marcrecord);
2099 $oldbiblio->{XSLTResultsRecord} = XSLTParse4Display($oldbiblio->{biblionumber}, $marcrecord, $xslsyspref, 1, \@hiddenitems, $sysxml, $xslfile, $lang, $xslt_variables);
2100 }
2101
2102 # if biblio level itypes are used and itemtype is notforloan, it can't be reserved either
2103 if (!C4::Context->preference("item-level_itypes")) {
2104 if ($itemtype && $itemtype->{notforloan}) {
2105 $can_place_holds = 0;
2106 }
2107 }
2108 $oldbiblio->{norequests} = 1 unless $can_place_holds;
2109 $oldbiblio->{items_count} = $items_count;
2110 $oldbiblio->{available_items_loop} = \@available_items_loop;
2111 $oldbiblio->{onloan_items_loop} = \@onloan_items_loop;
2112 $oldbiblio->{other_items_loop} = \@other_items_loop;
2113 $oldbiblio->{availablecount} = $available_count;
2114 $oldbiblio->{availableplural} = 1 if $available_count > 1;
2115 $oldbiblio->{onloancount} = $onloan_count;
2116 $oldbiblio->{onloanplural} = 1 if $onloan_count > 1;
2117 $oldbiblio->{othercount} = $other_count;
2118 $oldbiblio->{otherplural} = 1 if $other_count > 1;
2119 $oldbiblio->{withdrawncount} = $withdrawn_count;
2120 $oldbiblio->{itemlostcount} = $itemlost_count;
2121 $oldbiblio->{damagedcount} = $itemdamaged_count;
2122 $oldbiblio->{intransitcount} = $item_in_transit_count;
2123 $oldbiblio->{onholdcount} = $item_onhold_count;
2124 $oldbiblio->{orderedcount} = $ordered_count;
2125 $oldbiblio->{notforloancount} = $notforloan_count;
2126
2127 if (C4::Context->preference("AlternateHoldingsField") && $items_count == 0) {
2128 my $fieldspec = C4::Context->preference("AlternateHoldingsField");
2129 my $subfields = substr $fieldspec, 3;
2130 my $holdingsep = C4::Context->preference("AlternateHoldingsSeparator") || ' ';
2131 my @alternateholdingsinfo = ();
2132 my @holdingsfields = $marcrecord->field(substr $fieldspec, 0, 3);
2133 my $alternateholdingscount = 0;
2134
2135 for my $field (@holdingsfields) {
2136 my %holding = ( holding => '' );
2137 my $havesubfield = 0;
2138 for my $subfield ($field->subfields()) {
2139 if ((index $subfields, $$subfield[0]) >= 0) {
2140 $holding{'holding'} .= $holdingsep if (length $holding{'holding'} > 0);
2141 $holding{'holding'} .= $$subfield[1];
2142 $havesubfield++;
2143 }
2144 }
2145 if ($havesubfield) {
2146 push(@alternateholdingsinfo, \%holding);
2147 $alternateholdingscount++;
2148 }
2149 }
2150
2151 $oldbiblio->{'ALTERNATEHOLDINGS'} = \@alternateholdingsinfo;
2152 $oldbiblio->{'alternateholdings_count'} = $alternateholdingscount;
2153 }
2154
2155 $oldbiblio->{biblio_object} = Koha::Biblios->find( $oldbiblio->{biblionumber} );
2156
2157 push( @newresults, $oldbiblio );
2158 }
2159
2160 return @newresults;
2161 }
2162
2163 =head2 enabled_staff_search_views
2164
2165 %hash = enabled_staff_search_views()
2166
2167 This function returns a hash that contains three flags obtained from the system
2168 preferences, used to determine whether a particular staff search results view
2169 is enabled.
2170
2171 =over 2
2172
2173 =item C<Output arg:>
2174
2175 * $hash{can_view_MARC} is true only if the MARC view is enabled
2176 * $hash{can_view_ISBD} is true only if the ISBD view is enabled
2177 * $hash{can_view_labeledMARC} is true only if the Labeled MARC view is enabled
2178
2179 =item C<usage in the script:>
2180
2181 =back
2182
2183 $template->param ( C4::Search::enabled_staff_search_views );
2184
2185 =cut
2186
2187 sub enabled_staff_search_views
2188 {
2189 return (
2190 can_view_MARC => C4::Context->preference('viewMARC'), # 1 if the staff search allows the MARC view
2191 can_view_ISBD => C4::Context->preference('viewISBD'), # 1 if the staff search allows the ISBD view
2192 can_view_labeledMARC => C4::Context->preference('viewLabeledMARC'), # 1 if the staff search allows the Labeled MARC view
2193 );
2194 }
2195
2196 =head2 z3950_search_args
2197
2198 $arrayref = z3950_search_args($matchpoints)
2199
2200 This function returns an array reference that contains the search parameters to be
2201 passed to the Z39.50 search script (z3950_search.pl). The array elements
2202 are hash refs whose keys are name and value, and whose values are the
2203 name of a search parameter, the value of that search parameter and the URL encoded
2204 value of that parameter.
2205
2206 The search parameter names are lccn, isbn, issn, title, author, dewey and subject.
2207
2208 The search parameter values are obtained from the bibliographic record whose
2209 data is in a hash reference in $matchpoints, as returned by Biblio::GetBiblioData().
2210
2211 If $matchpoints is a scalar, it is assumed to be an unnamed query descriptor, e.g.
2212 a general purpose search argument. In this case, the returned array contains only
2213 entry: the key is 'title' and the value is derived from $matchpoints.
2214
2215 If a search parameter value is undefined or empty, it is not included in the returned
2216 array.
2217
2218 The returned array reference may be passed directly to the template parameters.
2219
2220 =over 2
2221
2222 =item C<Output arg:>
2223
2224 * $array containing hash refs as described above
2225
2226 =item C<usage in the script:>
2227
2228 =back
2229
2230 $data = Biblio::GetBiblioData($bibno);
2231 $template->param ( MYLOOP => C4::Search::z3950_search_args($data) )
2232
2233 *OR*
2234
2235 $template->param ( MYLOOP => C4::Search::z3950_search_args($searchscalar) )
2236
2237 =cut
2238
2239 sub z3950_search_args {
2240 my $bibrec = shift;
2241
2242 my $isbn_string = ref( $bibrec ) ? $bibrec->{title} : $bibrec;
2243 my $isbn = Business::ISBN->new( $isbn_string );
2244
2245 if (defined $isbn && $isbn->is_valid)
2246 {
2247 if ( ref($bibrec) ) {
2248 $bibrec->{isbn} = $isbn_string;
2249 $bibrec->{title} = undef;
2250 } else {
2251 $bibrec = { isbn => $isbn_string };
2252 }
2253 }
2254 else {
2255 $bibrec = { title => $bibrec } if !ref $bibrec;
2256 }
2257 my $array = [];
2258 for my $field (qw/ lccn isbn issn title author dewey subject /)
2259 {
2260 push @$array, { name => $field, value => $bibrec->{$field} }
2261 if defined $bibrec->{$field};
2262 }
2263 return $array;
2264 }
2265
2266 =head2 GetDistinctValues($field);
2267
2268 C<$field> is a reference to the fields array
2269
2270 =cut
2271
2272 sub GetDistinctValues {
2273 my ($fieldname,$string)=@_;
2274 # returns a reference to a hash of references to branches...
2275 if ($fieldname=~/\./){
2276 my ($table,$column)=split /\./, $fieldname;
2277 my $dbh = C4::Context->dbh;
2278 warn "select DISTINCT($column) as value, count(*) as cnt from $table group by lib order by $column " if $DEBUG;
2279 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 ");
2280 $sth->execute;
2281 my $elements=$sth->fetchall_arrayref({});
2282 return $elements;
2283 }
2284 else {
2285 $string||= qq("");
2286 my @servers=qw<biblioserver authorityserver>;
2287 my (@zconns,@results);
2288 for ( my $i = 0 ; $i < @servers ; $i++ ) {
2289 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
2290 $results[$i] =
2291 $zconns[$i]->scan(
2292 ZOOM::Query::CCL2RPN->new( qq"$fieldname $string", $zconns[$i])
2293 );
2294 }
2295 # The big moment: asynchronously retrieve results from all servers
2296 my @elements;
2297 _ZOOM_event_loop(
2298 \@zconns,
2299 \@results,
2300 sub {
2301 my ( $i, $size ) = @_;
2302 for ( my $j = 0 ; $j < $size ; $j++ ) {
2303 my %hashscan;
2304 @hashscan{qw(value cnt)} =
2305 $results[ $i - 1 ]->display_term($j);
2306 push @elements, \%hashscan;
2307 }
2308 }
2309 );
2310 return \@elements;
2311 }
2312 }
2313
2314 =head2 _ZOOM_event_loop
2315
2316 _ZOOM_event_loop(\@zconns, \@results, sub {
2317 my ( $i, $size ) = @_;
2318 ....
2319 } );
2320
2321 Processes a ZOOM event loop and passes control to a closure for
2322 processing the results, and destroying the resultsets.
2323
2324 =cut
2325
2326 sub _ZOOM_event_loop {
2327 my ($zconns, $results, $callback) = @_;
2328 while ( ( my $i = ZOOM::event( $zconns ) ) != 0 ) {
2329 my $ev = $zconns->[ $i - 1 ]->last_event();
2330 if ( $ev == ZOOM::Event::ZEND ) {
2331 next unless $results->[ $i - 1 ];
2332 my $size = $results->[ $i - 1 ]->size();
2333 if ( $size > 0 ) {
2334 $callback->($i, $size);
2335 }
2336 }
2337 }
2338
2339 foreach my $result (@$results) {
2340 $result->destroy();
2341 }
2342 }
2343
2344 =head2 new_record_from_zebra
2345
2346 Given raw data from a searchengine result set, return a MARC::Record object
2347
2348 This helper function is needed to take into account all the involved
2349 system preferences and configuration variables to properly create the
2350 MARC::Record object.
2351
2352 If we are using GRS-1, then the raw data we get from Zebra should be USMARC
2353 data. If we are using DOM, then it has to be MARCXML.
2354
2355 If we are using elasticsearch, it'll already be a MARC::Record and this
2356 function needs a new name.
2357
2358 =cut
2359
2360 sub new_record_from_zebra {
2361
2362 my $server = shift;
2363 my $raw_data = shift;
2364 # Set the default indexing modes
2365 my $search_engine = C4::Context->preference("SearchEngine");
2366 if ($search_engine eq 'Elasticsearch') {
2367 return ref $raw_data eq 'MARC::Record' ? $raw_data : MARC::Record->new_from_xml( $raw_data, 'UTF-8' );
2368 }
2369 my $index_mode = ( $server eq 'biblioserver' )
2370 ? C4::Context->config('zebra_bib_index_mode') // 'dom'
2371 : C4::Context->config('zebra_auth_index_mode') // 'dom';
2372
2373 my $marc_record = eval {
2374 if ( $index_mode eq 'dom' ) {
2375 MARC::Record->new_from_xml( $raw_data, 'UTF-8' );
2376 } else {
2377 MARC::Record->new_from_usmarc( $raw_data );
2378 }
2379 };
2380
2381 if ($@) {
2382 return;
2383 } else {
2384 return $marc_record;
2385 }
2386
2387 }
2388
2389 END { } # module clean-up code here (global destructor)
2390
2391 1;
2392 __END__
2393
2394 =head1 AUTHOR
2395
2396 Koha Development Team <http://koha-community.org/>
2397
2398 =cut
Koha ILS