?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
Bug 17301 - Add callnumber to label-edit-batch.pl
[koha.git] / C4 / Search.pm
bloba1296f9171d985ddd6ca5be742281a6af1be3392
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 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::Members qw(GetHideLostItemsPreference);
30 use C4::XSLT;
31 use C4::Reserves; # GetReserveStatus
32 use C4::Debug;
33 use C4::Charset;
34 use YAML;
35 use URI::Escape;
36 use Business::ISBN;
37 use MARC::Record;
38 use MARC::Field;
39 use vars qw(@ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
40
41 BEGIN {
42 $DEBUG = ($ENV{DEBUG}) ? 1 : 0;
43 }
44
45 =head1 NAME
46
47 C4::Search - Functions for searching the Koha catalog.
48
49 =head1 SYNOPSIS
50
51 See opac/opac-search.pl or catalogue/search.pl for example of usage
52
53 =head1 DESCRIPTION
54
55 This module provides searching functions for Koha's bibliographic databases
56
57 =head1 FUNCTIONS
58
59 =cut
60
61 @ISA = qw(Exporter);
62 @EXPORT = qw(
63 &FindDuplicate
64 &SimpleSearch
65 &searchResults
66 &getRecords
67 &buildQuery
68 &GetDistinctValues
69 &enabled_staff_search_views
70 );
71
72 # make all your functions, whether exported or not;
73
74 =head2 FindDuplicate
75
76 ($biblionumber,$biblionumber,$title) = FindDuplicate($record);
77
78 This function attempts to find duplicate records using a hard-coded, fairly simplistic algorithm
79
80 =cut
81
82 sub FindDuplicate {
83 my ($record) = @_;
84 my $dbh = C4::Context->dbh;
85 my $result = TransformMarcToKoha( $record, '' );
86 my $sth;
87 my $query;
88 my $search;
89 my $type;
90 my ( $biblionumber, $title );
91
92 # search duplicate on ISBN, easy and fast..
93 # ... normalize first
94 if ( $result->{isbn} ) {
95 $result->{isbn} =~ s/\(.*$//;
96 $result->{isbn} =~ s/\s+$//;
97 $query = "isbn:$result->{isbn}";
98 }
99 else {
100 my $QParser;
101 $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser'));
102 my $titleindex;
103 my $authorindex;
104 my $op;
105
106 if ($QParser) {
107 $titleindex = 'title|exact';
108 $authorindex = 'author|exact';
109 $op = '&&';
110 $QParser->custom_data->{'QueryAutoTruncate'} = C4::Context->preference('QueryAutoTruncate');
111 } else {
112 $titleindex = 'ti,ext';
113 $authorindex = 'au,ext';
114 $op = 'and';
115 }
116
117 $result->{title} =~ s /\\//g;
118 $result->{title} =~ s /\"//g;
119 $result->{title} =~ s /\(//g;
120 $result->{title} =~ s /\)//g;
121
122 # FIXME: instead of removing operators, could just do
123 # quotes around the value
124 $result->{title} =~ s/(and|or|not)//g;
125 $query = "$titleindex:\"$result->{title}\"";
126 if ( $result->{author} ) {
127 $result->{author} =~ s /\\//g;
128 $result->{author} =~ s /\"//g;
129 $result->{author} =~ s /\(//g;
130 $result->{author} =~ s /\)//g;
131
132 # remove valid operators
133 $result->{author} =~ s/(and|or|not)//g;
134 $query .= " $op $authorindex:\"$result->{author}\"";
135 }
136 }
137
138 my ( $error, $searchresults, undef ) = SimpleSearch($query); # FIXME :: hardcoded !
139 my @results;
140 if (!defined $error) {
141 foreach my $possible_duplicate_record (@{$searchresults}) {
142 my $marcrecord = new_record_from_zebra(
143 'biblioserver',
144 $possible_duplicate_record
145 );
146
147 my $result = TransformMarcToKoha( $marcrecord, '' );
148
149 # FIXME :: why 2 $biblionumber ?
150 if ($result) {
151 push @results, $result->{'biblionumber'};
152 push @results, $result->{'title'};
153 }
154 }
155 }
156 return @results;
157 }
158
159 =head2 SimpleSearch
160
161 ( $error, $results, $total_hits ) = SimpleSearch( $query, $offset, $max_results, [@servers] );
162
163 This function provides a simple search API on the bibliographic catalog
164
165 =over 2
166
167 =item C<input arg:>
168
169 * $query can be a simple keyword or a complete CCL query
170 * @servers is optional. Defaults to biblioserver as found in koha-conf.xml
171 * $offset - If present, represents the number of records at the beginning to omit. Defaults to 0
172 * $max_results - if present, determines the maximum number of records to fetch. undef is All. defaults to undef.
173
174
175 =item C<Return:>
176
177 Returns an array consisting of three elements
178 * $error is undefined unless an error is detected
179 * $results is a reference to an array of records.
180 * $total_hits is the number of hits that would have been returned with no limit
181
182 If an error is returned the two other return elements are undefined. If error itself is undefined
183 the other two elements are always defined
184
185 =item C<usage in the script:>
186
187 =back
188
189 my ( $error, $marcresults, $total_hits ) = SimpleSearch($query);
190
191 if (defined $error) {
192 $template->param(query_error => $error);
193 warn "error: ".$error;
194 output_html_with_http_headers $input, $cookie, $template->output;
195 exit;
196 }
197
198 my $hits = @{$marcresults};
199 my @results;
200
201 for my $r ( @{$marcresults} ) {
202 my $marcrecord = MARC::File::USMARC::decode($r);
203 my $biblio = TransformMarcToKoha($marcrecord,q{});
204
205 #build the iarray of hashs for the template.
206 push @results, {
207 title => $biblio->{'title'},
208 subtitle => $biblio->{'subtitle'},
209 biblionumber => $biblio->{'biblionumber'},
210 author => $biblio->{'author'},
211 publishercode => $biblio->{'publishercode'},
212 publicationyear => $biblio->{'publicationyear'},
213 };
214
215 }
216
217 $template->param(result=>\@results);
218
219 =cut
220
221 sub SimpleSearch {
222 my ( $query, $offset, $max_results, $servers ) = @_;
223
224 return ( 'No query entered', undef, undef ) unless $query;
225 # FIXME hardcoded value. See catalog/search.pl & opac-search.pl too.
226 my @servers = defined ( $servers ) ? @$servers : ( 'biblioserver' );
227 my @zoom_queries;
228 my @tmpresults;
229 my @zconns;
230 my $results = [];
231 my $total_hits = 0;
232
233 my $QParser;
234 $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser') && ! ($query =~ m/\w,\w|\w=\w/));
235 if ($QParser) {
236 $QParser->custom_data->{'QueryAutoTruncate'} = C4::Context->preference('QueryAutoTruncate');
237 }
238
239 # Initialize & Search Zebra
240 for ( my $i = 0 ; $i < @servers ; $i++ ) {
241 eval {
242 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
243 if ($QParser) {
244 $query =~ s/=/:/g;
245 $QParser->parse( $query );
246 $query = $QParser->target_syntax($servers[$i]);
247 $zoom_queries[$i] = new ZOOM::Query::PQF( $query, $zconns[$i]);
248 } else {
249 $query =~ s/:/=/g;
250 $zoom_queries[$i] = new ZOOM::Query::CCL2RPN( $query, $zconns[$i]);
251 }
252 $tmpresults[$i] = $zconns[$i]->search( $zoom_queries[$i] );
253
254 # error handling
255 my $error =
256 $zconns[$i]->errmsg() . " ("
257 . $zconns[$i]->errcode() . ") "
258 . $zconns[$i]->addinfo() . " "
259 . $zconns[$i]->diagset();
260
261 return ( $error, undef, undef ) if $zconns[$i]->errcode();
262 };
263 if ($@) {
264
265 # caught a ZOOM::Exception
266 my $error =
267 $@->message() . " ("
268 . $@->code() . ") "
269 . $@->addinfo() . " "
270 . $@->diagset();
271 warn $error." for query: $query";
272 return ( $error, undef, undef );
273 }
274 }
275
276 _ZOOM_event_loop(
277 \@zconns,
278 \@tmpresults,
279 sub {
280 my ($i, $size) = @_;
281 my $first_record = defined($offset) ? $offset + 1 : 1;
282 my $hits = $tmpresults[ $i - 1 ]->size();
283 $total_hits += $hits;
284 my $last_record = $hits;
285 if ( defined $max_results && $offset + $max_results < $hits ) {
286 $last_record = $offset + $max_results;
287 }
288
289 for my $j ( $first_record .. $last_record ) {
290 my $record = eval {
291 $tmpresults[ $i - 1 ]->record( $j - 1 )->raw()
292 ; # 0 indexed
293 };
294 push @{$results}, $record if defined $record;
295 }
296 }
297 );
298
299 foreach my $zoom_query (@zoom_queries) {
300 $zoom_query->destroy();
301 }
302
303 return ( undef, $results, $total_hits );
304 }
305
306 =head2 getRecords
307
308 ( undef, $results_hashref, \@facets_loop ) = getRecords (
309
310 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
311 $results_per_page, $offset, $expanded_facet, $branches,$itemtypes,
312 $query_type, $scan
313 );
314
315 The all singing, all dancing, multi-server, asynchronous, scanning,
316 searching, record nabbing, facet-building
317
318 See verbse embedded documentation.
319
320 =cut
321
322 sub getRecords {
323 my (
324 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
325 $results_per_page, $offset, $expanded_facet, $branches,
326 $itemtypes, $query_type, $scan, $opac
327 ) = @_;
328
329 my @servers = @$servers_ref;
330 my @sort_by = @$sort_by_ref;
331
332 # Initialize variables for the ZOOM connection and results object
333 my $zconn;
334 my @zconns;
335 my @results;
336 my $results_hashref = ();
337
338 # TODO simplify this structure ( { branchcode => $branchname } is enought) and remove this parameter
339 $branches ||= { map { $_->branchcode => { branchname => $_->branchname } } Koha::Libraries->search };
340
341 # Initialize variables for the faceted results objects
342 my $facets_counter = {};
343 my $facets_info = {};
344 my $facets = getFacets();
345
346 my @facets_loop; # stores the ref to array of hashes for template facets loop
347
348 ### LOOP THROUGH THE SERVERS
349 for ( my $i = 0 ; $i < @servers ; $i++ ) {
350 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
351
352 # perform the search, create the results objects
353 # if this is a local search, use the $koha-query, if it's a federated one, use the federated-query
354 my $query_to_use = ($servers[$i] =~ /biblioserver/) ? $koha_query : $simple_query;
355
356 #$query_to_use = $simple_query if $scan;
357 warn $simple_query if ( $scan and $DEBUG );
358
359 # Check if we've got a query_type defined, if so, use it
360 eval {
361 if ($query_type) {
362 if ($query_type =~ /^ccl/) {
363 $query_to_use =~ s/\:/\=/g; # change : to = last minute (FIXME)
364 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
365 } elsif ($query_type =~ /^cql/) {
366 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CQL($query_to_use, $zconns[$i]));
367 } elsif ($query_type =~ /^pqf/) {
368 $results[$i] = $zconns[$i]->search(new ZOOM::Query::PQF($query_to_use, $zconns[$i]));
369 } else {
370 warn "Unknown query_type '$query_type'. Results undetermined.";
371 }
372 } elsif ($scan) {
373 $results[$i] = $zconns[$i]->scan( new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
374 } else {
375 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
376 }
377 };
378 if ($@) {
379 warn "WARNING: query problem with $query_to_use " . $@;
380 }
381
382 # Concatenate the sort_by limits and pass them to the results object
383 # Note: sort will override rank
384 my $sort_by;
385 foreach my $sort (@sort_by) {
386 if ( $sort eq "author_az" || $sort eq "author_asc" ) {
387 $sort_by .= "1=1003 <i ";
388 }
389 elsif ( $sort eq "author_za" || $sort eq "author_dsc" ) {
390 $sort_by .= "1=1003 >i ";
391 }
392 elsif ( $sort eq "popularity_asc" ) {
393 $sort_by .= "1=9003 <i ";
394 }
395 elsif ( $sort eq "popularity_dsc" ) {
396 $sort_by .= "1=9003 >i ";
397 }
398 elsif ( $sort eq "call_number_asc" ) {
399 $sort_by .= "1=8007 <i ";
400 }
401 elsif ( $sort eq "call_number_dsc" ) {
402 $sort_by .= "1=8007 >i ";
403 }
404 elsif ( $sort eq "pubdate_asc" ) {
405 $sort_by .= "1=31 <i ";
406 }
407 elsif ( $sort eq "pubdate_dsc" ) {
408 $sort_by .= "1=31 >i ";
409 }
410 elsif ( $sort eq "acqdate_asc" ) {
411 $sort_by .= "1=32 <i ";
412 }
413 elsif ( $sort eq "acqdate_dsc" ) {
414 $sort_by .= "1=32 >i ";
415 }
416 elsif ( $sort eq "title_az" || $sort eq "title_asc" ) {
417 $sort_by .= "1=4 <i ";
418 }
419 elsif ( $sort eq "title_za" || $sort eq "title_dsc" ) {
420 $sort_by .= "1=4 >i ";
421 }
422 else {
423 warn "Ignoring unrecognized sort '$sort' requested" if $sort_by;
424 }
425 }
426 if ( $sort_by && !$scan && $results[$i] ) {
427 if ( $results[$i]->sort( "yaz", $sort_by ) < 0 ) {
428 warn "WARNING sort $sort_by failed";
429 }
430 }
431 } # finished looping through servers
432
433 # The big moment: asynchronously retrieve results from all servers
434 _ZOOM_event_loop(
435 \@zconns,
436 \@results,
437 sub {
438 my ( $i, $size ) = @_;
439 my $results_hash;
440
441 # loop through the results
442 $results_hash->{'hits'} = $size;
443 my $times;
444 if ( $offset + $results_per_page <= $size ) {
445 $times = $offset + $results_per_page;
446 }
447 else {
448 $times = $size;
449 }
450
451 for ( my $j = $offset ; $j < $times ; $j++ ) {
452 my $records_hash;
453 my $record;
454
455 ## Check if it's an index scan
456 if ($scan) {
457 my ( $term, $occ ) = $results[ $i - 1 ]->display_term($j);
458
459 # here we create a minimal MARC record and hand it off to the
460 # template just like a normal result ... perhaps not ideal, but
461 # it works for now
462 my $tmprecord = MARC::Record->new();
463 $tmprecord->encoding('UTF-8');
464 my $tmptitle;
465 my $tmpauthor;
466
467 # the minimal record in author/title (depending on MARC flavour)
468 if ( C4::Context->preference("marcflavour") eq
469 "UNIMARC" )
470 {
471 $tmptitle = MARC::Field->new(
472 '200', ' ', ' ',
473 a => $term,
474 f => $occ
475 );
476 $tmprecord->append_fields($tmptitle);
477 }
478 else {
479 $tmptitle =
480 MARC::Field->new( '245', ' ', ' ', a => $term, );
481 $tmpauthor =
482 MARC::Field->new( '100', ' ', ' ', a => $occ, );
483 $tmprecord->append_fields($tmptitle);
484 $tmprecord->append_fields($tmpauthor);
485 }
486 $results_hash->{'RECORDS'}[$j] =
487 $tmprecord->as_usmarc();
488 }
489
490 # not an index scan
491 else {
492 $record = $results[ $i - 1 ]->record($j)->raw();
493 # warn "RECORD $j:".$record;
494 $results_hash->{'RECORDS'}[$j] = $record;
495 }
496
497 }
498 $results_hashref->{ $servers[ $i - 1 ] } = $results_hash;
499
500 # Fill the facets while we're looping, but only for the
501 # biblioserver and not for a scan
502 if ( !$scan && $servers[ $i - 1 ] =~ /biblioserver/ ) {
503 $facets_counter = GetFacets( $results[ $i - 1 ] );
504 $facets_info = _get_facets_info( $facets );
505 }
506
507 # BUILD FACETS
508 if ( $servers[ $i - 1 ] =~ /biblioserver/ ) {
509 for my $link_value (
510 sort { $facets_counter->{$b} <=> $facets_counter->{$a} }
511 keys %$facets_counter
512 )
513 {
514 my $expandable;
515 my $number_of_facets;
516 my @this_facets_array;
517 for my $one_facet (
518 sort {
519 $facets_counter->{$link_value}
520 ->{$b} <=> $facets_counter->{$link_value}
521 ->{$a}
522 } keys %{ $facets_counter->{$link_value} }
523 )
524 {
525 $number_of_facets++;
526 if ( ( $number_of_facets <= 5 )
527 || ( $expanded_facet eq $link_value )
528 || ( $facets_info->{$link_value}->{'expanded'} )
529 )
530 {
531
532 # Sanitize the link value : parenthesis, question and exclamation mark will cause errors with CCL
533 my $facet_link_value = $one_facet;
534 $facet_link_value =~ s/[()!?¡¿؟]/ /g;
535
536 # fix the length that will display in the label,
537 my $facet_label_value = $one_facet;
538 my $facet_max_length = C4::Context->preference(
539 'FacetLabelTruncationLength')
540 || 20;
541 $facet_label_value =
542 substr( $one_facet, 0, $facet_max_length )
543 . "..."
544 if length($facet_label_value) >
545 $facet_max_length;
546
547 # if it's a branch, label by the name, not the code,
548 if ( $link_value =~ /branch/ ) {
549 if ( defined $branches
550 && ref($branches) eq "HASH"
551 && defined $branches->{$one_facet}
552 && ref( $branches->{$one_facet} ) eq
553 "HASH" )
554 {
555 $facet_label_value =
556 $branches->{$one_facet}
557 ->{'branchname'};
558 }
559 else {
560 $facet_label_value = "*";
561 }
562 }
563
564 # if it's a itemtype, label by the name, not the code,
565 if ( $link_value =~ /itype/ ) {
566 if ( defined $itemtypes
567 && ref($itemtypes) eq "HASH"
568 && defined $itemtypes->{$one_facet}
569 && ref( $itemtypes->{$one_facet} ) eq
570 "HASH" )
571 {
572 $facet_label_value =
573 $itemtypes->{$one_facet}
574 ->{translated_description};
575 }
576 }
577
578 # also, if it's a location code, use the name instead of the code
579 if ( $link_value =~ /location/ ) {
580 $facet_label_value =
581 GetKohaAuthorisedValueLib( 'LOC',
582 $one_facet, $opac );
583 }
584
585 # but we're down with the whole label being in the link's title.
586 push @this_facets_array,
587 {
588 facet_count =>
589 $facets_counter->{$link_value}
590 ->{$one_facet},
591 facet_label_value => $facet_label_value,
592 facet_title_value => $one_facet,
593 facet_link_value => $facet_link_value,
594 type_link_value => $link_value,
595 }
596 if ($facet_label_value);
597 }
598 }
599
600 # handle expanded option
601 unless ( $facets_info->{$link_value}->{'expanded'} ) {
602 $expandable = 1
603 if ( ( $number_of_facets > 5 )
604 && ( $expanded_facet ne $link_value ) );
605 }
606 push @facets_loop,
607 {
608 type_link_value => $link_value,
609 type_id => $link_value . "_id",
610 "type_label_"
611 . $facets_info->{$link_value}->{'label_value'} =>
612 1,
613 facets => \@this_facets_array,
614 expandable => $expandable,
615 expand => $link_value,
616 }
617 unless (
618 (
619 $facets_info->{$link_value}->{'label_value'} =~
620 /Libraries/
621 )
622 and ( Koha::Libraries->search->count == 1 )
623 );
624 }
625 }
626 }
627 );
628
629 # This sorts the facets into alphabetical order
630 if (@facets_loop) {
631 foreach my $f (@facets_loop) {
632 $f->{facets} = [ sort { uc($a->{facet_label_value}) cmp uc($b->{facet_label_value}) } @{ $f->{facets} } ];
633 }
634 @facets_loop = sort {$a->{expand} cmp $b->{expand}} @facets_loop;
635 }
636
637 return ( undef, $results_hashref, \@facets_loop );
638 }
639
640 sub GetFacets {
641
642 my $rs = shift;
643 my $facets;
644
645 my $indexing_mode = C4::Context->config('zebra_bib_index_mode') // 'dom';
646 my $use_zebra_facets = C4::Context->config('use_zebra_facets') // 0;
647
648 if ( $indexing_mode eq 'dom' &&
649 $use_zebra_facets ) {
650 $facets = _get_facets_from_zebra( $rs );
651 } else {
652 $facets = _get_facets_from_records( $rs );
653 }
654
655 return $facets;
656 }
657
658 sub _get_facets_from_records {
659
660 my $rs = shift;
661
662 my $facets_maxrecs = C4::Context->preference('maxRecordsForFacets') // 20;
663 my $facets_config = getFacets();
664 my $facets = {};
665 my $size = $rs->size();
666 my $jmax = $size > $facets_maxrecs
667 ? $facets_maxrecs
668 : $size;
669
670 for ( my $j = 0 ; $j < $jmax ; $j++ ) {
671
672 my $marc_record = new_record_from_zebra (
673 'biblioserver',
674 $rs->record( $j )->raw()
675 );
676
677 if ( ! defined $marc_record ) {
678 warn "ERROR DECODING RECORD - $@: " .
679 $rs->record( $j )->raw();
680 next;
681 }
682
683 _get_facets_data_from_record( $marc_record, $facets_config, $facets );
684 }
685
686 return $facets;
687 }
688
689 =head2 _get_facets_data_from_record
690
691 C4::Search::_get_facets_data_from_record( $marc_record, $facets, $facets_counter );
692
693 Internal function that extracts facets information from a MARC::Record object
694 and populates $facets_counter for using in getRecords.
695
696 $facets is expected to be filled with C4::Koha::getFacets output (i.e. the configured
697 facets for Zebra).
698
699 =cut
700
701 sub _get_facets_data_from_record {
702
703 my ( $marc_record, $facets, $facets_counter ) = @_;
704
705 for my $facet (@$facets) {
706
707 my @used_datas = ();
708
709 foreach my $tag ( @{ $facet->{ tags } } ) {
710
711 # tag number is the first three digits
712 my $tag_num = substr( $tag, 0, 3 );
713 # subfields are the remainder
714 my $subfield_letters = substr( $tag, 3 );
715
716 my @fields = $marc_record->field( $tag_num );
717 foreach my $field (@fields) {
718 # If $field->indicator(1) eq 'z', it means it is a 'see from'
719 # field introduced because of IncludeSeeFromInSearches, so skip it
720 next if $field->indicator(1) eq 'z';
721
722 my $data = $field->as_string( $subfield_letters, $facet->{ sep } );
723
724 unless ( grep { /^\Q$data\E$/ } @used_datas ) {
725 push @used_datas, $data;
726 $facets_counter->{ $facet->{ idx } }->{ $data }++;
727 }
728 }
729 }
730 }
731 }
732
733 =head2 _get_facets_from_zebra
734
735 my $facets = _get_facets_from_zebra( $result_set )
736
737 Retrieves facets for a specified result set. It loops through the facets defined
738 in C4::Koha::getFacets and returns a hash with the following structure:
739
740 { facet_idx => {
741 facet_value => count
742 },
743 ...
744 }
745
746 =cut
747
748 sub _get_facets_from_zebra {
749
750 my $rs = shift;
751
752 # save current elementSetName
753 my $elementSetName = $rs->option( 'elementSetName' );
754
755 my $facets_loop = getFacets();
756 my $facets_data = {};
757 # loop through defined facets and fill the facets hashref
758 foreach my $facet ( @$facets_loop ) {
759
760 my $idx = $facet->{ idx };
761 my $sep = $facet->{ sep };
762 my $facet_values = _get_facet_from_result_set( $idx, $rs, $sep );
763 if ( $facet_values ) {
764 # we've actually got a result
765 $facets_data->{ $idx } = $facet_values;
766 }
767 }
768 # set elementSetName to its previous value to avoid side effects
769 $rs->option( elementSetName => $elementSetName );
770
771 return $facets_data;
772 }
773
774 =head2 _get_facet_from_result_set
775
776 my $facet_values =
777 C4::Search::_get_facet_from_result_set( $facet_idx, $result_set, $sep )
778
779 Internal function that extracts facet information for a specific index ($facet_idx) and
780 returns a hash containing facet values and count:
781
782 {
783 $facet_value => $count ,
784 ...
785 }
786
787 Warning: this function has the side effect of changing the elementSetName for the result
788 set. It is a helper function for the main loop, which takes care of backing it up for
789 restoring.
790
791 =cut
792
793 sub _get_facet_from_result_set {
794
795 my $facet_idx = shift;
796 my $rs = shift;
797 my $sep = shift;
798
799 my $internal_sep = '<*>';
800 my $facetMaxCount = C4::Context->preference('FacetMaxCount') // 20;
801
802 return if ( ! defined $facet_idx || ! defined $rs );
803 # zebra's facet element, untokenized index
804 my $facet_element = 'zebra::facet::' . $facet_idx . ':0:' . $facetMaxCount;
805 # configure zebra results for retrieving the desired facet
806 $rs->option( elementSetName => $facet_element );
807 # get the facet record from result set
808 my $facet = $rs->record( 0 )->raw;
809 # if the facet has no restuls...
810 return if !defined $facet;
811 # TODO: benchmark DOM vs. SAX performance
812 my $facet_dom = XML::LibXML->load_xml(
813 string => ($facet)
814 );
815 my @terms = $facet_dom->getElementsByTagName('term');
816 return if ! @terms;
817
818 my $facets = {};
819 foreach my $term ( @terms ) {
820 my $facet_value = $term->textContent;
821 $facet_value =~ s/\Q$internal_sep\E/$sep/ if defined $sep;
822 $facets->{ $facet_value } = $term->getAttribute( 'occur' );
823 }
824
825 return $facets;
826 }
827
828 =head2 _get_facets_info
829
830 my $facets_info = C4::Search::_get_facets_info( $facets )
831
832 Internal function that extracts facets information and properly builds
833 the data structure needed to render facet labels.
834
835 =cut
836
837 sub _get_facets_info {
838
839 my $facets = shift;
840
841 my $facets_info = {};
842
843 for my $facet ( @$facets ) {
844 $facets_info->{ $facet->{ idx } }->{ label_value } = $facet->{ label };
845 $facets_info->{ $facet->{ idx } }->{ expanded } = $facet->{ expanded };
846 }
847
848 return $facets_info;
849 }
850
851 sub pazGetRecords {
852 my (
853 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
854 $results_per_page, $offset, $expanded_facet, $branches,
855 $query_type, $scan
856 ) = @_;
857 #NOTE: Parameter $branches is not used here !
858
859 my $paz = C4::Search::PazPar2->new(C4::Context->config('pazpar2url'));
860 $paz->init();
861 $paz->search($simple_query);
862 sleep 1; # FIXME: WHY?
863
864 # do results
865 my $results_hashref = {};
866 my $stats = XMLin($paz->stat);
867 my $results = XMLin($paz->show($offset, $results_per_page, 'work-title:1'), forcearray => 1);
868
869 # for a grouped search result, the number of hits
870 # is the number of groups returned; 'bib_hits' will have
871 # the total number of bibs.
872 $results_hashref->{'biblioserver'}->{'hits'} = $results->{'merged'}->[0];
873 $results_hashref->{'biblioserver'}->{'bib_hits'} = $stats->{'hits'};
874
875 HIT: foreach my $hit (@{ $results->{'hit'} }) {
876 my $recid = $hit->{recid}->[0];
877
878 my $work_title = $hit->{'md-work-title'}->[0];
879 my $work_author;
880 if (exists $hit->{'md-work-author'}) {
881 $work_author = $hit->{'md-work-author'}->[0];
882 }
883 my $group_label = (defined $work_author) ? "$work_title / $work_author" : $work_title;
884
885 my $result_group = {};
886 $result_group->{'group_label'} = $group_label;
887 $result_group->{'group_merge_key'} = $recid;
888
889 my $count = 1;
890 if (exists $hit->{count}) {
891 $count = $hit->{count}->[0];
892 }
893 $result_group->{'group_count'} = $count;
894
895 for (my $i = 0; $i < $count; $i++) {
896 # FIXME -- may need to worry about diacritics here
897 my $rec = $paz->record($recid, $i);
898 push @{ $result_group->{'RECORDS'} }, $rec;
899 }
900
901 push @{ $results_hashref->{'biblioserver'}->{'GROUPS'} }, $result_group;
902 }
903
904 # pass through facets
905 my $termlist_xml = $paz->termlist('author,subject');
906 my $terms = XMLin($termlist_xml, forcearray => 1);
907 my @facets_loop = ();
908 #die Dumper($results);
909 # foreach my $list (sort keys %{ $terms->{'list'} }) {
910 # my @facets = ();
911 # foreach my $facet (sort @{ $terms->{'list'}->{$list}->{'term'} } ) {
912 # push @facets, {
913 # facet_label_value => $facet->{'name'}->[0],
914 # };
915 # }
916 # push @facets_loop, ( {
917 # type_label => $list,
918 # facets => \@facets,
919 # } );
920 # }
921
922 return ( undef, $results_hashref, \@facets_loop );
923 }
924
925 # TRUNCATION
926 sub _detect_truncation {
927 my ( $operand, $index ) = @_;
928 my ( @nontruncated, @righttruncated, @lefttruncated, @rightlefttruncated,
929 @regexpr );
930 $operand =~ s/^ //g;
931 my @wordlist = split( /\s/, $operand );
932 foreach my $word (@wordlist) {
933 if ( $word =~ s/^\*([^\*]+)\*$/$1/ ) {
934 push @rightlefttruncated, $word;
935 }
936 elsif ( $word =~ s/^\*([^\*]+)$/$1/ ) {
937 push @lefttruncated, $word;
938 }
939 elsif ( $word =~ s/^([^\*]+)\*$/$1/ ) {
940 push @righttruncated, $word;
941 }
942 elsif ( index( $word, "*" ) < 0 ) {
943 push @nontruncated, $word;
944 }
945 else {
946 push @regexpr, $word;
947 }
948 }
949 return (
950 \@nontruncated, \@righttruncated, \@lefttruncated,
951 \@rightlefttruncated, \@regexpr
952 );
953 }
954
955 # STEMMING
956 sub _build_stemmed_operand {
957 my ($operand,$lang) = @_;
958 require Lingua::Stem::Snowball ;
959 my $stemmed_operand=q{};
960
961 # If operand contains a digit, it is almost certainly an identifier, and should
962 # not be stemmed. This is particularly relevant for ISBNs and ISSNs, which
963 # can contain the letter "X" - for example, _build_stemmend_operand would reduce
964 # "014100018X" to "x ", which for a MARC21 database would bring up irrelevant
965 # results (e.g., "23 x 29 cm." from the 300$c). Bug 2098.
966 return $operand if $operand =~ /\d/;
967
968 # FIXME: the locale should be set based on the user's language and/or search choice
969 #warn "$lang";
970 # Make sure we only use the first two letters from the language code
971 $lang = lc(substr($lang, 0, 2));
972 # The language codes for the two variants of Norwegian will now be "nb" and "nn",
973 # none of which Lingua::Stem::Snowball can use, so we need to "translate" them
974 if ($lang eq 'nb' || $lang eq 'nn') {
975 $lang = 'no';
976 }
977 my $stemmer = Lingua::Stem::Snowball->new( lang => $lang,
978 encoding => "UTF-8" );
979
980 my @words = split( / /, $operand );
981 my @stems = $stemmer->stem(\@words);
982 for my $stem (@stems) {
983 $stemmed_operand .= "$stem";
984 $stemmed_operand .= "?"
985 unless ( $stem =~ /(and$|or$|not$)/ ) || ( length($stem) < 3 );
986 $stemmed_operand .= " ";
987 }
988 warn "STEMMED OPERAND: $stemmed_operand" if $DEBUG;
989 return $stemmed_operand;
990 }
991
992 # FIELD WEIGHTING
993 sub _build_weighted_query {
994
995 # FIELD WEIGHTING - This is largely experimental stuff. What I'm committing works
996 # pretty well but could work much better if we had a smarter query parser
997 my ( $operand, $stemmed_operand, $index ) = @_;
998 my $stemming = C4::Context->preference("QueryStemming") || 0;
999 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
1000 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
1001 $operand =~ s/"/ /g; # Bug 7518: searches with quotation marks don't work
1002
1003 my $weighted_query .= "(rk=("; # Specifies that we're applying rank
1004
1005 # Keyword, or, no index specified
1006 if ( ( $index eq 'kw' ) || ( !$index ) ) {
1007 $weighted_query .=
1008 "Title-cover,ext,r1=\"$operand\""; # exact title-cover
1009 $weighted_query .= " or ti,ext,r2=\"$operand\""; # exact title
1010 $weighted_query .= " or Title-cover,phr,r3=\"$operand\""; # phrase title
1011 $weighted_query .= " or ti,wrdl,r4=\"$operand\""; # words in title
1012 #$weighted_query .= " or any,ext,r4=$operand"; # exact any
1013 #$weighted_query .=" or kw,wrdl,r5=\"$operand\""; # word list any
1014 $weighted_query .= " or wrdl,fuzzy,r8=\"$operand\""
1015 if $fuzzy_enabled; # add fuzzy, word list
1016 $weighted_query .= " or wrdl,right-Truncation,r9=\"$stemmed_operand\""
1017 if ( $stemming and $stemmed_operand )
1018 ; # add stemming, right truncation
1019 $weighted_query .= " or wrdl,r9=\"$operand\"";
1020
1021 # embedded sorting: 0 a-z; 1 z-a
1022 # $weighted_query .= ") or (sort1,aut=1";
1023 }
1024
1025 # Barcode searches should skip this process
1026 elsif ( $index eq 'bc' ) {
1027 $weighted_query .= "bc=\"$operand\"";
1028 }
1029
1030 # Authority-number searches should skip this process
1031 elsif ( $index eq 'an' ) {
1032 $weighted_query .= "an=\"$operand\"";
1033 }
1034
1035 # If the index is numeric, don't autoquote it.
1036 elsif ( $index =~ /,st-numeric$/ ) {
1037 $weighted_query .= " $index=$operand";
1038 }
1039
1040 # If the index already has more than one qualifier, wrap the operand
1041 # in quotes and pass it back (assumption is that the user knows what they
1042 # are doing and won't appreciate us mucking up their query
1043 elsif ( $index =~ ',' ) {
1044 $weighted_query .= " $index=\"$operand\"";
1045 }
1046
1047 #TODO: build better cases based on specific search indexes
1048 else {
1049 $weighted_query .= " $index,ext,r1=\"$operand\""; # exact index
1050 #$weighted_query .= " or (title-sort-az=0 or $index,startswithnt,st-word,r3=$operand #)";
1051 $weighted_query .= " or $index,phr,r3=\"$operand\""; # phrase index
1052 $weighted_query .= " or $index,wrdl,r6=\"$operand\""; # word list index
1053 $weighted_query .= " or $index,wrdl,fuzzy,r8=\"$operand\""
1054 if $fuzzy_enabled; # add fuzzy, word list
1055 $weighted_query .= " or $index,wrdl,rt,r9=\"$stemmed_operand\""
1056 if ( $stemming and $stemmed_operand ); # add stemming, right truncation
1057 }
1058
1059 $weighted_query .= "))"; # close rank specification
1060 return $weighted_query;
1061 }
1062
1063 =head2 getIndexes
1064
1065 Return an array with available indexes.
1066
1067 =cut
1068
1069 sub getIndexes{
1070 my @indexes = (
1071 # biblio indexes
1072 'ab',
1073 'Abstract',
1074 'acqdate',
1075 'allrecords',
1076 'an',
1077 'Any',
1078 'at',
1079 'au',
1080 'aub',
1081 'aud',
1082 'audience',
1083 'auo',
1084 'aut',
1085 'Author',
1086 'Author-in-order ',
1087 'Author-personal-bibliography',
1088 'Authority-Number',
1089 'authtype',
1090 'bc',
1091 'Bib-level',
1092 'biblionumber',
1093 'bio',
1094 'biography',
1095 'callnum',
1096 'cfn',
1097 'Chronological-subdivision',
1098 'cn-bib-source',
1099 'cn-bib-sort',
1100 'cn-class',
1101 'cn-item',
1102 'cn-prefix',
1103 'cn-suffix',
1104 'cpn',
1105 'Code-institution',
1106 'Conference-name',
1107 'Conference-name-heading',
1108 'Conference-name-see',
1109 'Conference-name-seealso',
1110 'Content-type',
1111 'Control-number',
1112 'copydate',
1113 'Corporate-name',
1114 'Corporate-name-heading',
1115 'Corporate-name-see',
1116 'Corporate-name-seealso',
1117 'Country-publication',
1118 'ctype',
1119 'curriculum',
1120 'date-entered-on-file',
1121 'Date-of-acquisition',
1122 'Date-of-publication',
1123 'Date-time-last-modified',
1124 'Dewey-classification',
1125 'Dissertation-information',
1126 'diss',
1127 'dtlm',
1128 'EAN',
1129 'extent',
1130 'fic',
1131 'fiction',
1132 'Form-subdivision',
1133 'format',
1134 'Geographic-subdivision',
1135 'he',
1136 'Heading',
1137 'Heading-use-main-or-added-entry',
1138 'Heading-use-series-added-entry ',
1139 'Heading-use-subject-added-entry',
1140 'Host-item',
1141 'id-other',
1142 'ident',
1143 'Identifier-standard',
1144 'Illustration-code',
1145 'Index-term-genre',
1146 'Index-term-uncontrolled',
1147 'Interest-age-level',
1148 'Interest-grade-level',
1149 'ISBN',
1150 'isbn',
1151 'ISSN',
1152 'issn',
1153 'itemtype',
1154 'kw',
1155 'Koha-Auth-Number',
1156 'l-format',
1157 'language',
1158 'language-original',
1159 'lc-card',
1160 'LC-card-number',
1161 'lcn',
1162 'lex',
1163 'lexile-number',
1164 'llength',
1165 'ln',
1166 'ln-audio',
1167 'ln-subtitle',
1168 'Local-classification',
1169 'Local-number',
1170 'Match-heading',
1171 'Match-heading-see-from',
1172 'Material-type',
1173 'mc-itemtype',
1174 'mc-rtype',
1175 'mus',
1176 'name',
1177 'Music-number',
1178 'Name-geographic',
1179 'Name-geographic-heading',
1180 'Name-geographic-see',
1181 'Name-geographic-seealso',
1182 'nb',
1183 'Note',
1184 'notes',
1185 'ns',
1186 'nt',
1187 'Other-control-number',
1188 'pb',
1189 'Personal-name',
1190 'Personal-name-heading',
1191 'Personal-name-see',
1192 'Personal-name-seealso',
1193 'pl',
1194 'Place-publication',
1195 'pn',
1196 'popularity',
1197 'pubdate',
1198 'Publisher',
1199 'Provider',
1200 'pv',
1201 'Reading-grade-level',
1202 'Record-control-number',
1203 'rcn',
1204 'Record-type',
1205 'rtype',
1206 'se',
1207 'See',
1208 'See-also',
1209 'sn',
1210 'Stock-number',
1211 'su',
1212 'Subject',
1213 'Subject-heading-thesaurus',
1214 'Subject-name-personal',
1215 'Subject-subdivision',
1216 'Summary',
1217 'Suppress',
1218 'su-geo',
1219 'su-na',
1220 'su-to',
1221 'su-ut',
1222 'ut',
1223 'Term-genre-form',
1224 'Term-genre-form-heading',
1225 'Term-genre-form-see',
1226 'Term-genre-form-seealso',
1227 'ti',
1228 'Title',
1229 'Title-cover',
1230 'Title-series',
1231 'Title-uniform',
1232 'Title-uniform-heading',
1233 'Title-uniform-see',
1234 'Title-uniform-seealso',
1235 'totalissues',
1236 'yr',
1237
1238 # items indexes
1239 'acqsource',
1240 'barcode',
1241 'bc',
1242 'branch',
1243 'ccode',
1244 'classification-source',
1245 'cn-sort',
1246 'coded-location-qualifier',
1247 'copynumber',
1248 'damaged',
1249 'datelastborrowed',
1250 'datelastseen',
1251 'holdingbranch',
1252 'homebranch',
1253 'issues',
1254 'item',
1255 'itemnumber',
1256 'itype',
1257 'Local-classification',
1258 'location',
1259 'lost',
1260 'materials-specified',
1261 'mc-ccode',
1262 'mc-itype',
1263 'mc-loc',
1264 'notforloan',
1265 'Number-local-acquisition',
1266 'onloan',
1267 'price',
1268 'renewals',
1269 'replacementprice',
1270 'replacementpricedate',
1271 'reserves',
1272 'restricted',
1273 'stack',
1274 'stocknumber',
1275 'inv',
1276 'uri',
1277 'withdrawn',
1278
1279 # subject related
1280 );
1281
1282 return \@indexes;
1283 }
1284
1285 =head2 _handle_exploding_index
1286
1287 my $query = _handle_exploding_index($index, $term)
1288
1289 Callback routine to generate the search for "exploding" indexes (i.e.
1290 those indexes which are turned into multiple or-connected searches based
1291 on authority data).
1292
1293 =cut
1294
1295 sub _handle_exploding_index {
1296 my ($QParser, $filter, $params, $negate, $server) = @_;
1297 my $index = $filter;
1298 my $term = join(' ', @$params);
1299
1300 return unless ($index =~ m/(su-br|su-na|su-rl)/ && $term);
1301
1302 my $marcflavour = C4::Context->preference('marcflavour');
1303
1304 my $codesubfield = $marcflavour eq 'UNIMARC' ? '5' : 'w';
1305 my $wantedcodes = '';
1306 my @subqueries = ( "\@attr 1=Subject \@attr 4=1 \"$term\"");
1307 my ($error, $results, $total_hits) = SimpleSearch( "he:$term", undef, undef, [ "authorityserver" ] );
1308 foreach my $auth (@$results) {
1309 my $record = MARC::Record->new_from_usmarc($auth);
1310 my @references = $record->field('5..');
1311 if (@references) {
1312 if ($index eq 'su-br') {
1313 $wantedcodes = 'g';
1314 } elsif ($index eq 'su-na') {
1315 $wantedcodes = 'h';
1316 } elsif ($index eq 'su-rl') {
1317 $wantedcodes = '';
1318 }
1319 foreach my $reference (@references) {
1320 my $codes = $reference->subfield($codesubfield);
1321 push @subqueries, '@attr 1=Subject @attr 4=1 "' . $reference->as_string('abcdefghijlmnopqrstuvxyz') . '"' if (($codes && $codes eq $wantedcodes) || !$wantedcodes);
1322 }
1323 }
1324 }
1325 my $query = ' @or ' x (scalar(@subqueries) - 1) . join(' ', @subqueries);
1326 return $query;
1327 }
1328
1329 =head2 parseQuery
1330
1331 ( $operators, $operands, $indexes, $limits,
1332 $sort_by, $scan, $lang ) =
1333 buildQuery ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1334
1335 Shim function to ease the transition from buildQuery to a new QueryParser.
1336 This function is called at the beginning of buildQuery, and modifies
1337 buildQuery's input. If it can handle the input, it returns a query that
1338 buildQuery will not try to parse.
1339
1340 =cut
1341
1342 sub parseQuery {
1343 my ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang) = @_;
1344
1345 my @operators = $operators ? @$operators : ();
1346 my @indexes = $indexes ? @$indexes : ();
1347 my @operands = $operands ? @$operands : ();
1348 my @limits = $limits ? @$limits : ();
1349 my @sort_by = $sort_by ? @$sort_by : ();
1350
1351 my $query = $operands[0];
1352 my $index;
1353 my $term;
1354 my $query_desc;
1355
1356 my $QParser;
1357 $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser') || $query =~ s/^qp=//);
1358 undef $QParser if ($query =~ m/^(ccl=|pqf=|cql=)/ || grep (/\w,\w|\w=\w/, @operands, @indexes) );
1359 undef $QParser if (scalar @limits > 0);
1360
1361 if ($QParser)
1362 {
1363 $QParser->custom_data->{'QueryAutoTruncate'} = C4::Context->preference('QueryAutoTruncate');
1364 $query = '';
1365 for ( my $ii = 0 ; $ii <= @operands ; $ii++ ) {
1366 next unless $operands[$ii];
1367 $query .= $operators[ $ii - 1 ] eq 'or' ? ' || ' : ' && '
1368 if ($query);
1369 if ( $operands[$ii] =~ /^[^"]\W*[-|_\w]*:\w.*[^"]$/ ) {
1370 $query .= $operands[$ii];
1371 }
1372 elsif ( $indexes[$ii] =~ m/su-/ ) {
1373 $query .= $indexes[$ii] . '(' . $operands[$ii] . ')';
1374 }
1375 else {
1376 $query .=
1377 ( $indexes[$ii] ? "$indexes[$ii]:" : '' ) . $operands[$ii];
1378 }
1379 }
1380 foreach my $limit (@limits) {
1381 }
1382 if ( scalar(@sort_by) > 0 ) {
1383 my $modifier_re =
1384 '#(' . join( '|', @{ $QParser->modifiers } ) . ')';
1385 $query =~ s/$modifier_re//g;
1386 foreach my $modifier (@sort_by) {
1387 $query .= " #$modifier";
1388 }
1389 }
1390
1391 $query_desc = $query;
1392 $query_desc =~ s/\s+/ /g;
1393 if ( C4::Context->preference("QueryWeightFields") ) {
1394 }
1395 $QParser->add_bib1_filter_map( 'su-br' => 'biblioserver' =>
1396 { 'target_syntax_callback' => \&_handle_exploding_index } );
1397 $QParser->add_bib1_filter_map( 'su-na' => 'biblioserver' =>
1398 { 'target_syntax_callback' => \&_handle_exploding_index } );
1399 $QParser->add_bib1_filter_map( 'su-rl' => 'biblioserver' =>
1400 { 'target_syntax_callback' => \&_handle_exploding_index } );
1401 $QParser->parse($query);
1402 $operands[0] = "pqf=" . $QParser->target_syntax('biblioserver');
1403 }
1404 else {
1405 require Koha::QueryParser::Driver::PQF;
1406 my $modifier_re = '#(' . join( '|', @{Koha::QueryParser::Driver::PQF->modifiers}) . ')';
1407 s/$modifier_re//g for @operands;
1408 }
1409
1410 return ( $operators, \@operands, $indexes, $limits, $sort_by, $scan, $lang, $query_desc);
1411 }
1412
1413 =head2 buildQuery
1414
1415 ( $error, $query,
1416 $simple_query, $query_cgi,
1417 $query_desc, $limit,
1418 $limit_cgi, $limit_desc,
1419 $query_type ) = buildQuery ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1420
1421 Build queries and limits in CCL, CGI, Human,
1422 handle truncation, stemming, field weighting, fuzziness, etc.
1423
1424 See verbose embedded documentation.
1425
1426
1427 =cut
1428
1429 sub buildQuery {
1430 my ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang) = @_;
1431
1432 warn "---------\nEnter buildQuery\n---------" if $DEBUG;
1433
1434 my $query_desc;
1435 ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang, $query_desc) = parseQuery($operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1436
1437 # dereference
1438 my @operators = $operators ? @$operators : ();
1439 my @indexes = $indexes ? @$indexes : ();
1440 my @operands = $operands ? @$operands : ();
1441 my @limits = $limits ? @$limits : ();
1442 my @sort_by = $sort_by ? @$sort_by : ();
1443
1444 my $stemming = C4::Context->preference("QueryStemming") || 0;
1445 my $auto_truncation = C4::Context->preference("QueryAutoTruncate") || 0;
1446 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
1447 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
1448
1449 my $query = $operands[0];
1450 my $simple_query = $operands[0];
1451
1452 # initialize the variables we're passing back
1453 my $query_cgi;
1454 my $query_type;
1455
1456 my $limit;
1457 my $limit_cgi;
1458 my $limit_desc;
1459
1460 my $cclq = 0;
1461 my $cclindexes = getIndexes();
1462 if ( $query !~ /\s*(ccl=|pqf=|cql=)/ ) {
1463 while ( !$cclq && $query =~ /(?:^|\W)([\w-]+)(,[\w-]+)*[:=]/g ) {
1464 my $dx = lc($1);
1465 $cclq = grep { lc($_) eq $dx } @$cclindexes;
1466 }
1467 $query = "ccl=$query" if $cclq;
1468 }
1469
1470 # for handling ccl, cql, pqf queries in diagnostic mode, skip the rest of the steps
1471 # DIAGNOSTIC ONLY!!
1472 if ( $query =~ /^ccl=/ ) {
1473 my $q=$';
1474 # This is needed otherwise ccl= and &limit won't work together, and
1475 # this happens when selecting a subject on the opac-detail page
1476 @limits = grep {!/^$/} @limits;
1477 if ( @limits ) {
1478 $q .= ' and '.join(' and ', @limits);
1479 }
1480 return ( undef, $q, $q, "q=ccl=".uri_escape_utf8($q), $q, '', '', '', 'ccl' );
1481 }
1482 if ( $query =~ /^cql=/ ) {
1483 return ( undef, $', $', "q=cql=".uri_escape_utf8($'), $', '', '', '', 'cql' );
1484 }
1485 if ( $query =~ /^pqf=/ ) {
1486 if ($query_desc) {
1487 $query_cgi = "q=".uri_escape_utf8($query_desc);
1488 } else {
1489 $query_desc = $';
1490 $query_cgi = "q=pqf=".uri_escape_utf8($');
1491 }
1492 return ( undef, $', $', $query_cgi, $query_desc, '', '', '', 'pqf' );
1493 }
1494
1495 # pass nested queries directly
1496 # FIXME: need better handling of some of these variables in this case
1497 # Nested queries aren't handled well and this implementation is flawed and causes users to be
1498 # unable to search for anything containing () commenting out, will be rewritten for 3.4.0
1499 # if ( $query =~ /(\(|\))/ ) {
1500 # return (
1501 # undef, $query, $simple_query, $query_cgi,
1502 # $query, $limit, $limit_cgi, $limit_desc,
1503 # 'ccl'
1504 # );
1505 # }
1506
1507 # Form-based queries are non-nested and fixed depth, so we can easily modify the incoming
1508 # query operands and indexes and add stemming, truncation, field weighting, etc.
1509 # Once we do so, we'll end up with a value in $query, just like if we had an
1510 # incoming $query from the user
1511 else {
1512 $query = ""
1513 ; # clear it out so we can populate properly with field-weighted, stemmed, etc. query
1514 my $previous_operand
1515 ; # a flag used to keep track if there was a previous query
1516 # if there was, we can apply the current operator
1517 # for every operand
1518 for ( my $i = 0 ; $i <= @operands ; $i++ ) {
1519
1520 # COMBINE OPERANDS, INDEXES AND OPERATORS
1521 if ( $operands[$i] ) {
1522 $operands[$i]=~s/^\s+//;
1523
1524 # A flag to determine whether or not to add the index to the query
1525 my $indexes_set;
1526
1527 # If the user is sophisticated enough to specify an index, turn off field weighting, and stemming handling
1528 if ( $operands[$i] =~ /\w(:|=)/ || $scan ) {
1529 $weight_fields = 0;
1530 $stemming = 0;
1531 } else {
1532 $operands[$i] =~ s/\?/{?}/g; # need to escape question marks
1533 }
1534 my $operand = $operands[$i];
1535 my $index = $indexes[$i];
1536
1537 # Add index-specific attributes
1538
1539 #Afaik, this 'yr' condition will only ever be met in the staff client advanced search
1540 #for "Publication date", since typing 'yr:YYYY' into the search box produces a CCL query,
1541 #which is processed higher up in this sub. Other than that, year searches are typically
1542 #handled as limits which are not processed her either.
1543
1544 # Search ranges: Date of Publication, st-numeric
1545 if ( $index =~ /(yr|st-numeric)/ ) {
1546 #weight_fields/relevance search causes errors with date ranges
1547 #In the case of YYYY-, it will only return records with a 'yr' of YYYY (not the range)
1548 #In the case of YYYY-YYYY, it will return no results
1549 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = 0;
1550 }
1551
1552 # Date of Acquisition
1553 elsif ( $index =~ /acqdate/ ) {
1554 #stemming and auto_truncation would have zero impact since it already is YYYY-MM-DD format
1555 #Weight_fields probably SHOULD be turned OFF, otherwise you'll get records floating to the
1556 #top of the results just because they have lots of item records matching that date.
1557 #Fuzzy actually only applies during _build_weighted_query, and is reset there anyway, so
1558 #irrelevant here
1559 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = 0;
1560 }
1561 # ISBN,ISSN,Standard Number, don't need special treatment
1562 elsif ( $index eq 'nb' || $index eq 'ns' ) {
1563 (
1564 $stemming, $auto_truncation,
1565 $weight_fields, $fuzzy_enabled
1566 ) = ( 0, 0, 0, 0 );
1567
1568 if ( $index eq 'nb' ) {
1569 if ( C4::Context->preference("SearchWithISBNVariations") ) {
1570 my @isbns = C4::Koha::GetVariationsOfISBN( $operand );
1571 $operands[$i] = $operand = '(nb=' . join(' OR nb=', @isbns) . ')';
1572 $indexes[$i] = $index = '';
1573 }
1574 }
1575 }
1576
1577 if(not $index){
1578 $index = 'kw';
1579 }
1580
1581 # Set default structure attribute (word list)
1582 my $struct_attr = q{};
1583 unless ( $indexes_set || !$index || $index =~ /,(st-|phr|ext|wrdl)/ || $index =~ /^(nb|ns)$/ ) {
1584 $struct_attr = ",wrdl";
1585 }
1586
1587 # Some helpful index variants
1588 my $index_plus = $index . $struct_attr . ':';
1589 my $index_plus_comma = $index . $struct_attr . ',';
1590
1591 if ($auto_truncation){
1592 unless ( $index =~ /,(st-|phr|ext)/ ) {
1593 #FIXME only valid with LTR scripts
1594 $operand=join(" ",map{
1595 (index($_,"*")>0?"$_":"$_*")
1596 }split (/\s+/,$operand));
1597 warn $operand if $DEBUG;
1598 }
1599 }
1600
1601 # Detect Truncation
1602 my $truncated_operand;
1603 my( $nontruncated, $righttruncated, $lefttruncated,
1604 $rightlefttruncated, $regexpr
1605 ) = _detect_truncation( $operand, $index );
1606 warn
1607 "TRUNCATION: NON:>@$nontruncated< RIGHT:>@$righttruncated< LEFT:>@$lefttruncated< RIGHTLEFT:>@$rightlefttruncated< REGEX:>@$regexpr<"
1608 if $DEBUG;
1609
1610 # Apply Truncation
1611 if (
1612 scalar(@$righttruncated) + scalar(@$lefttruncated) +
1613 scalar(@$rightlefttruncated) > 0 )
1614 {
1615
1616 # Don't field weight or add the index to the query, we do it here
1617 $indexes_set = 1;
1618 undef $weight_fields;
1619 my $previous_truncation_operand;
1620 if (scalar @$nontruncated) {
1621 $truncated_operand .= "$index_plus @$nontruncated ";
1622 $previous_truncation_operand = 1;
1623 }
1624 if (scalar @$righttruncated) {
1625 $truncated_operand .= "and " if $previous_truncation_operand;
1626 $truncated_operand .= $index_plus_comma . "rtrn:@$righttruncated ";
1627 $previous_truncation_operand = 1;
1628 }
1629 if (scalar @$lefttruncated) {
1630 $truncated_operand .= "and " if $previous_truncation_operand;
1631 $truncated_operand .= $index_plus_comma . "ltrn:@$lefttruncated ";
1632 $previous_truncation_operand = 1;
1633 }
1634 if (scalar @$rightlefttruncated) {
1635 $truncated_operand .= "and " if $previous_truncation_operand;
1636 $truncated_operand .= $index_plus_comma . "rltrn:@$rightlefttruncated ";
1637 $previous_truncation_operand = 1;
1638 }
1639 }
1640 $operand = $truncated_operand if $truncated_operand;
1641 warn "TRUNCATED OPERAND: >$truncated_operand<" if $DEBUG;
1642
1643 # Handle Stemming
1644 my $stemmed_operand;
1645 $stemmed_operand = _build_stemmed_operand($operand, $lang)
1646 if $stemming;
1647
1648 warn "STEMMED OPERAND: >$stemmed_operand<" if $DEBUG;
1649
1650 # Handle Field Weighting
1651 my $weighted_operand;
1652 if ($weight_fields) {
1653 $weighted_operand = _build_weighted_query( $operand, $stemmed_operand, $index );
1654 $operand = $weighted_operand;
1655 $indexes_set = 1;
1656 }
1657
1658 warn "FIELD WEIGHTED OPERAND: >$weighted_operand<" if $DEBUG;
1659
1660 ($query,$query_cgi,$query_desc,$previous_operand) = _build_initial_query({
1661 query => $query,
1662 query_cgi => $query_cgi,
1663 query_desc => $query_desc,
1664 operator => ($operators[ $i - 1 ]) ? $operators[ $i - 1 ] : '',
1665 parsed_operand => $operand,
1666 original_operand => ($operands[$i]) ? $operands[$i] : '',
1667 index => $index,
1668 index_plus => $index_plus,
1669 indexes_set => $indexes_set,
1670 previous_operand => $previous_operand,
1671 });
1672
1673 } #/if $operands
1674 } # /for
1675 }
1676 warn "QUERY BEFORE LIMITS: >$query<" if $DEBUG;
1677
1678 # add limits
1679 my %group_OR_limits;
1680 my $availability_limit;
1681 foreach my $this_limit (@limits) {
1682 next unless $this_limit;
1683 if ( $this_limit =~ /available/ ) {
1684 #
1685 ## 'available' is defined as (items.onloan is NULL) and (items.itemlost = 0)
1686 ## In English:
1687 ## all records not indexed in the onloan register (zebra) and all records with a value of lost equal to 0
1688 $availability_limit .=
1689 "( ( allrecords,AlwaysMatches='' not onloan,AlwaysMatches='') and (lost,st-numeric=0) )"; #or ( allrecords,AlwaysMatches='' not lost,AlwaysMatches='')) )";
1690 $limit_cgi .= "&limit=available";
1691 $limit_desc .= "";
1692 }
1693
1694 # group_OR_limits, prefixed by mc-
1695 # OR every member of the group
1696 elsif ( $this_limit =~ /mc/ ) {
1697 my ($k,$v) = split(/:/, $this_limit,2);
1698 if ( $k !~ /mc-i(tem)?type/ ) {
1699 # in case the mc-ccode value has complicating chars like ()'s inside it we wrap in quotes
1700 $this_limit =~ tr/"//d;
1701 $this_limit = $k.":'".$v."'";
1702 }
1703
1704 $group_OR_limits{$k} .= " or " if $group_OR_limits{$k};
1705 $limit_desc .= " or " if $group_OR_limits{$k};
1706 $group_OR_limits{$k} .= "$this_limit";
1707 $limit_cgi .= "&limit=" . uri_escape_utf8($this_limit);
1708 $limit_desc .= " $this_limit";
1709 }
1710
1711 # Regular old limits
1712 else {
1713 $limit .= " and " if $limit || $query;
1714 $limit .= "$this_limit";
1715 $limit_cgi .= "&limit=" . uri_escape_utf8($this_limit);
1716 if ($this_limit =~ /^branch:(.+)/) {
1717 my $branchcode = $1;
1718 my $library = Koha::Libraries->find( $branchcode );
1719 if (defined $library) {
1720 $limit_desc .= " branch:" . $library->branchname;
1721 } else {
1722 $limit_desc .= " $this_limit";
1723 }
1724 } else {
1725 $limit_desc .= " $this_limit";
1726 }
1727 }
1728 }
1729 foreach my $k (keys (%group_OR_limits)) {
1730 $limit .= " and " if ( $query || $limit );
1731 $limit .= "($group_OR_limits{$k})";
1732 }
1733 if ($availability_limit) {
1734 $limit .= " and " if ( $query || $limit );
1735 $limit .= "($availability_limit)";
1736 }
1737
1738 # Normalize the query and limit strings
1739 # This is flawed , means we can't search anything with : in it
1740 # if user wants to do ccl or cql, start the query with that
1741 # $query =~ s/:/=/g;
1742 #NOTE: We use several several different regexps here as you can't have variable length lookback assertions
1743 $query =~ s/(?<=(ti|au|pb|su|an|kw|mc|nb|ns)):/=/g;
1744 $query =~ s/(?<=(wrdl)):/=/g;
1745 $query =~ s/(?<=(trn|phr)):/=/g;
1746 $query =~ s/(?<=(st-numeric)):/=/g;
1747 $query =~ s/(?<=(st-year)):/=/g;
1748 $query =~ s/(?<=(st-date-normalized)):/=/g;
1749 $limit =~ s/:/=/g;
1750 for ( $query, $query_desc, $limit, $limit_desc ) {
1751 s/ +/ /g; # remove extra spaces
1752 s/^ //g; # remove any beginning spaces
1753 s/ $//g; # remove any ending spaces
1754 s/==/=/g; # remove double == from query
1755 }
1756 $query_cgi =~ s/^&//; # remove unnecessary & from beginning of the query cgi
1757
1758 for ($query_cgi,$simple_query) {
1759 s/"//g;
1760 }
1761 # append the limit to the query
1762 $query .= " " . $limit;
1763
1764 # Warnings if DEBUG
1765 if ($DEBUG) {
1766 warn "QUERY:" . $query;
1767 warn "QUERY CGI:" . $query_cgi;
1768 warn "QUERY DESC:" . $query_desc;
1769 warn "LIMIT:" . $limit;
1770 warn "LIMIT CGI:" . $limit_cgi;
1771 warn "LIMIT DESC:" . $limit_desc;
1772 warn "---------\nLeave buildQuery\n---------";
1773 }
1774
1775 return (
1776 undef, $query, $simple_query, $query_cgi,
1777 $query_desc, $limit, $limit_cgi, $limit_desc,
1778 $query_type
1779 );
1780 }
1781
1782 =head2 _build_initial_query
1783
1784 ($query, $query_cgi, $query_desc, $previous_operand) = _build_initial_query($initial_query_params);
1785
1786 Build a section of the initial query containing indexes, operators, and operands.
1787
1788 =cut
1789
1790 sub _build_initial_query {
1791 my ($params) = @_;
1792
1793 my $operator = "";
1794 if ($params->{previous_operand}){
1795 #If there is a previous operand, add a supplied operator or the default 'and'
1796 $operator = ($params->{operator}) ? " ".($params->{operator})." " : ' and ';
1797 }
1798
1799 #NOTE: indexes_set is typically set when doing truncation or field weighting
1800 my $operand = ($params->{indexes_set}) ? $params->{parsed_operand} : $params->{index_plus}.$params->{parsed_operand};
1801
1802 #e.g. "kw,wrdl:test"
1803 #e.g. " and kw,wrdl:test"
1804 $params->{query} .= $operator . $operand;
1805
1806 $params->{query_cgi} .= "&op=".uri_escape_utf8($operator) if $operator;
1807 $params->{query_cgi} .= "&idx=".uri_escape_utf8($params->{index}) if $params->{index};
1808 $params->{query_cgi} .= "&q=".uri_escape_utf8($params->{original_operand}) if $params->{original_operand};
1809
1810 #e.g. " and kw,wrdl: test"
1811 $params->{query_desc} .= $operator . $params->{index_plus} . " " . $params->{original_operand};
1812
1813 $params->{previous_operand} = 1 unless $params->{previous_operand}; #If there is no previous operand, mark this as one
1814
1815 return ($params->{query}, $params->{query_cgi}, $params->{query_desc}, $params->{previous_operand});
1816 }
1817
1818 =head2 searchResults
1819
1820 my @search_results = searchResults($search_context, $searchdesc, $hits,
1821 $results_per_page, $offset, $scan,
1822 @marcresults);
1823
1824 Format results in a form suitable for passing to the template
1825
1826 =cut
1827
1828 # IMO this subroutine is pretty messy still -- it's responsible for
1829 # building the HTML output for the template
1830 sub searchResults {
1831 my ( $search_context, $searchdesc, $hits, $results_per_page, $offset, $scan, $marcresults ) = @_;
1832 my $dbh = C4::Context->dbh;
1833 my @newresults;
1834
1835 require C4::Items;
1836
1837 $search_context = 'opac' if !$search_context || $search_context ne 'intranet';
1838 my ($is_opac, $hidelostitems);
1839 if ($search_context eq 'opac') {
1840 $hidelostitems = C4::Context->preference('hidelostitems');
1841 $is_opac = 1;
1842 }
1843
1844 #Build branchnames hash
1845 my %branches = map { $_->branchcode => $_->branchname } Koha::Libraries->search({}, { order_by => 'branchname' });
1846
1847 # FIXME - We build an authorised values hash here, using the default framework
1848 # though it is possible to have different authvals for different fws.
1849
1850 my $shelflocations =GetKohaAuthorisedValues('items.location','');
1851
1852 # get notforloan authorised value list (see $shelflocations FIXME)
1853 my $notforloan_authorised_value = GetAuthValCode('items.notforloan','');
1854
1855 #Get itemtype hash
1856 my %itemtypes = %{ GetItemTypes() };
1857
1858 #search item field code
1859 my ($itemtag, undef) = &GetMarcFromKohaField( "items.itemnumber", "" );
1860
1861 ## find column names of items related to MARC
1862 my %subfieldstosearch;
1863 my @columns = Koha::Database->new()->schema()->resultset('Item')->result_source->columns;
1864 for my $column ( @columns ) {
1865 my ( $tagfield, $tagsubfield ) =
1866 &GetMarcFromKohaField( "items." . $column, "" );
1867 if ( defined $tagsubfield ) {
1868 $subfieldstosearch{$column} = $tagsubfield;
1869 }
1870 }
1871
1872 # handle which records to actually retrieve
1873 my $times;
1874 if ( $hits && $offset + $results_per_page <= $hits ) {
1875 $times = $offset + $results_per_page;
1876 }
1877 else {
1878 $times = $hits; # FIXME: if $hits is undefined, why do we want to equal it?
1879 }
1880
1881 my $marcflavour = C4::Context->preference("marcflavour");
1882 # We get the biblionumber position in MARC
1883 my ($bibliotag,$bibliosubf)=GetMarcFromKohaField('biblio.biblionumber','');
1884
1885 # set stuff for XSLT processing here once, not later again for every record we retrieved
1886 my $interface = $search_context eq 'opac' ? 'OPAC' : '';
1887 my $xslsyspref = $interface . "XSLTResultsDisplay";
1888 my $xslfile = C4::Context->preference($xslsyspref);
1889 my $lang = $xslfile ? C4::Languages::getlanguage() : undef;
1890 my $sysxml = $xslfile ? C4::XSLT::get_xslt_sysprefs() : undef;
1891
1892 # loop through all of the records we've retrieved
1893 for ( my $i = $offset ; $i <= $times - 1 ; $i++ ) {
1894
1895 my $marcrecord;
1896 if ($scan) {
1897 # For Scan searches we built USMARC data
1898 $marcrecord = MARC::Record->new_from_usmarc( $marcresults->[$i]);
1899 } else {
1900 # Normal search, render from Zebra's output
1901 $marcrecord = new_record_from_zebra(
1902 'biblioserver',
1903 $marcresults->[$i]
1904 );
1905
1906 if ( ! defined $marcrecord ) {
1907 warn "ERROR DECODING RECORD - $@: " . $marcresults->[$i];
1908 next;
1909 }
1910 }
1911
1912 my $fw = $scan
1913 ? undef
1914 : $bibliotag < 10
1915 ? GetFrameworkCode($marcrecord->field($bibliotag)->data)
1916 : GetFrameworkCode($marcrecord->subfield($bibliotag,$bibliosubf));
1917
1918 SetUTF8Flag($marcrecord);
1919 my $oldbiblio = TransformMarcToKoha( $marcrecord, $fw );
1920 $oldbiblio->{subtitle} = GetRecordValue('subtitle', $marcrecord, $fw);
1921 $oldbiblio->{result_number} = $i + 1;
1922
1923 # add imageurl to itemtype if there is one
1924 $oldbiblio->{imageurl} = getitemtypeimagelocation( $search_context, $itemtypes{ $oldbiblio->{itemtype} }->{imageurl} );
1925
1926 $oldbiblio->{normalized_upc} = GetNormalizedUPC( $marcrecord,$marcflavour);
1927 $oldbiblio->{normalized_ean} = GetNormalizedEAN( $marcrecord,$marcflavour);
1928 $oldbiblio->{normalized_oclc} = GetNormalizedOCLCNumber($marcrecord,$marcflavour);
1929 $oldbiblio->{normalized_isbn} = GetNormalizedISBN(undef,$marcrecord,$marcflavour);
1930 $oldbiblio->{content_identifier_exists} = 1 if ($oldbiblio->{normalized_isbn} or $oldbiblio->{normalized_oclc} or $oldbiblio->{normalized_ean} or $oldbiblio->{normalized_upc});
1931
1932 # edition information, if any
1933 $oldbiblio->{edition} = $oldbiblio->{editionstatement};
1934 $oldbiblio->{description} = $itemtypes{ $oldbiblio->{itemtype} }->{translated_description};
1935 # Build summary if there is one (the summary is defined in the itemtypes table)
1936 # FIXME: is this used anywhere, I think it can be commented out? -- JF
1937 if ( $itemtypes{ $oldbiblio->{itemtype} }->{summary} ) {
1938 my $summary = $itemtypes{ $oldbiblio->{itemtype} }->{summary};
1939 my @fields = $marcrecord->fields();
1940
1941 my $newsummary;
1942 foreach my $line ( "$summary\n" =~ /(.*)\n/g ){
1943 my $tags = {};
1944 foreach my $tag ( $line =~ /\[(\d{3}[\w|\d])\]/ ) {
1945 $tag =~ /(.{3})(.)/;
1946 if($marcrecord->field($1)){
1947 my @abc = $marcrecord->field($1)->subfield($2);
1948 $tags->{$tag} = $#abc + 1 ;
1949 }
1950 }
1951
1952 # We catch how many times to repeat this line
1953 my $max = 0;
1954 foreach my $tag (keys(%$tags)){
1955 $max = $tags->{$tag} if($tags->{$tag} > $max);
1956 }
1957
1958 # we replace, and repeat each line
1959 for (my $i = 0 ; $i < $max ; $i++){
1960 my $newline = $line;
1961
1962 foreach my $tag ( $newline =~ /\[(\d{3}[\w|\d])\]/g ) {
1963 $tag =~ /(.{3})(.)/;
1964
1965 if($marcrecord->field($1)){
1966 my @repl = $marcrecord->field($1)->subfield($2);
1967 my $subfieldvalue = $repl[$i];
1968 $newline =~ s/\[$tag\]/$subfieldvalue/g;
1969 }
1970 }
1971 $newsummary .= "$newline\n";
1972 }
1973 }
1974
1975 $newsummary =~ s/\[(.*?)]//g;
1976 $newsummary =~ s/\n/<br\/>/g;
1977 $oldbiblio->{summary} = $newsummary;
1978 }
1979
1980 # Pull out the items fields
1981 my @fields = $marcrecord->field($itemtag);
1982 my $marcflavor = C4::Context->preference("marcflavour");
1983 # adding linked items that belong to host records
1984 my $analyticsfield = '773';
1985 if ($marcflavor eq 'MARC21' || $marcflavor eq 'NORMARC') {
1986 $analyticsfield = '773';
1987 } elsif ($marcflavor eq 'UNIMARC') {
1988 $analyticsfield = '461';
1989 }
1990 foreach my $hostfield ( $marcrecord->field($analyticsfield)) {
1991 my $hostbiblionumber = $hostfield->subfield("0");
1992 my $linkeditemnumber = $hostfield->subfield("9");
1993 if(!$hostbiblionumber eq undef){
1994 my $hostbiblio = GetMarcBiblio($hostbiblionumber, 1);
1995 my ($itemfield, undef) = GetMarcFromKohaField( 'items.itemnumber', GetFrameworkCode($hostbiblionumber) );
1996 if(!$hostbiblio eq undef){
1997 my @hostitems = $hostbiblio->field($itemfield);
1998 foreach my $hostitem (@hostitems){
1999 if ($hostitem->subfield("9") eq $linkeditemnumber){
2000 my $linkeditem =$hostitem;
2001 # append linked items if they exist
2002 if (!$linkeditem eq undef){
2003 push (@fields, $linkeditem);}
2004 }
2005 }
2006 }
2007 }
2008 }
2009
2010 # Setting item statuses for display
2011 my @available_items_loop;
2012 my @onloan_items_loop;
2013 my @other_items_loop;
2014
2015 my $available_items;
2016 my $onloan_items;
2017 my $other_items;
2018
2019 my $ordered_count = 0;
2020 my $available_count = 0;
2021 my $onloan_count = 0;
2022 my $longoverdue_count = 0;
2023 my $other_count = 0;
2024 my $withdrawn_count = 0;
2025 my $itemlost_count = 0;
2026 my $hideatopac_count = 0;
2027 my $itembinding_count = 0;
2028 my $itemdamaged_count = 0;
2029 my $item_in_transit_count = 0;
2030 my $can_place_holds = 0;
2031 my $item_onhold_count = 0;
2032 my $notforloan_count = 0;
2033 my $items_count = scalar(@fields);
2034 my $maxitems_pref = C4::Context->preference('maxItemsinSearchResults');
2035 my $maxitems = $maxitems_pref ? $maxitems_pref - 1 : 1;
2036 my @hiddenitems; # hidden itemnumbers based on OpacHiddenItems syspref
2037
2038 # loop through every item
2039 foreach my $field (@fields) {
2040 my $item;
2041
2042 # populate the items hash
2043 foreach my $code ( keys %subfieldstosearch ) {
2044 $item->{$code} = $field->subfield( $subfieldstosearch{$code} );
2045 }
2046 $item->{description} = $itemtypes{ $item->{itype} }{translated_description};
2047
2048 # OPAC hidden items
2049 if ($is_opac) {
2050 # hidden because lost
2051 if ($hidelostitems && $item->{itemlost}) {
2052 $hideatopac_count++;
2053 next;
2054 }
2055 # hidden based on OpacHiddenItems syspref
2056 my @hi = C4::Items::GetHiddenItemnumbers($item);
2057 if (scalar @hi) {
2058 push @hiddenitems, @hi;
2059 $hideatopac_count++;
2060 next;
2061 }
2062 }
2063
2064 my $hbranch = C4::Context->preference('StaffSearchResultsDisplayBranch');
2065 my $otherbranch = $hbranch eq 'homebranch' ? 'holdingbranch' : 'homebranch';
2066
2067 # set item's branch name, use HomeOrHoldingBranch syspref first, fall back to the other one
2068 if ($item->{$hbranch}) {
2069 $item->{'branchname'} = $branches{$item->{$hbranch}};
2070 }
2071 elsif ($item->{$otherbranch}) { # Last resort
2072 $item->{'branchname'} = $branches{$item->{$otherbranch}};
2073 }
2074
2075 my $prefix = $item->{$hbranch} . '--' . $item->{location} . $item->{itype} . $item->{itemcallnumber};
2076 # For each grouping of items (onloan, available, unavailable), we build a key to store relevant info about that item
2077 my $userenv = C4::Context->userenv;
2078 if ( $item->{onloan}
2079 && !( C4::Members::GetHideLostItemsPreference( $userenv->{'number'} ) && $item->{itemlost} ) )
2080 {
2081 $onloan_count++;
2082 my $key = $prefix . $item->{onloan} . $item->{barcode};
2083 $onloan_items->{$key}->{due_date} = output_pref( { dt => dt_from_string( $item->{onloan} ), dateonly => 1 } );
2084 $onloan_items->{$key}->{count}++ if $item->{$hbranch};
2085 $onloan_items->{$key}->{branchname} = $item->{branchname};
2086 $onloan_items->{$key}->{location} = $shelflocations->{ $item->{location} };
2087 $onloan_items->{$key}->{itemcallnumber} = $item->{itemcallnumber};
2088 $onloan_items->{$key}->{description} = $item->{description};
2089 $onloan_items->{$key}->{imageurl} =
2090 getitemtypeimagelocation( $search_context, $itemtypes{ $item->{itype} }->{imageurl} );
2091
2092 # if something's checked out and lost, mark it as 'long overdue'
2093 if ( $item->{itemlost} ) {
2094 $onloan_items->{$key}->{longoverdue}++;
2095 $longoverdue_count++;
2096 }
2097 else { # can place holds as long as item isn't lost
2098 $can_place_holds = 1;
2099 }
2100 }
2101
2102 # items not on loan, but still unavailable ( lost, withdrawn, damaged )
2103 else {
2104
2105 $item->{notforloan}=1 if !$item->{notforloan} && $itemtypes{ C4::Context->preference("item-level_itypes")? $item->{itype}: $oldbiblio->{itemtype} }->{notforloan};
2106
2107 # item is on order
2108 if ( $item->{notforloan} < 0 ) {
2109 $ordered_count++;
2110 } elsif ( $item->{notforloan} > 0 ) {
2111 $notforloan_count++;
2112 }
2113
2114 # is item in transit?
2115 my $transfertwhen = '';
2116 my ($transfertfrom, $transfertto);
2117
2118 # is item on the reserve shelf?
2119 my $reservestatus = '';
2120
2121 unless ($item->{withdrawn}
2122 || $item->{itemlost}
2123 || $item->{damaged}
2124 || $item->{notforloan}
2125 || ( C4::Context->preference('MaxSearchResultsItemsPerRecordStatusCheck')
2126 && $items_count > C4::Context->preference('MaxSearchResultsItemsPerRecordStatusCheck') ) ) {
2127
2128 # A couple heuristics to limit how many times
2129 # we query the database for item transfer information, sacrificing
2130 # accuracy in some cases for speed;
2131 #
2132 # 1. don't query if item has one of the other statuses
2133 # 2. don't check transit status if the bib has
2134 # more than 20 items
2135 #
2136 # FIXME: to avoid having the query the database like this, and to make
2137 # the in transit status count as unavailable for search limiting,
2138 # should map transit status to record indexed in Zebra.
2139 #
2140 ($transfertwhen, $transfertfrom, $transfertto) = C4::Circulation::GetTransfers($item->{itemnumber});
2141 $reservestatus = C4::Reserves::GetReserveStatus( $item->{itemnumber} );
2142 }
2143
2144 # item is withdrawn, lost, damaged, not for loan, reserved or in transit
2145 if ( $item->{withdrawn}
2146 || $item->{itemlost}
2147 || $item->{damaged}
2148 || $item->{notforloan}
2149 || $reservestatus eq 'Waiting'
2150 || ($transfertwhen ne ''))
2151 {
2152 $withdrawn_count++ if $item->{withdrawn};
2153 $itemlost_count++ if $item->{itemlost};
2154 $itemdamaged_count++ if $item->{damaged};
2155 $item_in_transit_count++ if $transfertwhen ne '';
2156 $item_onhold_count++ if $reservestatus eq 'Waiting';
2157 $item->{status} = $item->{withdrawn} . "-" . $item->{itemlost} . "-" . $item->{damaged} . "-" . $item->{notforloan};
2158
2159 # can place a hold on a item if
2160 # not lost nor withdrawn
2161 # not damaged unless AllowHoldsOnDamagedItems is true
2162 # item is either for loan or on order (notforloan < 0)
2163 $can_place_holds = 1
2164 if (
2165 !$item->{itemlost}
2166 && !$item->{withdrawn}
2167 && ( !$item->{damaged} || C4::Context->preference('AllowHoldsOnDamagedItems') )
2168 && ( !$item->{notforloan} || $item->{notforloan} < 0 )
2169 );
2170
2171 $other_count++;
2172
2173 my $key = $prefix . $item->{status};
2174 foreach (qw(withdrawn itemlost damaged branchname itemcallnumber)) {
2175 $other_items->{$key}->{$_} = $item->{$_};
2176 }
2177 $other_items->{$key}->{intransit} = ( $transfertwhen ne '' ) ? 1 : 0;
2178 $other_items->{$key}->{onhold} = ($reservestatus) ? 1 : 0;
2179 $other_items->{$key}->{notforloan} = GetAuthorisedValueDesc('','',$item->{notforloan},'','',$notforloan_authorised_value) if $notforloan_authorised_value and $item->{notforloan};
2180 $other_items->{$key}->{count}++ if $item->{$hbranch};
2181 $other_items->{$key}->{location} = $shelflocations->{ $item->{location} };
2182 $other_items->{$key}->{description} = $item->{description};
2183 $other_items->{$key}->{imageurl} = getitemtypeimagelocation( $search_context, $itemtypes{ $item->{itype} }->{imageurl} );
2184 }
2185 # item is available
2186 else {
2187 $can_place_holds = 1;
2188 $available_count++;
2189 $available_items->{$prefix}->{count}++ if $item->{$hbranch};
2190 foreach (qw(branchname itemcallnumber description)) {
2191 $available_items->{$prefix}->{$_} = $item->{$_};
2192 }
2193 $available_items->{$prefix}->{location} = $shelflocations->{ $item->{location} };
2194 $available_items->{$prefix}->{imageurl} = getitemtypeimagelocation( $search_context, $itemtypes{ $item->{itype} }->{imageurl} );
2195 }
2196 }
2197 } # notforloan, item level and biblioitem level
2198
2199 # if all items are hidden, do not show the record
2200 if ($items_count > 0 && $hideatopac_count == $items_count) {
2201 next;
2202 }
2203
2204 my ( $availableitemscount, $onloanitemscount, $otheritemscount );
2205 for my $key ( sort keys %$onloan_items ) {
2206 (++$onloanitemscount > $maxitems) and last;
2207 push @onloan_items_loop, $onloan_items->{$key};
2208 }
2209 for my $key ( sort keys %$other_items ) {
2210 (++$otheritemscount > $maxitems) and last;
2211 push @other_items_loop, $other_items->{$key};
2212 }
2213 for my $key ( sort keys %$available_items ) {
2214 (++$availableitemscount > $maxitems) and last;
2215 push @available_items_loop, $available_items->{$key}
2216 }
2217
2218 # XSLT processing of some stuff
2219 # we fetched the sysprefs already before the loop through all retrieved record!
2220 if (!$scan && $xslfile) {
2221 $oldbiblio->{XSLTResultsRecord} = XSLTParse4Display($oldbiblio->{biblionumber}, $marcrecord, $xslsyspref, 1, \@hiddenitems, $sysxml, $xslfile, $lang);
2222 # the last parameter tells Koha to clean up the problematic ampersand entities that Zebra outputs
2223 }
2224
2225 # if biblio level itypes are used and itemtype is notforloan, it can't be reserved either
2226 if (!C4::Context->preference("item-level_itypes")) {
2227 if ($itemtypes{ $oldbiblio->{itemtype} }->{notforloan}) {
2228 $can_place_holds = 0;
2229 }
2230 }
2231 $oldbiblio->{norequests} = 1 unless $can_place_holds;
2232 $oldbiblio->{itemsplural} = 1 if $items_count > 1;
2233 $oldbiblio->{items_count} = $items_count;
2234 $oldbiblio->{available_items_loop} = \@available_items_loop;
2235 $oldbiblio->{onloan_items_loop} = \@onloan_items_loop;
2236 $oldbiblio->{other_items_loop} = \@other_items_loop;
2237 $oldbiblio->{availablecount} = $available_count;
2238 $oldbiblio->{availableplural} = 1 if $available_count > 1;
2239 $oldbiblio->{onloancount} = $onloan_count;
2240 $oldbiblio->{onloanplural} = 1 if $onloan_count > 1;
2241 $oldbiblio->{othercount} = $other_count;
2242 $oldbiblio->{otherplural} = 1 if $other_count > 1;
2243 $oldbiblio->{withdrawncount} = $withdrawn_count;
2244 $oldbiblio->{itemlostcount} = $itemlost_count;
2245 $oldbiblio->{damagedcount} = $itemdamaged_count;
2246 $oldbiblio->{intransitcount} = $item_in_transit_count;
2247 $oldbiblio->{onholdcount} = $item_onhold_count;
2248 $oldbiblio->{orderedcount} = $ordered_count;
2249 $oldbiblio->{notforloancount} = $notforloan_count;
2250
2251 if (C4::Context->preference("AlternateHoldingsField") && $items_count == 0) {
2252 my $fieldspec = C4::Context->preference("AlternateHoldingsField");
2253 my $subfields = substr $fieldspec, 3;
2254 my $holdingsep = C4::Context->preference("AlternateHoldingsSeparator") || ' ';
2255 my @alternateholdingsinfo = ();
2256 my @holdingsfields = $marcrecord->field(substr $fieldspec, 0, 3);
2257 my $alternateholdingscount = 0;
2258
2259 for my $field (@holdingsfields) {
2260 my %holding = ( holding => '' );
2261 my $havesubfield = 0;
2262 for my $subfield ($field->subfields()) {
2263 if ((index $subfields, $$subfield[0]) >= 0) {
2264 $holding{'holding'} .= $holdingsep if (length $holding{'holding'} > 0);
2265 $holding{'holding'} .= $$subfield[1];
2266 $havesubfield++;
2267 }
2268 }
2269 if ($havesubfield) {
2270 push(@alternateholdingsinfo, \%holding);
2271 $alternateholdingscount++;
2272 }
2273 }
2274
2275 $oldbiblio->{'ALTERNATEHOLDINGS'} = \@alternateholdingsinfo;
2276 $oldbiblio->{'alternateholdings_count'} = $alternateholdingscount;
2277 }
2278
2279 push( @newresults, $oldbiblio );
2280 }
2281
2282 return @newresults;
2283 }
2284
2285 =head2 enabled_staff_search_views
2286
2287 %hash = enabled_staff_search_views()
2288
2289 This function returns a hash that contains three flags obtained from the system
2290 preferences, used to determine whether a particular staff search results view
2291 is enabled.
2292
2293 =over 2
2294
2295 =item C<Output arg:>
2296
2297 * $hash{can_view_MARC} is true only if the MARC view is enabled
2298 * $hash{can_view_ISBD} is true only if the ISBD view is enabled
2299 * $hash{can_view_labeledMARC} is true only if the Labeled MARC view is enabled
2300
2301 =item C<usage in the script:>
2302
2303 =back
2304
2305 $template->param ( C4::Search::enabled_staff_search_views );
2306
2307 =cut
2308
2309 sub enabled_staff_search_views
2310 {
2311 return (
2312 can_view_MARC => C4::Context->preference('viewMARC'), # 1 if the staff search allows the MARC view
2313 can_view_ISBD => C4::Context->preference('viewISBD'), # 1 if the staff search allows the ISBD view
2314 can_view_labeledMARC => C4::Context->preference('viewLabeledMARC'), # 1 if the staff search allows the Labeled MARC view
2315 );
2316 }
2317
2318 =head2 z3950_search_args
2319
2320 $arrayref = z3950_search_args($matchpoints)
2321
2322 This function returns an array reference that contains the search parameters to be
2323 passed to the Z39.50 search script (z3950_search.pl). The array elements
2324 are hash refs whose keys are name and value, and whose values are the
2325 name of a search parameter, the value of that search parameter and the URL encoded
2326 value of that parameter.
2327
2328 The search parameter names are lccn, isbn, issn, title, author, dewey and subject.
2329
2330 The search parameter values are obtained from the bibliographic record whose
2331 data is in a hash reference in $matchpoints, as returned by Biblio::GetBiblioData().
2332
2333 If $matchpoints is a scalar, it is assumed to be an unnamed query descriptor, e.g.
2334 a general purpose search argument. In this case, the returned array contains only
2335 entry: the key is 'title' and the value is derived from $matchpoints.
2336
2337 If a search parameter value is undefined or empty, it is not included in the returned
2338 array.
2339
2340 The returned array reference may be passed directly to the template parameters.
2341
2342 =over 2
2343
2344 =item C<Output arg:>
2345
2346 * $array containing hash refs as described above
2347
2348 =item C<usage in the script:>
2349
2350 =back
2351
2352 $data = Biblio::GetBiblioData($bibno);
2353 $template->param ( MYLOOP => C4::Search::z3950_search_args($data) )
2354
2355 *OR*
2356
2357 $template->param ( MYLOOP => C4::Search::z3950_search_args($searchscalar) )
2358
2359 =cut
2360
2361 sub z3950_search_args {
2362 my $bibrec = shift;
2363
2364 my $isbn_string = ref( $bibrec ) ? $bibrec->{title} : $bibrec;
2365 my $isbn = Business::ISBN->new( $isbn_string );
2366
2367 if (defined $isbn && $isbn->is_valid)
2368 {
2369 if ( ref($bibrec) ) {
2370 $bibrec->{isbn} = $isbn_string;
2371 $bibrec->{title} = undef;
2372 } else {
2373 $bibrec = { isbn => $isbn_string };
2374 }
2375 }
2376 else {
2377 $bibrec = { title => $bibrec } if !ref $bibrec;
2378 }
2379 my $array = [];
2380 for my $field (qw/ lccn isbn issn title author dewey subject /)
2381 {
2382 push @$array, { name => $field, value => $bibrec->{$field} }
2383 if defined $bibrec->{$field};
2384 }
2385 return $array;
2386 }
2387
2388 =head2 GetDistinctValues($field);
2389
2390 C<$field> is a reference to the fields array
2391
2392 =cut
2393
2394 sub GetDistinctValues {
2395 my ($fieldname,$string)=@_;
2396 # returns a reference to a hash of references to branches...
2397 if ($fieldname=~/\./){
2398 my ($table,$column)=split /\./, $fieldname;
2399 my $dbh = C4::Context->dbh;
2400 warn "select DISTINCT($column) as value, count(*) as cnt from $table group by lib order by $column " if $DEBUG;
2401 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 ");
2402 $sth->execute;
2403 my $elements=$sth->fetchall_arrayref({});
2404 return $elements;
2405 }
2406 else {
2407 $string||= qq("");
2408 my @servers=qw<biblioserver authorityserver>;
2409 my (@zconns,@results);
2410 for ( my $i = 0 ; $i < @servers ; $i++ ) {
2411 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
2412 $results[$i] =
2413 $zconns[$i]->scan(
2414 ZOOM::Query::CCL2RPN->new( qq"$fieldname $string", $zconns[$i])
2415 );
2416 }
2417 # The big moment: asynchronously retrieve results from all servers
2418 my @elements;
2419 _ZOOM_event_loop(
2420 \@zconns,
2421 \@results,
2422 sub {
2423 my ( $i, $size ) = @_;
2424 for ( my $j = 0 ; $j < $size ; $j++ ) {
2425 my %hashscan;
2426 @hashscan{qw(value cnt)} =
2427 $results[ $i - 1 ]->display_term($j);
2428 push @elements, \%hashscan;
2429 }
2430 }
2431 );
2432 return \@elements;
2433 }
2434 }
2435
2436 =head2 _ZOOM_event_loop
2437
2438 _ZOOM_event_loop(\@zconns, \@results, sub {
2439 my ( $i, $size ) = @_;
2440 ....
2441 } );
2442
2443 Processes a ZOOM event loop and passes control to a closure for
2444 processing the results, and destroying the resultsets.
2445
2446 =cut
2447
2448 sub _ZOOM_event_loop {
2449 my ($zconns, $results, $callback) = @_;
2450 while ( ( my $i = ZOOM::event( $zconns ) ) != 0 ) {
2451 my $ev = $zconns->[ $i - 1 ]->last_event();
2452 if ( $ev == ZOOM::Event::ZEND ) {
2453 next unless $results->[ $i - 1 ];
2454 my $size = $results->[ $i - 1 ]->size();
2455 if ( $size > 0 ) {
2456 $callback->($i, $size);
2457 }
2458 }
2459 }
2460
2461 foreach my $result (@$results) {
2462 $result->destroy();
2463 }
2464 }
2465
2466 =head2 new_record_from_zebra
2467
2468 Given raw data from a searchengine result set, return a MARC::Record object
2469
2470 This helper function is needed to take into account all the involved
2471 system preferences and configuration variables to properly create the
2472 MARC::Record object.
2473
2474 If we are using GRS-1, then the raw data we get from Zebra should be USMARC
2475 data. If we are using DOM, then it has to be MARCXML.
2476
2477 If we are using elasticsearch, it'll already be a MARC::Record and this
2478 function needs a new name.
2479
2480 =cut
2481
2482 sub new_record_from_zebra {
2483
2484 my $server = shift;
2485 my $raw_data = shift;
2486 # Set the default indexing modes
2487 my $search_engine = C4::Context->preference("SearchEngine");
2488 if ($search_engine eq 'Elasticsearch') {
2489 return $raw_data;
2490 }
2491 my $index_mode = ( $server eq 'biblioserver' )
2492 ? C4::Context->config('zebra_bib_index_mode') // 'dom'
2493 : C4::Context->config('zebra_auth_index_mode') // 'dom';
2494
2495 my $marc_record = eval {
2496 if ( $index_mode eq 'dom' ) {
2497 MARC::Record->new_from_xml( $raw_data, 'UTF-8' );
2498 } else {
2499 MARC::Record->new_from_usmarc( $raw_data );
2500 }
2501 };
2502
2503 if ($@) {
2504 return;
2505 } else {
2506 return $marc_record;
2507 }
2508
2509 }
2510
2511 END { } # module clean-up code here (global destructor)
2512
2513 1;
2514 __END__
2515
2516 =head1 AUTHOR
2517
2518 Koha Development Team <http://koha-community.org/>
2519
2520 =cut
Koha ILS