Bug 11030 : add fields 947,969,359 in unimarc_complete french framework
[koha.git] / C4 / Search.pm
blob1b5c6fa52454d59e8c9540ab4a7795ad4c6b6c07
1 package C4::Search;
3 # This file is part of Koha.
5 # Koha is free software; you can redistribute it and/or modify it under the
6 # terms of the GNU General Public License as published by the Free Software
7 # Foundation; either version 2 of the License, or (at your option) any later
8 # version.
10 # Koha is distributed in the hope that it will be useful, but WITHOUT ANY
11 # WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
12 # A PARTICULAR PURPOSE. See the GNU General Public License for more details.
14 # You should have received a copy of the GNU General Public License along with
15 # Koha; if not, write to the Free Software Foundation, Inc., 59 Temple Place,
16 # Suite 330, Boston, MA 02111-1307 USA
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 Lingua::Stem;
25 use C4::Search::PazPar2;
26 use XML::Simple;
27 use C4::Dates qw(format_date);
28 use C4::Members qw(GetHideLostItemsPreference);
29 use C4::XSLT;
30 use C4::Branch;
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 utf8;
40 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
42 # set the version for version checking
43 BEGIN {
44 $VERSION = 3.07.00.049;
45 $DEBUG = ($ENV{DEBUG}) ? 1 : 0;
48 =head1 NAME
50 C4::Search - Functions for searching the Koha catalog.
52 =head1 SYNOPSIS
54 See opac/opac-search.pl or catalogue/search.pl for example of usage
56 =head1 DESCRIPTION
58 This module provides searching functions for Koha's bibliographic databases
60 =head1 FUNCTIONS
62 =cut
64 @ISA = qw(Exporter);
65 @EXPORT = qw(
66 &FindDuplicate
67 &SimpleSearch
68 &searchResults
69 &getRecords
70 &buildQuery
71 &GetDistinctValues
72 &enabled_staff_search_views
73 &PurgeSearchHistory
76 # make all your functions, whether exported or not;
78 =head2 FindDuplicate
80 ($biblionumber,$biblionumber,$title) = FindDuplicate($record);
82 This function attempts to find duplicate records using a hard-coded, fairly simplistic algorithm
84 =cut
86 sub FindDuplicate {
87 my ($record) = @_;
88 my $dbh = C4::Context->dbh;
89 my $result = TransformMarcToKoha( $dbh, $record, '' );
90 my $sth;
91 my $query;
92 my $search;
93 my $type;
94 my ( $biblionumber, $title );
96 # search duplicate on ISBN, easy and fast..
97 # ... normalize first
98 if ( $result->{isbn} ) {
99 $result->{isbn} =~ s/\(.*$//;
100 $result->{isbn} =~ s/\s+$//;
101 $query = "isbn:$result->{isbn}";
103 else {
104 my $QParser;
105 $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser'));
106 my $titleindex;
107 my $authorindex;
108 my $op;
110 if ($QParser) {
111 $titleindex = 'title|exact';
112 $authorindex = 'author|exact';
113 $op = '&&';
114 $QParser->custom_data->{'QueryAutoTruncate'} = C4::Context->preference('QueryAutoTruncate');
115 } else {
116 $titleindex = 'ti,ext';
117 $authorindex = 'au,ext';
118 $op = 'and';
121 $result->{title} =~ s /\\//g;
122 $result->{title} =~ s /\"//g;
123 $result->{title} =~ s /\(//g;
124 $result->{title} =~ s /\)//g;
126 # FIXME: instead of removing operators, could just do
127 # quotes around the value
128 $result->{title} =~ s/(and|or|not)//g;
129 $query = "$titleindex:\"$result->{title}\"";
130 if ( $result->{author} ) {
131 $result->{author} =~ s /\\//g;
132 $result->{author} =~ s /\"//g;
133 $result->{author} =~ s /\(//g;
134 $result->{author} =~ s /\)//g;
136 # remove valid operators
137 $result->{author} =~ s/(and|or|not)//g;
138 $query .= " $op $authorindex:\"$result->{author}\"";
142 my ( $error, $searchresults, undef ) = SimpleSearch($query); # FIXME :: hardcoded !
143 my @results;
144 if (!defined $error) {
145 foreach my $possible_duplicate_record (@{$searchresults}) {
146 my $marcrecord = new_record_from_zebra(
147 'biblioserver',
148 $possible_duplicate_record
151 my $result = TransformMarcToKoha( $dbh, $marcrecord, '' );
153 # FIXME :: why 2 $biblionumber ?
154 if ($result) {
155 push @results, $result->{'biblionumber'};
156 push @results, $result->{'title'};
160 return @results;
163 =head2 SimpleSearch
165 ( $error, $results, $total_hits ) = SimpleSearch( $query, $offset, $max_results, [@servers] );
167 This function provides a simple search API on the bibliographic catalog
169 =over 2
171 =item C<input arg:>
173 * $query can be a simple keyword or a complete CCL query
174 * @servers is optional. Defaults to biblioserver as found in koha-conf.xml
175 * $offset - If present, represents the number of records at the beggining to omit. Defaults to 0
176 * $max_results - if present, determines the maximum number of records to fetch. undef is All. defaults to undef.
179 =item C<Return:>
181 Returns an array consisting of three elements
182 * $error is undefined unless an error is detected
183 * $results is a reference to an array of records.
184 * $total_hits is the number of hits that would have been returned with no limit
186 If an error is returned the two other return elements are undefined. If error itself is undefined
187 the other two elements are always defined
189 =item C<usage in the script:>
191 =back
193 my ( $error, $marcresults, $total_hits ) = SimpleSearch($query);
195 if (defined $error) {
196 $template->param(query_error => $error);
197 warn "error: ".$error;
198 output_html_with_http_headers $input, $cookie, $template->output;
199 exit;
202 my $hits = @{$marcresults};
203 my @results;
205 for my $r ( @{$marcresults} ) {
206 my $marcrecord = MARC::File::USMARC::decode($r);
207 my $biblio = TransformMarcToKoha(C4::Context->dbh,$marcrecord,q{});
209 #build the iarray of hashs for the template.
210 push @results, {
211 title => $biblio->{'title'},
212 subtitle => $biblio->{'subtitle'},
213 biblionumber => $biblio->{'biblionumber'},
214 author => $biblio->{'author'},
215 publishercode => $biblio->{'publishercode'},
216 publicationyear => $biblio->{'publicationyear'},
221 $template->param(result=>\@results);
223 =cut
225 sub SimpleSearch {
226 my ( $query, $offset, $max_results, $servers ) = @_;
228 return ( 'No query entered', undef, undef ) unless $query;
229 # FIXME hardcoded value. See catalog/search.pl & opac-search.pl too.
230 my @servers = defined ( $servers ) ? @$servers : ( 'biblioserver' );
231 my @zoom_queries;
232 my @tmpresults;
233 my @zconns;
234 my $results = [];
235 my $total_hits = 0;
237 my $QParser;
238 $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser') && ! ($query =~ m/\w,\w|\w=\w/));
239 if ($QParser) {
240 $QParser->custom_data->{'QueryAutoTruncate'} = C4::Context->preference('QueryAutoTruncate');
243 # Initialize & Search Zebra
244 for ( my $i = 0 ; $i < @servers ; $i++ ) {
245 eval {
246 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
247 if ($QParser) {
248 $query =~ s/=/:/g;
249 $QParser->parse( $query );
250 $query = $QParser->target_syntax($servers[$i]);
251 $zoom_queries[$i] = new ZOOM::Query::PQF( $query, $zconns[$i]);
252 } else {
253 $query =~ s/:/=/g;
254 $zoom_queries[$i] = new ZOOM::Query::CCL2RPN( $query, $zconns[$i]);
256 $tmpresults[$i] = $zconns[$i]->search( $zoom_queries[$i] );
258 # error handling
259 my $error =
260 $zconns[$i]->errmsg() . " ("
261 . $zconns[$i]->errcode() . ") "
262 . $zconns[$i]->addinfo() . " "
263 . $zconns[$i]->diagset();
265 return ( $error, undef, undef ) if $zconns[$i]->errcode();
267 if ($@) {
269 # caught a ZOOM::Exception
270 my $error =
271 $@->message() . " ("
272 . $@->code() . ") "
273 . $@->addinfo() . " "
274 . $@->diagset();
275 warn $error." for query: $query";
276 return ( $error, undef, undef );
280 _ZOOM_event_loop(
281 \@zconns,
282 \@tmpresults,
283 sub {
284 my ($i, $size) = @_;
285 my $first_record = defined($offset) ? $offset + 1 : 1;
286 my $hits = $tmpresults[ $i - 1 ]->size();
287 $total_hits += $hits;
288 my $last_record = $hits;
289 if ( defined $max_results && $offset + $max_results < $hits ) {
290 $last_record = $offset + $max_results;
293 for my $j ( $first_record .. $last_record ) {
294 my $record = eval {
295 $tmpresults[ $i - 1 ]->record( $j - 1 )->raw()
296 ; # 0 indexed
298 push @{$results}, $record if defined $record;
303 foreach my $zoom_query (@zoom_queries) {
304 $zoom_query->destroy();
307 return ( undef, $results, $total_hits );
310 =head2 getRecords
312 ( undef, $results_hashref, \@facets_loop ) = getRecords (
314 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
315 $results_per_page, $offset, $expanded_facet, $branches,$itemtypes,
316 $query_type, $scan
319 The all singing, all dancing, multi-server, asynchronous, scanning,
320 searching, record nabbing, facet-building
322 See verbse embedded documentation.
324 =cut
326 sub getRecords {
327 my (
328 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
329 $results_per_page, $offset, $expanded_facet, $branches,
330 $itemtypes, $query_type, $scan, $opac
331 ) = @_;
333 my @servers = @$servers_ref;
334 my @sort_by = @$sort_by_ref;
336 # Initialize variables for the ZOOM connection and results object
337 my $zconn;
338 my @zconns;
339 my @results;
340 my $results_hashref = ();
342 # Initialize variables for the faceted results objects
343 my $facets_counter = ();
344 my $facets_info = ();
345 my $facets = getFacets();
346 my $facets_maxrecs = C4::Context->preference('maxRecordsForFacets')||20;
348 my @facets_loop; # stores the ref to array of hashes for template facets loop
350 ### LOOP THROUGH THE SERVERS
351 for ( my $i = 0 ; $i < @servers ; $i++ ) {
352 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
354 # perform the search, create the results objects
355 # if this is a local search, use the $koha-query, if it's a federated one, use the federated-query
356 my $query_to_use = ($servers[$i] =~ /biblioserver/) ? $koha_query : $simple_query;
358 #$query_to_use = $simple_query if $scan;
359 warn $simple_query if ( $scan and $DEBUG );
361 # Check if we've got a query_type defined, if so, use it
362 eval {
363 if ($query_type) {
364 if ($query_type =~ /^ccl/) {
365 $query_to_use =~ s/\:/\=/g; # change : to = last minute (FIXME)
366 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
367 } elsif ($query_type =~ /^cql/) {
368 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CQL($query_to_use, $zconns[$i]));
369 } elsif ($query_type =~ /^pqf/) {
370 $results[$i] = $zconns[$i]->search(new ZOOM::Query::PQF($query_to_use, $zconns[$i]));
371 } else {
372 warn "Unknown query_type '$query_type'. Results undetermined.";
374 } elsif ($scan) {
375 $results[$i] = $zconns[$i]->scan( new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
376 } else {
377 $results[$i] = $zconns[$i]->search(new ZOOM::Query::CCL2RPN($query_to_use, $zconns[$i]));
380 if ($@) {
381 warn "WARNING: query problem with $query_to_use " . $@;
384 # Concatenate the sort_by limits and pass them to the results object
385 # Note: sort will override rank
386 my $sort_by;
387 foreach my $sort (@sort_by) {
388 if ( $sort eq "author_az" || $sort eq "author_asc" ) {
389 $sort_by .= "1=1003 <i ";
391 elsif ( $sort eq "author_za" || $sort eq "author_dsc" ) {
392 $sort_by .= "1=1003 >i ";
394 elsif ( $sort eq "popularity_asc" ) {
395 $sort_by .= "1=9003 <i ";
397 elsif ( $sort eq "popularity_dsc" ) {
398 $sort_by .= "1=9003 >i ";
400 elsif ( $sort eq "call_number_asc" ) {
401 $sort_by .= "1=8007 <i ";
403 elsif ( $sort eq "call_number_dsc" ) {
404 $sort_by .= "1=8007 >i ";
406 elsif ( $sort eq "pubdate_asc" ) {
407 $sort_by .= "1=31 <i ";
409 elsif ( $sort eq "pubdate_dsc" ) {
410 $sort_by .= "1=31 >i ";
412 elsif ( $sort eq "acqdate_asc" ) {
413 $sort_by .= "1=32 <i ";
415 elsif ( $sort eq "acqdate_dsc" ) {
416 $sort_by .= "1=32 >i ";
418 elsif ( $sort eq "title_az" || $sort eq "title_asc" ) {
419 $sort_by .= "1=4 <i ";
421 elsif ( $sort eq "title_za" || $sort eq "title_dsc" ) {
422 $sort_by .= "1=4 >i ";
424 else {
425 warn "Ignoring unrecognized sort '$sort' requested" if $sort_by;
428 if ( $sort_by && !$scan && $results[$i] ) {
429 if ( $results[$i]->sort( "yaz", $sort_by ) < 0 ) {
430 warn "WARNING sort $sort_by failed";
433 } # finished looping through servers
435 # The big moment: asynchronously retrieve results from all servers
436 _ZOOM_event_loop(
437 \@zconns,
438 \@results,
439 sub {
440 my ( $i, $size ) = @_;
441 my $results_hash;
443 # loop through the results
444 $results_hash->{'hits'} = $size;
445 my $times;
446 if ( $offset + $results_per_page <= $size ) {
447 $times = $offset + $results_per_page;
449 else {
450 $times = $size;
453 for ( my $j = $offset ; $j < $times ; $j++ ) {
454 my $records_hash;
455 my $record;
457 ## Check if it's an index scan
458 if ($scan) {
459 my ( $term, $occ ) = $results[ $i - 1 ]->display_term($j);
461 # here we create a minimal MARC record and hand it off to the
462 # template just like a normal result ... perhaps not ideal, but
463 # it works for now
464 my $tmprecord = MARC::Record->new();
465 $tmprecord->encoding('UTF-8');
466 my $tmptitle;
467 my $tmpauthor;
469 # the minimal record in author/title (depending on MARC flavour)
470 if ( C4::Context->preference("marcflavour") eq
471 "UNIMARC" )
473 $tmptitle = MARC::Field->new(
474 '200', ' ', ' ',
475 a => $term,
476 f => $occ
478 $tmprecord->append_fields($tmptitle);
480 else {
481 $tmptitle =
482 MARC::Field->new( '245', ' ', ' ', a => $term, );
483 $tmpauthor =
484 MARC::Field->new( '100', ' ', ' ', a => $occ, );
485 $tmprecord->append_fields($tmptitle);
486 $tmprecord->append_fields($tmpauthor);
488 $results_hash->{'RECORDS'}[$j] =
489 $tmprecord->as_usmarc();
492 # not an index scan
493 else {
494 $record = $results[ $i - 1 ]->record($j)->raw();
495 # warn "RECORD $j:".$record;
496 $results_hash->{'RECORDS'}[$j] = $record;
500 $results_hashref->{ $servers[ $i - 1 ] } = $results_hash;
502 # Fill the facets while we're looping, but only for the biblioserver and not for a scan
503 if ( !$scan && $servers[ $i - 1 ] =~ /biblioserver/ ) {
505 my $jmax =
506 $size > $facets_maxrecs ? $facets_maxrecs : $size;
507 for my $facet (@$facets) {
508 for ( my $j = 0 ; $j < $jmax ; $j++ ) {
510 my $marc_record = new_record_from_zebra (
511 'biblioserver',
512 $results[ $i - 1 ]->record($j)->raw()
515 if ( ! defined $marc_record ) {
516 warn "ERROR DECODING RECORD - $@: " .
517 $results[ $i - 1 ]->record($j)->raw();
518 next;
521 my @used_datas = ();
523 foreach my $tag ( @{ $facet->{tags} } ) {
525 # avoid first line
526 my $tag_num = substr( $tag, 0, 3 );
527 my $subfield_letters = substr( $tag, 3 );
528 # Removed when as_string fixed
529 my @subfields = $subfield_letters =~ /./sg;
531 my @fields = $marc_record->field($tag_num);
532 foreach my $field (@fields) {
533 my $data = $field->as_string( $subfield_letters, $facet->{sep} );
535 unless ( grep { /^$data$/ } @used_datas ) {
536 push @used_datas, $data;
537 $facets_counter->{ $facet->{idx} }->{$data}++;
539 } # fields
540 } # field codes
541 } # records
542 $facets_info->{ $facet->{idx} }->{label_value} =
543 $facet->{label};
544 $facets_info->{ $facet->{idx} }->{expanded} =
545 $facet->{expanded};
546 } # facets
549 # warn "connection ", $i-1, ": $size hits";
550 # warn $results[$i-1]->record(0)->render() if $size > 0;
552 # BUILD FACETS
553 if ( $servers[ $i - 1 ] =~ /biblioserver/ ) {
554 for my $link_value (
555 sort { $facets_counter->{$b} <=> $facets_counter->{$a} }
556 keys %$facets_counter
559 my $expandable;
560 my $number_of_facets;
561 my @this_facets_array;
562 for my $one_facet (
563 sort {
564 $facets_counter->{$link_value}
565 ->{$b} <=> $facets_counter->{$link_value}
566 ->{$a}
567 } keys %{ $facets_counter->{$link_value} }
570 $number_of_facets++;
571 if ( ( $number_of_facets <= 5 )
572 || ( $expanded_facet eq $link_value )
573 || ( $facets_info->{$link_value}->{'expanded'} )
577 # Sanitize the link value : parenthesis, question and exclamation mark will cause errors with CCL
578 my $facet_link_value = $one_facet;
579 $facet_link_value =~ s/[()!?¡¿؟]/ /g;
581 # fix the length that will display in the label,
582 my $facet_label_value = $one_facet;
583 my $facet_max_length = C4::Context->preference(
584 'FacetLabelTruncationLength')
585 || 20;
586 $facet_label_value =
587 substr( $one_facet, 0, $facet_max_length )
588 . "..."
589 if length($facet_label_value) >
590 $facet_max_length;
592 # if it's a branch, label by the name, not the code,
593 if ( $link_value =~ /branch/ ) {
594 if ( defined $branches
595 && ref($branches) eq "HASH"
596 && defined $branches->{$one_facet}
597 && ref( $branches->{$one_facet} ) eq
598 "HASH" )
600 $facet_label_value =
601 $branches->{$one_facet}
602 ->{'branchname'};
604 else {
605 $facet_label_value = "*";
609 # if it's a itemtype, label by the name, not the code,
610 if ( $link_value =~ /itype/ ) {
611 if ( defined $itemtypes
612 && ref($itemtypes) eq "HASH"
613 && defined $itemtypes->{$one_facet}
614 && ref( $itemtypes->{$one_facet} ) eq
615 "HASH" )
617 $facet_label_value =
618 $itemtypes->{$one_facet}
619 ->{'description'};
623 # also, if it's a location code, use the name instead of the code
624 if ( $link_value =~ /location/ ) {
625 $facet_label_value =
626 GetKohaAuthorisedValueLib( 'LOC',
627 $one_facet, $opac );
630 # but we're down with the whole label being in the link's title.
631 push @this_facets_array,
633 facet_count =>
634 $facets_counter->{$link_value}
635 ->{$one_facet},
636 facet_label_value => $facet_label_value,
637 facet_title_value => $one_facet,
638 facet_link_value => $facet_link_value,
639 type_link_value => $link_value,
641 if ($facet_label_value);
645 # handle expanded option
646 unless ( $facets_info->{$link_value}->{'expanded'} ) {
647 $expandable = 1
648 if ( ( $number_of_facets > 5 )
649 && ( $expanded_facet ne $link_value ) );
651 push @facets_loop,
653 type_link_value => $link_value,
654 type_id => $link_value . "_id",
655 "type_label_"
656 . $facets_info->{$link_value}->{'label_value'} =>
658 facets => \@this_facets_array,
659 expandable => $expandable,
660 expand => $link_value,
662 unless (
664 $facets_info->{$link_value}->{'label_value'} =~
665 /Libraries/
667 and ( C4::Context->preference('singleBranchMode') )
673 return ( undef, $results_hashref, \@facets_loop );
676 sub pazGetRecords {
677 my (
678 $koha_query, $simple_query, $sort_by_ref, $servers_ref,
679 $results_per_page, $offset, $expanded_facet, $branches,
680 $query_type, $scan
681 ) = @_;
683 my $paz = C4::Search::PazPar2->new(C4::Context->config('pazpar2url'));
684 $paz->init();
685 $paz->search($simple_query);
686 sleep 1; # FIXME: WHY?
688 # do results
689 my $results_hashref = {};
690 my $stats = XMLin($paz->stat);
691 my $results = XMLin($paz->show($offset, $results_per_page, 'work-title:1'), forcearray => 1);
693 # for a grouped search result, the number of hits
694 # is the number of groups returned; 'bib_hits' will have
695 # the total number of bibs.
696 $results_hashref->{'biblioserver'}->{'hits'} = $results->{'merged'}->[0];
697 $results_hashref->{'biblioserver'}->{'bib_hits'} = $stats->{'hits'};
699 HIT: foreach my $hit (@{ $results->{'hit'} }) {
700 my $recid = $hit->{recid}->[0];
702 my $work_title = $hit->{'md-work-title'}->[0];
703 my $work_author;
704 if (exists $hit->{'md-work-author'}) {
705 $work_author = $hit->{'md-work-author'}->[0];
707 my $group_label = (defined $work_author) ? "$work_title / $work_author" : $work_title;
709 my $result_group = {};
710 $result_group->{'group_label'} = $group_label;
711 $result_group->{'group_merge_key'} = $recid;
713 my $count = 1;
714 if (exists $hit->{count}) {
715 $count = $hit->{count}->[0];
717 $result_group->{'group_count'} = $count;
719 for (my $i = 0; $i < $count; $i++) {
720 # FIXME -- may need to worry about diacritics here
721 my $rec = $paz->record($recid, $i);
722 push @{ $result_group->{'RECORDS'} }, $rec;
725 push @{ $results_hashref->{'biblioserver'}->{'GROUPS'} }, $result_group;
728 # pass through facets
729 my $termlist_xml = $paz->termlist('author,subject');
730 my $terms = XMLin($termlist_xml, forcearray => 1);
731 my @facets_loop = ();
732 #die Dumper($results);
733 # foreach my $list (sort keys %{ $terms->{'list'} }) {
734 # my @facets = ();
735 # foreach my $facet (sort @{ $terms->{'list'}->{$list}->{'term'} } ) {
736 # push @facets, {
737 # facet_label_value => $facet->{'name'}->[0],
738 # };
740 # push @facets_loop, ( {
741 # type_label => $list,
742 # facets => \@facets,
743 # } );
746 return ( undef, $results_hashref, \@facets_loop );
749 # STOPWORDS
750 sub _remove_stopwords {
751 my ( $operand, $index ) = @_;
752 my @stopwords_removed;
754 # phrase and exact-qualified indexes shouldn't have stopwords removed
755 if ( $index !~ m/,(phr|ext)/ ) {
757 # remove stopwords from operand : parse all stopwords & remove them (case insensitive)
758 # we use IsAlpha unicode definition, to deal correctly with diacritics.
759 # otherwise, a French word like "leçon" woudl be split into "le" "çon", "le"
760 # is a stopword, we'd get "çon" and wouldn't find anything...
762 foreach ( keys %{ C4::Context->stopwords } ) {
763 next if ( $_ =~ /(and|or|not)/ ); # don't remove operators
764 if ( my ($matched) = ($operand =~
765 /([^\X\p{isAlnum}]\Q$_\E[^\X\p{isAlnum}]|[^\X\p{isAlnum}]\Q$_\E$|^\Q$_\E[^\X\p{isAlnum}])/gi))
767 $operand =~ s/\Q$matched\E/ /gi;
768 push @stopwords_removed, $_;
772 return ( $operand, \@stopwords_removed );
775 # TRUNCATION
776 sub _detect_truncation {
777 my ( $operand, $index ) = @_;
778 my ( @nontruncated, @righttruncated, @lefttruncated, @rightlefttruncated,
779 @regexpr );
780 $operand =~ s/^ //g;
781 my @wordlist = split( /\s/, $operand );
782 foreach my $word (@wordlist) {
783 if ( $word =~ s/^\*([^\*]+)\*$/$1/ ) {
784 push @rightlefttruncated, $word;
786 elsif ( $word =~ s/^\*([^\*]+)$/$1/ ) {
787 push @lefttruncated, $word;
789 elsif ( $word =~ s/^([^\*]+)\*$/$1/ ) {
790 push @righttruncated, $word;
792 elsif ( index( $word, "*" ) < 0 ) {
793 push @nontruncated, $word;
795 else {
796 push @regexpr, $word;
799 return (
800 \@nontruncated, \@righttruncated, \@lefttruncated,
801 \@rightlefttruncated, \@regexpr
805 # STEMMING
806 sub _build_stemmed_operand {
807 my ($operand,$lang) = @_;
808 require Lingua::Stem::Snowball ;
809 my $stemmed_operand=q{};
811 # If operand contains a digit, it is almost certainly an identifier, and should
812 # not be stemmed. This is particularly relevant for ISBNs and ISSNs, which
813 # can contain the letter "X" - for example, _build_stemmend_operand would reduce
814 # "014100018X" to "x ", which for a MARC21 database would bring up irrelevant
815 # results (e.g., "23 x 29 cm." from the 300$c). Bug 2098.
816 return $operand if $operand =~ /\d/;
818 # FIXME: the locale should be set based on the user's language and/or search choice
819 #warn "$lang";
820 # Make sure we only use the first two letters from the language code
821 $lang = lc(substr($lang, 0, 2));
822 # The language codes for the two variants of Norwegian will now be "nb" and "nn",
823 # none of which Lingua::Stem::Snowball can use, so we need to "translate" them
824 if ($lang eq 'nb' || $lang eq 'nn') {
825 $lang = 'no';
827 my $stemmer = Lingua::Stem::Snowball->new( lang => $lang,
828 encoding => "UTF-8" );
830 my @words = split( / /, $operand );
831 my @stems = $stemmer->stem(\@words);
832 for my $stem (@stems) {
833 $stemmed_operand .= "$stem";
834 $stemmed_operand .= "?"
835 unless ( $stem =~ /(and$|or$|not$)/ ) || ( length($stem) < 3 );
836 $stemmed_operand .= " ";
838 warn "STEMMED OPERAND: $stemmed_operand" if $DEBUG;
839 return $stemmed_operand;
842 # FIELD WEIGHTING
843 sub _build_weighted_query {
845 # FIELD WEIGHTING - This is largely experimental stuff. What I'm committing works
846 # pretty well but could work much better if we had a smarter query parser
847 my ( $operand, $stemmed_operand, $index ) = @_;
848 my $stemming = C4::Context->preference("QueryStemming") || 0;
849 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
850 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
851 $operand =~ s/"/ /g; # Bug 7518: searches with quotation marks don't work
853 my $weighted_query .= "(rk=("; # Specifies that we're applying rank
855 # Keyword, or, no index specified
856 if ( ( $index eq 'kw' ) || ( !$index ) ) {
857 $weighted_query .=
858 "Title-cover,ext,r1=\"$operand\""; # exact title-cover
859 $weighted_query .= " or ti,ext,r2=\"$operand\""; # exact title
860 $weighted_query .= " or Title-cover,phr,r3=\"$operand\""; # phrase title
861 $weighted_query .= " or ti,wrdl,r4=\"$operand\""; # words in title
862 #$weighted_query .= " or any,ext,r4=$operand"; # exact any
863 #$weighted_query .=" or kw,wrdl,r5=\"$operand\""; # word list any
864 $weighted_query .= " or wrdl,fuzzy,r8=\"$operand\""
865 if $fuzzy_enabled; # add fuzzy, word list
866 $weighted_query .= " or wrdl,right-Truncation,r9=\"$stemmed_operand\""
867 if ( $stemming and $stemmed_operand )
868 ; # add stemming, right truncation
869 $weighted_query .= " or wrdl,r9=\"$operand\"";
871 # embedded sorting: 0 a-z; 1 z-a
872 # $weighted_query .= ") or (sort1,aut=1";
875 # Barcode searches should skip this process
876 elsif ( $index eq 'bc' ) {
877 $weighted_query .= "bc=\"$operand\"";
880 # Authority-number searches should skip this process
881 elsif ( $index eq 'an' ) {
882 $weighted_query .= "an=\"$operand\"";
885 # If the index already has more than one qualifier, wrap the operand
886 # in quotes and pass it back (assumption is that the user knows what they
887 # are doing and won't appreciate us mucking up their query
888 elsif ( $index =~ ',' ) {
889 $weighted_query .= " $index=\"$operand\"";
892 #TODO: build better cases based on specific search indexes
893 else {
894 $weighted_query .= " $index,ext,r1=\"$operand\""; # exact index
895 #$weighted_query .= " or (title-sort-az=0 or $index,startswithnt,st-word,r3=$operand #)";
896 $weighted_query .= " or $index,phr,r3=\"$operand\""; # phrase index
897 $weighted_query .= " or $index,wrdl,r6=\"$operand\""; # word list index
898 $weighted_query .= " or $index,wrdl,fuzzy,r8=\"$operand\""
899 if $fuzzy_enabled; # add fuzzy, word list
900 $weighted_query .= " or $index,wrdl,rt,r9=\"$stemmed_operand\""
901 if ( $stemming and $stemmed_operand ); # add stemming, right truncation
904 $weighted_query .= "))"; # close rank specification
905 return $weighted_query;
908 =head2 getIndexes
910 Return an array with available indexes.
912 =cut
914 sub getIndexes{
915 my @indexes = (
916 # biblio indexes
917 'ab',
918 'Abstract',
919 'acqdate',
920 'allrecords',
921 'an',
922 'Any',
923 'at',
924 'au',
925 'aub',
926 'aud',
927 'audience',
928 'auo',
929 'aut',
930 'Author',
931 'Author-in-order ',
932 'Author-personal-bibliography',
933 'Authority-Number',
934 'authtype',
935 'bc',
936 'Bib-level',
937 'biblionumber',
938 'bio',
939 'biography',
940 'callnum',
941 'cfn',
942 'Chronological-subdivision',
943 'cn-bib-source',
944 'cn-bib-sort',
945 'cn-class',
946 'cn-item',
947 'cn-prefix',
948 'cn-suffix',
949 'cpn',
950 'Code-institution',
951 'Conference-name',
952 'Conference-name-heading',
953 'Conference-name-see',
954 'Conference-name-seealso',
955 'Content-type',
956 'Control-number',
957 'copydate',
958 'Corporate-name',
959 'Corporate-name-heading',
960 'Corporate-name-see',
961 'Corporate-name-seealso',
962 'Country-publication',
963 'ctype',
964 'curriculum',
965 'date-entered-on-file',
966 'Date-of-acquisition',
967 'Date-of-publication',
968 'Dewey-classification',
969 'Dissertation-information',
970 'EAN',
971 'extent',
972 'fic',
973 'fiction',
974 'Form-subdivision',
975 'format',
976 'Geographic-subdivision',
977 'he',
978 'Heading',
979 'Heading-use-main-or-added-entry',
980 'Heading-use-series-added-entry ',
981 'Heading-use-subject-added-entry',
982 'Host-item',
983 'id-other',
984 'Illustration-code',
985 'Index-term-genre',
986 'Index-term-uncontrolled',
987 'ISBN',
988 'isbn',
989 'ISSN',
990 'issn',
991 'itemtype',
992 'kw',
993 'Koha-Auth-Number',
994 'l-format',
995 'language',
996 'language-original',
997 'lc-card',
998 'LC-card-number',
999 'lcn',
1000 'lex',
1001 'llength',
1002 'ln',
1003 'ln-audio',
1004 'ln-subtitle',
1005 'Local-classification',
1006 'Local-number',
1007 'Match-heading',
1008 'Match-heading-see-from',
1009 'Material-type',
1010 'mc-itemtype',
1011 'mc-rtype',
1012 'mus',
1013 'name',
1014 'Music-number',
1015 'Name-geographic',
1016 'Name-geographic-heading',
1017 'Name-geographic-see',
1018 'Name-geographic-seealso',
1019 'nb',
1020 'Note',
1021 'notes',
1022 'ns',
1023 'nt',
1024 'pb',
1025 'Personal-name',
1026 'Personal-name-heading',
1027 'Personal-name-see',
1028 'Personal-name-seealso',
1029 'pl',
1030 'Place-publication',
1031 'pn',
1032 'popularity',
1033 'pubdate',
1034 'Publisher',
1035 'Record-control-number',
1036 'rcn',
1037 'Record-type',
1038 'rtype',
1039 'se',
1040 'See',
1041 'See-also',
1042 'sn',
1043 'Stock-number',
1044 'su',
1045 'Subject',
1046 'Subject-heading-thesaurus',
1047 'Subject-name-personal',
1048 'Subject-subdivision',
1049 'Summary',
1050 'Suppress',
1051 'su-geo',
1052 'su-na',
1053 'su-to',
1054 'su-ut',
1055 'ut',
1056 'Term-genre-form',
1057 'Term-genre-form-heading',
1058 'Term-genre-form-see',
1059 'Term-genre-form-seealso',
1060 'ti',
1061 'Title',
1062 'Title-cover',
1063 'Title-series',
1064 'Title-uniform',
1065 'Title-uniform-heading',
1066 'Title-uniform-see',
1067 'Title-uniform-seealso',
1068 'totalissues',
1069 'yr',
1071 # items indexes
1072 'acqsource',
1073 'barcode',
1074 'bc',
1075 'branch',
1076 'ccode',
1077 'classification-source',
1078 'cn-sort',
1079 'coded-location-qualifier',
1080 'copynumber',
1081 'damaged',
1082 'datelastborrowed',
1083 'datelastseen',
1084 'holdingbranch',
1085 'homebranch',
1086 'issues',
1087 'item',
1088 'itemnumber',
1089 'itype',
1090 'Local-classification',
1091 'location',
1092 'lost',
1093 'materials-specified',
1094 'mc-ccode',
1095 'mc-itype',
1096 'mc-loc',
1097 'notforloan',
1098 'Number-local-acquisition',
1099 'onloan',
1100 'price',
1101 'renewals',
1102 'replacementprice',
1103 'replacementpricedate',
1104 'reserves',
1105 'restricted',
1106 'stack',
1107 'stocknumber',
1108 'inv',
1109 'uri',
1110 'withdrawn',
1112 # subject related
1115 return \@indexes;
1118 =head2 _handle_exploding_index
1120 my $query = _handle_exploding_index($index, $term)
1122 Callback routine to generate the search for "exploding" indexes (i.e.
1123 those indexes which are turned into multiple or-connected searches based
1124 on authority data).
1126 =cut
1128 sub _handle_exploding_index {
1129 my ($QParser, $filter, $params, $negate, $server) = @_;
1130 my $index = $filter;
1131 my $term = join(' ', @$params);
1133 return unless ($index =~ m/(su-br|su-na|su-rl)/ && $term);
1135 my $marcflavour = C4::Context->preference('marcflavour');
1137 my $codesubfield = $marcflavour eq 'UNIMARC' ? '5' : 'w';
1138 my $wantedcodes = '';
1139 my @subqueries = ( "\@attr 1=Subject \@attr 4=1 \"$term\"");
1140 my ($error, $results, $total_hits) = SimpleSearch( "he:$term", undef, undef, [ "authorityserver" ] );
1141 foreach my $auth (@$results) {
1142 my $record = MARC::Record->new_from_usmarc($auth);
1143 my @references = $record->field('5..');
1144 if (@references) {
1145 if ($index eq 'su-br') {
1146 $wantedcodes = 'g';
1147 } elsif ($index eq 'su-na') {
1148 $wantedcodes = 'h';
1149 } elsif ($index eq 'su-rl') {
1150 $wantedcodes = '';
1152 foreach my $reference (@references) {
1153 my $codes = $reference->subfield($codesubfield);
1154 push @subqueries, '@attr 1=Subject @attr 4=1 "' . $reference->as_string('abcdefghijlmnopqrstuvxyz') . '"' if (($codes && $codes eq $wantedcodes) || !$wantedcodes);
1158 my $query = ' @or ' x (scalar(@subqueries) - 1) . join(' ', @subqueries);
1159 return $query;
1162 =head2 parseQuery
1164 ( $operators, $operands, $indexes, $limits,
1165 $sort_by, $scan, $lang ) =
1166 buildQuery ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1168 Shim function to ease the transition from buildQuery to a new QueryParser.
1169 This function is called at the beginning of buildQuery, and modifies
1170 buildQuery's input. If it can handle the input, it returns a query that
1171 buildQuery will not try to parse.
1172 =cut
1174 sub parseQuery {
1175 my ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang) = @_;
1177 my @operators = $operators ? @$operators : ();
1178 my @indexes = $indexes ? @$indexes : ();
1179 my @operands = $operands ? @$operands : ();
1180 my @limits = $limits ? @$limits : ();
1181 my @sort_by = $sort_by ? @$sort_by : ();
1183 my $query = $operands[0];
1184 my $index;
1185 my $term;
1186 my $query_desc;
1188 my $QParser;
1189 $QParser = C4::Context->queryparser if (C4::Context->preference('UseQueryParser') || $query =~ s/^qp=//);
1190 undef $QParser if ($query =~ m/^(ccl=|pqf=|cql=)/ || grep (/\w,\w|\w=\w/, @operands, @indexes) );
1191 undef $QParser if (scalar @limits > 0);
1193 if ($QParser)
1195 $QParser->custom_data->{'QueryAutoTruncate'} = C4::Context->preference('QueryAutoTruncate');
1196 $query = '';
1197 for ( my $ii = 0 ; $ii <= @operands ; $ii++ ) {
1198 next unless $operands[$ii];
1199 $query .= $operators[ $ii - 1 ] eq 'or' ? ' || ' : ' && '
1200 if ($query);
1201 if ( $operands[$ii] =~ /^[^"]\W*[-|_\w]*:\w.*[^"]$/ ) {
1202 $query .= $operands[$ii];
1204 elsif ( $indexes[$ii] =~ m/su-/ ) {
1205 $query .= $indexes[$ii] . '(' . $operands[$ii] . ')';
1207 else {
1208 $query .=
1209 ( $indexes[$ii] ? "$indexes[$ii]:" : '' ) . $operands[$ii];
1212 foreach my $limit (@limits) {
1214 if ( scalar(@sort_by) > 0 ) {
1215 my $modifier_re =
1216 '#(' . join( '|', @{ $QParser->modifiers } ) . ')';
1217 $query =~ s/$modifier_re//g;
1218 foreach my $modifier (@sort_by) {
1219 $query .= " #$modifier";
1223 $query_desc = $query;
1224 $query_desc =~ s/\s+/ /g;
1225 if ( C4::Context->preference("QueryWeightFields") ) {
1227 $QParser->add_bib1_filter_map( 'su-br' => 'biblioserver' =>
1228 { 'target_syntax_callback' => \&_handle_exploding_index } );
1229 $QParser->add_bib1_filter_map( 'su-na' => 'biblioserver' =>
1230 { 'target_syntax_callback' => \&_handle_exploding_index } );
1231 $QParser->add_bib1_filter_map( 'su-rl' => 'biblioserver' =>
1232 { 'target_syntax_callback' => \&_handle_exploding_index } );
1233 $QParser->parse($query);
1234 $operands[0] = "pqf=" . $QParser->target_syntax('biblioserver');
1236 else {
1237 require Koha::QueryParser::Driver::PQF;
1238 my $modifier_re = '#(' . join( '|', @{Koha::QueryParser::Driver::PQF->modifiers}) . ')';
1239 s/$modifier_re//g for @operands;
1242 return ( $operators, \@operands, $indexes, $limits, $sort_by, $scan, $lang, $query_desc);
1245 =head2 buildQuery
1247 ( $error, $query,
1248 $simple_query, $query_cgi,
1249 $query_desc, $limit,
1250 $limit_cgi, $limit_desc,
1251 $stopwords_removed, $query_type ) = buildQuery ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1253 Build queries and limits in CCL, CGI, Human,
1254 handle truncation, stemming, field weighting, stopwords, fuzziness, etc.
1256 See verbose embedded documentation.
1259 =cut
1261 sub buildQuery {
1262 my ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang) = @_;
1264 warn "---------\nEnter buildQuery\n---------" if $DEBUG;
1266 my $query_desc;
1267 ( $operators, $operands, $indexes, $limits, $sort_by, $scan, $lang, $query_desc) = parseQuery($operators, $operands, $indexes, $limits, $sort_by, $scan, $lang);
1269 # dereference
1270 my @operators = $operators ? @$operators : ();
1271 my @indexes = $indexes ? @$indexes : ();
1272 my @operands = $operands ? @$operands : ();
1273 my @limits = $limits ? @$limits : ();
1274 my @sort_by = $sort_by ? @$sort_by : ();
1276 my $stemming = C4::Context->preference("QueryStemming") || 0;
1277 my $auto_truncation = C4::Context->preference("QueryAutoTruncate") || 0;
1278 my $weight_fields = C4::Context->preference("QueryWeightFields") || 0;
1279 my $fuzzy_enabled = C4::Context->preference("QueryFuzzy") || 0;
1280 my $remove_stopwords = C4::Context->preference("QueryRemoveStopwords") || 0;
1282 my $query = $operands[0];
1283 my $simple_query = $operands[0];
1285 # initialize the variables we're passing back
1286 my $query_cgi;
1287 my $query_type;
1289 my $limit;
1290 my $limit_cgi;
1291 my $limit_desc;
1293 my $stopwords_removed; # flag to determine if stopwords have been removed
1295 my $cclq = 0;
1296 my $cclindexes = getIndexes();
1297 if ( $query !~ /\s*(ccl=|pqf=|cql=)/ ) {
1298 while ( !$cclq && $query =~ /(?:^|\W)([\w-]+)(,[\w-]+)*[:=]/g ) {
1299 my $dx = lc($1);
1300 $cclq = grep { lc($_) eq $dx } @$cclindexes;
1302 $query = "ccl=$query" if $cclq;
1305 # for handling ccl, cql, pqf queries in diagnostic mode, skip the rest of the steps
1306 # DIAGNOSTIC ONLY!!
1307 if ( $query =~ /^ccl=/ ) {
1308 my $q=$';
1309 # This is needed otherwise ccl= and &limit won't work together, and
1310 # this happens when selecting a subject on the opac-detail page
1311 @limits = grep {!/^$/} @limits;
1312 if ( @limits ) {
1313 $q .= ' and '.join(' and ', @limits);
1315 return ( undef, $q, $q, "q=ccl=".uri_escape($q), $q, '', '', '', '', 'ccl' );
1317 if ( $query =~ /^cql=/ ) {
1318 return ( undef, $', $', "q=cql=".uri_escape($'), $', '', '', '', '', 'cql' );
1320 if ( $query =~ /^pqf=/ ) {
1321 if ($query_desc) {
1322 $query_cgi = "q=".uri_escape($query_desc);
1323 } else {
1324 $query_desc = $';
1325 $query_cgi = "q=pqf=".uri_escape($');
1327 return ( undef, $', $', $query_cgi, $query_desc, '', '', '', '', 'pqf' );
1330 # pass nested queries directly
1331 # FIXME: need better handling of some of these variables in this case
1332 # Nested queries aren't handled well and this implementation is flawed and causes users to be
1333 # unable to search for anything containing () commenting out, will be rewritten for 3.4.0
1334 # if ( $query =~ /(\(|\))/ ) {
1335 # return (
1336 # undef, $query, $simple_query, $query_cgi,
1337 # $query, $limit, $limit_cgi, $limit_desc,
1338 # $stopwords_removed, 'ccl'
1339 # );
1342 # Form-based queries are non-nested and fixed depth, so we can easily modify the incoming
1343 # query operands and indexes and add stemming, truncation, field weighting, etc.
1344 # Once we do so, we'll end up with a value in $query, just like if we had an
1345 # incoming $query from the user
1346 else {
1347 $query = ""
1348 ; # clear it out so we can populate properly with field-weighted, stemmed, etc. query
1349 my $previous_operand
1350 ; # a flag used to keep track if there was a previous query
1351 # if there was, we can apply the current operator
1352 # for every operand
1353 for ( my $i = 0 ; $i <= @operands ; $i++ ) {
1355 # COMBINE OPERANDS, INDEXES AND OPERATORS
1356 if ( $operands[$i] ) {
1357 $operands[$i]=~s/^\s+//;
1359 # A flag to determine whether or not to add the index to the query
1360 my $indexes_set;
1362 # If the user is sophisticated enough to specify an index, turn off field weighting, stemming, and stopword handling
1363 if ( $operands[$i] =~ /\w(:|=)/ || $scan ) {
1364 $weight_fields = 0;
1365 $stemming = 0;
1366 $remove_stopwords = 0;
1367 } else {
1368 $operands[$i] =~ s/\?/{?}/g; # need to escape question marks
1370 my $operand = $operands[$i];
1371 my $index = $indexes[$i];
1373 # Add index-specific attributes
1374 # Date of Publication
1375 if ( $index eq 'yr' ) {
1376 $index .= ",st-numeric";
1377 $indexes_set++;
1378 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = $remove_stopwords = 0;
1381 # Date of Acquisition
1382 elsif ( $index eq 'acqdate' ) {
1383 $index .= ",st-date-normalized";
1384 $indexes_set++;
1385 $stemming = $auto_truncation = $weight_fields = $fuzzy_enabled = $remove_stopwords = 0;
1387 # ISBN,ISSN,Standard Number, don't need special treatment
1388 elsif ( $index eq 'nb' || $index eq 'ns' ) {
1390 $stemming, $auto_truncation,
1391 $weight_fields, $fuzzy_enabled,
1392 $remove_stopwords
1393 ) = ( 0, 0, 0, 0, 0 );
1397 if(not $index){
1398 $index = 'kw';
1401 # Set default structure attribute (word list)
1402 my $struct_attr = q{};
1403 unless ( $indexes_set || !$index || $index =~ /,(st-|phr|ext|wrdl)/ || $index =~ /^(nb|ns)$/ ) {
1404 $struct_attr = ",wrdl";
1407 # Some helpful index variants
1408 my $index_plus = $index . $struct_attr . ':';
1409 my $index_plus_comma = $index . $struct_attr . ',';
1411 # Remove Stopwords
1412 if ($remove_stopwords) {
1413 ( $operand, $stopwords_removed ) =
1414 _remove_stopwords( $operand, $index );
1415 warn "OPERAND w/out STOPWORDS: >$operand<" if $DEBUG;
1416 warn "REMOVED STOPWORDS: @$stopwords_removed"
1417 if ( $stopwords_removed && $DEBUG );
1420 if ($auto_truncation){
1421 unless ( $index =~ /,(st-|phr|ext)/ ) {
1422 #FIXME only valid with LTR scripts
1423 $operand=join(" ",map{
1424 (index($_,"*")>0?"$_":"$_*")
1425 }split (/\s+/,$operand));
1426 warn $operand if $DEBUG;
1430 # Detect Truncation
1431 my $truncated_operand;
1432 my( $nontruncated, $righttruncated, $lefttruncated,
1433 $rightlefttruncated, $regexpr
1434 ) = _detect_truncation( $operand, $index );
1435 warn
1436 "TRUNCATION: NON:>@$nontruncated< RIGHT:>@$righttruncated< LEFT:>@$lefttruncated< RIGHTLEFT:>@$rightlefttruncated< REGEX:>@$regexpr<"
1437 if $DEBUG;
1439 # Apply Truncation
1440 if (
1441 scalar(@$righttruncated) + scalar(@$lefttruncated) +
1442 scalar(@$rightlefttruncated) > 0 )
1445 # Don't field weight or add the index to the query, we do it here
1446 $indexes_set = 1;
1447 undef $weight_fields;
1448 my $previous_truncation_operand;
1449 if (scalar @$nontruncated) {
1450 $truncated_operand .= "$index_plus @$nontruncated ";
1451 $previous_truncation_operand = 1;
1453 if (scalar @$righttruncated) {
1454 $truncated_operand .= "and " if $previous_truncation_operand;
1455 $truncated_operand .= $index_plus_comma . "rtrn:@$righttruncated ";
1456 $previous_truncation_operand = 1;
1458 if (scalar @$lefttruncated) {
1459 $truncated_operand .= "and " if $previous_truncation_operand;
1460 $truncated_operand .= $index_plus_comma . "ltrn:@$lefttruncated ";
1461 $previous_truncation_operand = 1;
1463 if (scalar @$rightlefttruncated) {
1464 $truncated_operand .= "and " if $previous_truncation_operand;
1465 $truncated_operand .= $index_plus_comma . "rltrn:@$rightlefttruncated ";
1466 $previous_truncation_operand = 1;
1469 $operand = $truncated_operand if $truncated_operand;
1470 warn "TRUNCATED OPERAND: >$truncated_operand<" if $DEBUG;
1472 # Handle Stemming
1473 my $stemmed_operand;
1474 $stemmed_operand = _build_stemmed_operand($operand, $lang)
1475 if $stemming;
1477 warn "STEMMED OPERAND: >$stemmed_operand<" if $DEBUG;
1479 # Handle Field Weighting
1480 my $weighted_operand;
1481 if ($weight_fields) {
1482 $weighted_operand = _build_weighted_query( $operand, $stemmed_operand, $index );
1483 $operand = $weighted_operand;
1484 $indexes_set = 1;
1487 warn "FIELD WEIGHTED OPERAND: >$weighted_operand<" if $DEBUG;
1489 ($query,$query_cgi,$query_desc,$previous_operand) = _build_initial_query({
1490 query => $query,
1491 query_cgi => $query_cgi,
1492 query_desc => $query_desc,
1493 operator => ($operators[ $i - 1 ]) ? $operators[ $i - 1 ] : '',
1494 parsed_operand => $operand,
1495 original_operand => ($operands[$i]) ? $operands[$i] : '',
1496 index => $index,
1497 index_plus => $index_plus,
1498 indexes_set => $indexes_set,
1499 previous_operand => $previous_operand,
1502 } #/if $operands
1503 } # /for
1505 warn "QUERY BEFORE LIMITS: >$query<" if $DEBUG;
1507 # add limits
1508 my %group_OR_limits;
1509 my $availability_limit;
1510 foreach my $this_limit (@limits) {
1511 next unless $this_limit;
1512 if ( $this_limit =~ /available/ ) {
1514 ## 'available' is defined as (items.onloan is NULL) and (items.itemlost = 0)
1515 ## In English:
1516 ## all records not indexed in the onloan register (zebra) and all records with a value of lost equal to 0
1517 $availability_limit .=
1518 "( ( allrecords,AlwaysMatches='' not onloan,AlwaysMatches='') and (lost,st-numeric=0) )"; #or ( allrecords,AlwaysMatches='' not lost,AlwaysMatches='')) )";
1519 $limit_cgi .= "&limit=available";
1520 $limit_desc .= "";
1523 # group_OR_limits, prefixed by mc-
1524 # OR every member of the group
1525 elsif ( $this_limit =~ /mc/ ) {
1526 my ($k,$v) = split(/:/, $this_limit,2);
1527 if ( $k !~ /mc-i(tem)?type/ ) {
1528 # in case the mc-ccode value has complicating chars like ()'s inside it we wrap in quotes
1529 $this_limit =~ tr/"//d;
1530 $this_limit = $k.":\"".$v."\"";
1533 $group_OR_limits{$k} .= " or " if $group_OR_limits{$k};
1534 $limit_desc .= " or " if $group_OR_limits{$k};
1535 $group_OR_limits{$k} .= "$this_limit";
1536 $limit_cgi .= "&limit=" . uri_escape($this_limit);
1537 $limit_desc .= " $this_limit";
1540 # Regular old limits
1541 else {
1542 $limit .= " and " if $limit || $query;
1543 $limit .= "$this_limit";
1544 $limit_cgi .= "&limit=" . uri_escape($this_limit);
1545 if ($this_limit =~ /^branch:(.+)/) {
1546 my $branchcode = $1;
1547 my $branchname = GetBranchName($branchcode);
1548 if (defined $branchname) {
1549 $limit_desc .= " branch:$branchname";
1550 } else {
1551 $limit_desc .= " $this_limit";
1553 } else {
1554 $limit_desc .= " $this_limit";
1558 foreach my $k (keys (%group_OR_limits)) {
1559 $limit .= " and " if ( $query || $limit );
1560 $limit .= "($group_OR_limits{$k})";
1562 if ($availability_limit) {
1563 $limit .= " and " if ( $query || $limit );
1564 $limit .= "($availability_limit)";
1567 # Normalize the query and limit strings
1568 # This is flawed , means we can't search anything with : in it
1569 # if user wants to do ccl or cql, start the query with that
1570 # $query =~ s/:/=/g;
1571 $query =~ s/(?<=(ti|au|pb|su|an|kw|mc|nb|ns)):/=/g;
1572 $query =~ s/(?<=(wrdl)):/=/g;
1573 $query =~ s/(?<=(trn|phr)):/=/g;
1574 $limit =~ s/:/=/g;
1575 for ( $query, $query_desc, $limit, $limit_desc ) {
1576 s/ +/ /g; # remove extra spaces
1577 s/^ //g; # remove any beginning spaces
1578 s/ $//g; # remove any ending spaces
1579 s/==/=/g; # remove double == from query
1581 $query_cgi =~ s/^&//; # remove unnecessary & from beginning of the query cgi
1583 for ($query_cgi,$simple_query) {
1584 s/"//g;
1586 # append the limit to the query
1587 $query .= " " . $limit;
1589 # Warnings if DEBUG
1590 if ($DEBUG) {
1591 warn "QUERY:" . $query;
1592 warn "QUERY CGI:" . $query_cgi;
1593 warn "QUERY DESC:" . $query_desc;
1594 warn "LIMIT:" . $limit;
1595 warn "LIMIT CGI:" . $limit_cgi;
1596 warn "LIMIT DESC:" . $limit_desc;
1597 warn "---------\nLeave buildQuery\n---------";
1599 return (
1600 undef, $query, $simple_query, $query_cgi,
1601 $query_desc, $limit, $limit_cgi, $limit_desc,
1602 $stopwords_removed, $query_type
1606 =head2 _build_initial_query
1608 ($query, $query_cgi, $query_desc, $previous_operand) = _build_initial_query($initial_query_params);
1610 Build a section of the initial query containing indexes, operators, and operands.
1612 =cut
1614 sub _build_initial_query {
1615 my ($params) = @_;
1617 my $operator = "";
1618 if ($params->{previous_operand}){
1619 #If there is a previous operand, add a supplied operator or the default 'and'
1620 $operator = ($params->{operator}) ? " ".($params->{operator})." " : ' and ';
1623 #NOTE: indexes_set is typically set when doing truncation or field weighting
1624 my $operand = ($params->{indexes_set}) ? $params->{parsed_operand} : $params->{index_plus}.$params->{parsed_operand};
1626 #e.g. "kw,wrdl:test"
1627 #e.g. " and kw,wrdl:test"
1628 $params->{query} .= $operator . $operand;
1630 $params->{query_cgi} .= "&op=".uri_escape($operator) if $operator;
1631 $params->{query_cgi} .= "&idx=".uri_escape($params->{index}) if $params->{index};
1632 $params->{query_cgi} .= "&q=".uri_escape($params->{original_operand}) if $params->{original_operand};
1634 #e.g. " and kw,wrdl: test"
1635 $params->{query_desc} .= $operator . $params->{index_plus} . " " . $params->{original_operand};
1637 $params->{previous_operand} = 1 unless $params->{previous_operand}; #If there is no previous operand, mark this as one
1639 return ($params->{query}, $params->{query_cgi}, $params->{query_desc}, $params->{previous_operand});
1642 =head2 searchResults
1644 my @search_results = searchResults($search_context, $searchdesc, $hits,
1645 $results_per_page, $offset, $scan,
1646 @marcresults);
1648 Format results in a form suitable for passing to the template
1650 =cut
1652 # IMO this subroutine is pretty messy still -- it's responsible for
1653 # building the HTML output for the template
1654 sub searchResults {
1655 my ( $search_context, $searchdesc, $hits, $results_per_page, $offset, $scan, $marcresults ) = @_;
1656 my $dbh = C4::Context->dbh;
1657 my @newresults;
1659 require C4::Items;
1661 $search_context = 'opac' if !$search_context || $search_context ne 'intranet';
1662 my ($is_opac, $hidelostitems);
1663 if ($search_context eq 'opac') {
1664 $hidelostitems = C4::Context->preference('hidelostitems');
1665 $is_opac = 1;
1668 #Build branchnames hash
1669 #find branchname
1670 #get branch information.....
1671 my %branches;
1672 my $bsth =$dbh->prepare("SELECT branchcode,branchname FROM branches"); # FIXME : use C4::Branch::GetBranches
1673 $bsth->execute();
1674 while ( my $bdata = $bsth->fetchrow_hashref ) {
1675 $branches{ $bdata->{'branchcode'} } = $bdata->{'branchname'};
1677 # FIXME - We build an authorised values hash here, using the default framework
1678 # though it is possible to have different authvals for different fws.
1680 my $shelflocations =GetKohaAuthorisedValues('items.location','');
1682 # get notforloan authorised value list (see $shelflocations FIXME)
1683 my $notforloan_authorised_value = GetAuthValCode('items.notforloan','');
1685 #Build itemtype hash
1686 #find itemtype & itemtype image
1687 my %itemtypes;
1688 $bsth =
1689 $dbh->prepare(
1690 "SELECT itemtype,description,imageurl,summary,notforloan FROM itemtypes"
1692 $bsth->execute();
1693 while ( my $bdata = $bsth->fetchrow_hashref ) {
1694 foreach (qw(description imageurl summary notforloan)) {
1695 $itemtypes{ $bdata->{'itemtype'} }->{$_} = $bdata->{$_};
1699 #search item field code
1700 my ($itemtag, undef) = &GetMarcFromKohaField( "items.itemnumber", "" );
1702 ## find column names of items related to MARC
1703 my $sth2 = $dbh->prepare("SHOW COLUMNS FROM items");
1704 $sth2->execute;
1705 my %subfieldstosearch;
1706 while ( ( my $column ) = $sth2->fetchrow ) {
1707 my ( $tagfield, $tagsubfield ) =
1708 &GetMarcFromKohaField( "items." . $column, "" );
1709 if ( defined $tagsubfield ) {
1710 $subfieldstosearch{$column} = $tagsubfield;
1714 # handle which records to actually retrieve
1715 my $times;
1716 if ( $hits && $offset + $results_per_page <= $hits ) {
1717 $times = $offset + $results_per_page;
1719 else {
1720 $times = $hits; # FIXME: if $hits is undefined, why do we want to equal it?
1723 my $marcflavour = C4::Context->preference("marcflavour");
1724 # We get the biblionumber position in MARC
1725 my ($bibliotag,$bibliosubf)=GetMarcFromKohaField('biblio.biblionumber','');
1727 # loop through all of the records we've retrieved
1728 for ( my $i = $offset ; $i <= $times - 1 ; $i++ ) {
1730 my $marcrecord;
1731 if ($scan) {
1732 # For Scan searches we built USMARC data
1733 $marcrecord = MARC::Record->new_from_usmarc( $marcresults->[$i]);
1734 } else {
1735 # Normal search, render from Zebra's output
1736 $marcrecord = new_record_from_zebra(
1737 'biblioserver',
1738 $marcresults->[$i]
1741 if ( ! defined $marcrecord ) {
1742 warn "ERROR DECODING RECORD - $@: " . $marcresults->[$i];
1743 next;
1747 my $fw = $scan
1748 ? undef
1749 : $bibliotag < 10
1750 ? GetFrameworkCode($marcrecord->field($bibliotag)->data)
1751 : GetFrameworkCode($marcrecord->subfield($bibliotag,$bibliosubf));
1752 my $oldbiblio = TransformMarcToKoha( $dbh, $marcrecord, $fw );
1753 $oldbiblio->{subtitle} = GetRecordValue('subtitle', $marcrecord, $fw);
1754 $oldbiblio->{result_number} = $i + 1;
1756 # add imageurl to itemtype if there is one
1757 $oldbiblio->{imageurl} = getitemtypeimagelocation( $search_context, $itemtypes{ $oldbiblio->{itemtype} }->{imageurl} );
1759 $oldbiblio->{'authorised_value_images'} = ($search_context eq 'opac' && C4::Context->preference('AuthorisedValueImages')) || ($search_context eq 'intranet' && C4::Context->preference('StaffAuthorisedValueImages')) ? C4::Items::get_authorised_value_images( C4::Biblio::get_biblio_authorised_values( $oldbiblio->{'biblionumber'}, $marcrecord ) ) : [];
1760 $oldbiblio->{normalized_upc} = GetNormalizedUPC( $marcrecord,$marcflavour);
1761 $oldbiblio->{normalized_ean} = GetNormalizedEAN( $marcrecord,$marcflavour);
1762 $oldbiblio->{normalized_oclc} = GetNormalizedOCLCNumber($marcrecord,$marcflavour);
1763 $oldbiblio->{normalized_isbn} = GetNormalizedISBN(undef,$marcrecord,$marcflavour);
1764 $oldbiblio->{content_identifier_exists} = 1 if ($oldbiblio->{normalized_isbn} or $oldbiblio->{normalized_oclc} or $oldbiblio->{normalized_ean} or $oldbiblio->{normalized_upc});
1766 # edition information, if any
1767 $oldbiblio->{edition} = $oldbiblio->{editionstatement};
1768 $oldbiblio->{description} = $itemtypes{ $oldbiblio->{itemtype} }->{description};
1769 # Build summary if there is one (the summary is defined in the itemtypes table)
1770 # FIXME: is this used anywhere, I think it can be commented out? -- JF
1771 if ( $itemtypes{ $oldbiblio->{itemtype} }->{summary} ) {
1772 my $summary = $itemtypes{ $oldbiblio->{itemtype} }->{summary};
1773 my @fields = $marcrecord->fields();
1775 my $newsummary;
1776 foreach my $line ( "$summary\n" =~ /(.*)\n/g ){
1777 my $tags = {};
1778 foreach my $tag ( $line =~ /\[(\d{3}[\w|\d])\]/ ) {
1779 $tag =~ /(.{3})(.)/;
1780 if($marcrecord->field($1)){
1781 my @abc = $marcrecord->field($1)->subfield($2);
1782 $tags->{$tag} = $#abc + 1 ;
1786 # We catch how many times to repeat this line
1787 my $max = 0;
1788 foreach my $tag (keys(%$tags)){
1789 $max = $tags->{$tag} if($tags->{$tag} > $max);
1792 # we replace, and repeat each line
1793 for (my $i = 0 ; $i < $max ; $i++){
1794 my $newline = $line;
1796 foreach my $tag ( $newline =~ /\[(\d{3}[\w|\d])\]/g ) {
1797 $tag =~ /(.{3})(.)/;
1799 if($marcrecord->field($1)){
1800 my @repl = $marcrecord->field($1)->subfield($2);
1801 my $subfieldvalue = $repl[$i];
1803 if (! utf8::is_utf8($subfieldvalue)) {
1804 utf8::decode($subfieldvalue);
1807 $newline =~ s/\[$tag\]/$subfieldvalue/g;
1810 $newsummary .= "$newline\n";
1814 $newsummary =~ s/\[(.*?)]//g;
1815 $newsummary =~ s/\n/<br\/>/g;
1816 $oldbiblio->{summary} = $newsummary;
1819 # Pull out the items fields
1820 my @fields = $marcrecord->field($itemtag);
1821 my $marcflavor = C4::Context->preference("marcflavour");
1822 # adding linked items that belong to host records
1823 my $analyticsfield = '773';
1824 if ($marcflavor eq 'MARC21' || $marcflavor eq 'NORMARC') {
1825 $analyticsfield = '773';
1826 } elsif ($marcflavor eq 'UNIMARC') {
1827 $analyticsfield = '461';
1829 foreach my $hostfield ( $marcrecord->field($analyticsfield)) {
1830 my $hostbiblionumber = $hostfield->subfield("0");
1831 my $linkeditemnumber = $hostfield->subfield("9");
1832 if(!$hostbiblionumber eq undef){
1833 my $hostbiblio = GetMarcBiblio($hostbiblionumber, 1);
1834 my ($itemfield, undef) = GetMarcFromKohaField( 'items.itemnumber', GetFrameworkCode($hostbiblionumber) );
1835 if(!$hostbiblio eq undef){
1836 my @hostitems = $hostbiblio->field($itemfield);
1837 foreach my $hostitem (@hostitems){
1838 if ($hostitem->subfield("9") eq $linkeditemnumber){
1839 my $linkeditem =$hostitem;
1840 # append linked items if they exist
1841 if (!$linkeditem eq undef){
1842 push (@fields, $linkeditem);}
1849 # Setting item statuses for display
1850 my @available_items_loop;
1851 my @onloan_items_loop;
1852 my @other_items_loop;
1854 my $available_items;
1855 my $onloan_items;
1856 my $other_items;
1858 my $ordered_count = 0;
1859 my $available_count = 0;
1860 my $onloan_count = 0;
1861 my $longoverdue_count = 0;
1862 my $other_count = 0;
1863 my $withdrawn_count = 0;
1864 my $itemlost_count = 0;
1865 my $hideatopac_count = 0;
1866 my $itembinding_count = 0;
1867 my $itemdamaged_count = 0;
1868 my $item_in_transit_count = 0;
1869 my $can_place_holds = 0;
1870 my $item_onhold_count = 0;
1871 my $notforloan_count = 0;
1872 my $items_count = scalar(@fields);
1873 my $maxitems_pref = C4::Context->preference('maxItemsinSearchResults');
1874 my $maxitems = $maxitems_pref ? $maxitems_pref - 1 : 1;
1875 my @hiddenitems; # hidden itemnumbers based on OpacHiddenItems syspref
1877 # loop through every item
1878 foreach my $field (@fields) {
1879 my $item;
1881 # populate the items hash
1882 foreach my $code ( keys %subfieldstosearch ) {
1883 $item->{$code} = $field->subfield( $subfieldstosearch{$code} );
1885 $item->{description} = $itemtypes{ $item->{itype} }{description};
1887 # OPAC hidden items
1888 if ($is_opac) {
1889 # hidden because lost
1890 if ($hidelostitems && $item->{itemlost}) {
1891 $hideatopac_count++;
1892 next;
1894 # hidden based on OpacHiddenItems syspref
1895 my @hi = C4::Items::GetHiddenItemnumbers($item);
1896 if (scalar @hi) {
1897 push @hiddenitems, @hi;
1898 $hideatopac_count++;
1899 next;
1903 my $hbranch = C4::Context->preference('HomeOrHoldingBranch') eq 'homebranch' ? 'homebranch' : 'holdingbranch';
1904 my $otherbranch = C4::Context->preference('HomeOrHoldingBranch') eq 'homebranch' ? 'holdingbranch' : 'homebranch';
1906 # set item's branch name, use HomeOrHoldingBranch syspref first, fall back to the other one
1907 if ($item->{$hbranch}) {
1908 $item->{'branchname'} = $branches{$item->{$hbranch}};
1910 elsif ($item->{$otherbranch}) { # Last resort
1911 $item->{'branchname'} = $branches{$item->{$otherbranch}};
1914 my $prefix = $item->{$hbranch} . '--' . $item->{location} . $item->{itype} . $item->{itemcallnumber};
1915 # For each grouping of items (onloan, available, unavailable), we build a key to store relevant info about that item
1916 my $userenv = C4::Context->userenv;
1917 if ( $item->{onloan} && !(C4::Members::GetHideLostItemsPreference($userenv->{'number'}) && $item->{itemlost}) ) {
1918 $onloan_count++;
1919 my $key = $prefix . $item->{onloan} . $item->{barcode};
1920 $onloan_items->{$key}->{due_date} = format_date($item->{onloan});
1921 $onloan_items->{$key}->{count}++ if $item->{$hbranch};
1922 $onloan_items->{$key}->{branchname} = $item->{branchname};
1923 $onloan_items->{$key}->{location} = $shelflocations->{ $item->{location} };
1924 $onloan_items->{$key}->{itemcallnumber} = $item->{itemcallnumber};
1925 $onloan_items->{$key}->{description} = $item->{description};
1926 $onloan_items->{$key}->{imageurl} = getitemtypeimagelocation( $search_context, $itemtypes{ $item->{itype} }->{imageurl} );
1927 # if something's checked out and lost, mark it as 'long overdue'
1928 if ( $item->{itemlost} ) {
1929 $onloan_items->{$prefix}->{longoverdue}++;
1930 $longoverdue_count++;
1931 } else { # can place holds as long as item isn't lost
1932 $can_place_holds = 1;
1936 # items not on loan, but still unavailable ( lost, withdrawn, damaged )
1937 else {
1939 # item is on order
1940 if ( $item->{notforloan} < 0 ) {
1941 $ordered_count++;
1942 } elsif ( $item->{notforloan} > 0 ) {
1943 $notforloan_count++;
1946 # is item in transit?
1947 my $transfertwhen = '';
1948 my ($transfertfrom, $transfertto);
1950 # is item on the reserve shelf?
1951 my $reservestatus = '';
1953 unless ($item->{withdrawn}
1954 || $item->{itemlost}
1955 || $item->{damaged}
1956 || $item->{notforloan}
1957 || $items_count > 20) {
1959 # A couple heuristics to limit how many times
1960 # we query the database for item transfer information, sacrificing
1961 # accuracy in some cases for speed;
1963 # 1. don't query if item has one of the other statuses
1964 # 2. don't check transit status if the bib has
1965 # more than 20 items
1967 # FIXME: to avoid having the query the database like this, and to make
1968 # the in transit status count as unavailable for search limiting,
1969 # should map transit status to record indexed in Zebra.
1971 ($transfertwhen, $transfertfrom, $transfertto) = C4::Circulation::GetTransfers($item->{itemnumber});
1972 $reservestatus = C4::Reserves::GetReserveStatus( $item->{itemnumber}, $oldbiblio->{biblionumber} );
1975 # item is withdrawn, lost, damaged, not for loan, reserved or in transit
1976 if ( $item->{withdrawn}
1977 || $item->{itemlost}
1978 || $item->{damaged}
1979 || $item->{notforloan}
1980 || $reservestatus eq 'Waiting'
1981 || ($transfertwhen ne ''))
1983 $withdrawn_count++ if $item->{withdrawn};
1984 $itemlost_count++ if $item->{itemlost};
1985 $itemdamaged_count++ if $item->{damaged};
1986 $item_in_transit_count++ if $transfertwhen ne '';
1987 $item_onhold_count++ if $reservestatus eq 'Waiting';
1988 $item->{status} = $item->{withdrawn} . "-" . $item->{itemlost} . "-" . $item->{damaged} . "-" . $item->{notforloan};
1990 # can place a hold on a item if
1991 # not lost nor withdrawn
1992 # not damaged unless AllowHoldsOnDamagedItems is true
1993 # item is either for loan or on order (notforloan < 0)
1994 $can_place_holds = 1
1995 if (
1996 !$item->{itemlost}
1997 && !$item->{withdrawn}
1998 && ( !$item->{damaged} || C4::Context->preference('AllowHoldsOnDamagedItems') )
1999 && ( !$item->{notforloan} || $item->{notforloan} < 0 )
2002 $other_count++;
2004 my $key = $prefix . $item->{status};
2005 foreach (qw(withdrawn itemlost damaged branchname itemcallnumber)) {
2006 $other_items->{$key}->{$_} = $item->{$_};
2008 $other_items->{$key}->{intransit} = ( $transfertwhen ne '' ) ? 1 : 0;
2009 $other_items->{$key}->{onhold} = ($reservestatus) ? 1 : 0;
2010 $other_items->{$key}->{notforloan} = GetAuthorisedValueDesc('','',$item->{notforloan},'','',$notforloan_authorised_value) if $notforloan_authorised_value and $item->{notforloan};
2011 $other_items->{$key}->{count}++ if $item->{$hbranch};
2012 $other_items->{$key}->{location} = $shelflocations->{ $item->{location} };
2013 $other_items->{$key}->{description} = $item->{description};
2014 $other_items->{$key}->{imageurl} = getitemtypeimagelocation( $search_context, $itemtypes{ $item->{itype} }->{imageurl} );
2016 # item is available
2017 else {
2018 $can_place_holds = 1;
2019 $available_count++;
2020 $available_items->{$prefix}->{count}++ if $item->{$hbranch};
2021 foreach (qw(branchname itemcallnumber description)) {
2022 $available_items->{$prefix}->{$_} = $item->{$_};
2024 $available_items->{$prefix}->{location} = $shelflocations->{ $item->{location} };
2025 $available_items->{$prefix}->{imageurl} = getitemtypeimagelocation( $search_context, $itemtypes{ $item->{itype} }->{imageurl} );
2028 } # notforloan, item level and biblioitem level
2030 # if all items are hidden, do not show the record
2031 if ($items_count > 0 && $hideatopac_count == $items_count) {
2032 next;
2035 my ( $availableitemscount, $onloanitemscount, $otheritemscount );
2036 for my $key ( sort keys %$onloan_items ) {
2037 (++$onloanitemscount > $maxitems) and last;
2038 push @onloan_items_loop, $onloan_items->{$key};
2040 for my $key ( sort keys %$other_items ) {
2041 (++$otheritemscount > $maxitems) and last;
2042 push @other_items_loop, $other_items->{$key};
2044 for my $key ( sort keys %$available_items ) {
2045 (++$availableitemscount > $maxitems) and last;
2046 push @available_items_loop, $available_items->{$key}
2049 # XSLT processing of some stuff
2050 use C4::Charset;
2051 SetUTF8Flag($marcrecord);
2052 warn $marcrecord->as_formatted if $DEBUG;
2053 my $interface = $search_context eq 'opac' ? 'OPAC' : '';
2054 if (!$scan && C4::Context->preference($interface . "XSLTResultsDisplay")) {
2055 $oldbiblio->{XSLTResultsRecord} = XSLTParse4Display($oldbiblio->{biblionumber}, $marcrecord, $interface."XSLTResultsDisplay", 1, \@hiddenitems);
2056 # the last parameter tells Koha to clean up the problematic ampersand entities that Zebra outputs
2059 # if biblio level itypes are used and itemtype is notforloan, it can't be reserved either
2060 if (!C4::Context->preference("item-level_itypes")) {
2061 if ($itemtypes{ $oldbiblio->{itemtype} }->{notforloan}) {
2062 $can_place_holds = 0;
2065 $oldbiblio->{norequests} = 1 unless $can_place_holds;
2066 $oldbiblio->{itemsplural} = 1 if $items_count > 1;
2067 $oldbiblio->{items_count} = $items_count;
2068 $oldbiblio->{available_items_loop} = \@available_items_loop;
2069 $oldbiblio->{onloan_items_loop} = \@onloan_items_loop;
2070 $oldbiblio->{other_items_loop} = \@other_items_loop;
2071 $oldbiblio->{availablecount} = $available_count;
2072 $oldbiblio->{availableplural} = 1 if $available_count > 1;
2073 $oldbiblio->{onloancount} = $onloan_count;
2074 $oldbiblio->{onloanplural} = 1 if $onloan_count > 1;
2075 $oldbiblio->{othercount} = $other_count;
2076 $oldbiblio->{otherplural} = 1 if $other_count > 1;
2077 $oldbiblio->{withdrawncount} = $withdrawn_count;
2078 $oldbiblio->{itemlostcount} = $itemlost_count;
2079 $oldbiblio->{damagedcount} = $itemdamaged_count;
2080 $oldbiblio->{intransitcount} = $item_in_transit_count;
2081 $oldbiblio->{onholdcount} = $item_onhold_count;
2082 $oldbiblio->{orderedcount} = $ordered_count;
2083 $oldbiblio->{notforloancount} = $notforloan_count;
2085 if (C4::Context->preference("AlternateHoldingsField") && $items_count == 0) {
2086 my $fieldspec = C4::Context->preference("AlternateHoldingsField");
2087 my $subfields = substr $fieldspec, 3;
2088 my $holdingsep = C4::Context->preference("AlternateHoldingsSeparator") || ' ';
2089 my @alternateholdingsinfo = ();
2090 my @holdingsfields = $marcrecord->field(substr $fieldspec, 0, 3);
2091 my $alternateholdingscount = 0;
2093 for my $field (@holdingsfields) {
2094 my %holding = ( holding => '' );
2095 my $havesubfield = 0;
2096 for my $subfield ($field->subfields()) {
2097 if ((index $subfields, $$subfield[0]) >= 0) {
2098 $holding{'holding'} .= $holdingsep if (length $holding{'holding'} > 0);
2099 $holding{'holding'} .= $$subfield[1];
2100 $havesubfield++;
2103 if ($havesubfield) {
2104 push(@alternateholdingsinfo, \%holding);
2105 $alternateholdingscount++;
2109 $oldbiblio->{'ALTERNATEHOLDINGS'} = \@alternateholdingsinfo;
2110 $oldbiblio->{'alternateholdings_count'} = $alternateholdingscount;
2113 push( @newresults, $oldbiblio );
2116 return @newresults;
2119 =head2 SearchAcquisitions
2120 Search for acquisitions
2121 =cut
2123 sub SearchAcquisitions{
2124 my ($datebegin, $dateend, $itemtypes,$criteria, $orderby) = @_;
2126 my $dbh=C4::Context->dbh;
2127 # Variable initialization
2128 my $str=qq|
2129 SELECT marcxml
2130 FROM biblio
2131 LEFT JOIN biblioitems ON biblioitems.biblionumber=biblio.biblionumber
2132 LEFT JOIN items ON items.biblionumber=biblio.biblionumber
2133 WHERE dateaccessioned BETWEEN ? AND ?
2136 my (@params,@loopcriteria);
2138 push @params, $datebegin->output("iso");
2139 push @params, $dateend->output("iso");
2141 if (scalar(@$itemtypes)>0 and $criteria ne "itemtype" ){
2142 if(C4::Context->preference("item-level_itypes")){
2143 $str .= "AND items.itype IN (?".( ',?' x scalar @$itemtypes - 1 ).") ";
2144 }else{
2145 $str .= "AND biblioitems.itemtype IN (?".( ',?' x scalar @$itemtypes - 1 ).") ";
2147 push @params, @$itemtypes;
2150 if ($criteria =~/itemtype/){
2151 if(C4::Context->preference("item-level_itypes")){
2152 $str .= "AND items.itype=? ";
2153 }else{
2154 $str .= "AND biblioitems.itemtype=? ";
2157 if(scalar(@$itemtypes) == 0){
2158 my $itypes = GetItemTypes();
2159 for my $key (keys %$itypes){
2160 push @$itemtypes, $key;
2164 @loopcriteria= @$itemtypes;
2165 }elsif ($criteria=~/itemcallnumber/){
2166 $str .= "AND (items.itemcallnumber LIKE CONCAT(?,'%')
2167 OR items.itemcallnumber is NULL
2168 OR items.itemcallnumber = '')";
2170 @loopcriteria = ("AA".."ZZ", "") unless (scalar(@loopcriteria)>0);
2171 }else {
2172 $str .= "AND biblio.title LIKE CONCAT(?,'%') ";
2173 @loopcriteria = ("A".."z") unless (scalar(@loopcriteria)>0);
2176 if ($orderby =~ /date_desc/){
2177 $str.=" ORDER BY dateaccessioned DESC";
2178 } else {
2179 $str.=" ORDER BY title";
2182 my $qdataacquisitions=$dbh->prepare($str);
2184 my @loopacquisitions;
2185 foreach my $value(@loopcriteria){
2186 push @params,$value;
2187 my %cell;
2188 $cell{"title"}=$value;
2189 $cell{"titlecode"}=$value;
2191 eval{$qdataacquisitions->execute(@params);};
2193 if ($@){ warn "recentacquisitions Error :$@";}
2194 else {
2195 my @loopdata;
2196 while (my $data=$qdataacquisitions->fetchrow_hashref){
2197 push @loopdata, {"summary"=>GetBiblioSummary( $data->{'marcxml'} ) };
2199 $cell{"loopdata"}=\@loopdata;
2201 push @loopacquisitions,\%cell if (scalar(@{$cell{loopdata}})>0);
2202 pop @params;
2204 $qdataacquisitions->finish;
2205 return \@loopacquisitions;
2208 =head2 enabled_staff_search_views
2210 %hash = enabled_staff_search_views()
2212 This function returns a hash that contains three flags obtained from the system
2213 preferences, used to determine whether a particular staff search results view
2214 is enabled.
2216 =over 2
2218 =item C<Output arg:>
2220 * $hash{can_view_MARC} is true only if the MARC view is enabled
2221 * $hash{can_view_ISBD} is true only if the ISBD view is enabled
2222 * $hash{can_view_labeledMARC} is true only if the Labeled MARC view is enabled
2224 =item C<usage in the script:>
2226 =back
2228 $template->param ( C4::Search::enabled_staff_search_views );
2230 =cut
2232 sub enabled_staff_search_views
2234 return (
2235 can_view_MARC => C4::Context->preference('viewMARC'), # 1 if the staff search allows the MARC view
2236 can_view_ISBD => C4::Context->preference('viewISBD'), # 1 if the staff search allows the ISBD view
2237 can_view_labeledMARC => C4::Context->preference('viewLabeledMARC'), # 1 if the staff search allows the Labeled MARC view
2241 sub PurgeSearchHistory{
2242 my ($pSearchhistory)=@_;
2243 my $dbh = C4::Context->dbh;
2244 my $sth = $dbh->prepare("DELETE FROM search_history WHERE time < DATE_SUB( NOW(), INTERVAL ? DAY )");
2245 $sth->execute($pSearchhistory) or die $dbh->errstr;
2248 =head2 z3950_search_args
2250 $arrayref = z3950_search_args($matchpoints)
2252 This function returns an array reference that contains the search parameters to be
2253 passed to the Z39.50 search script (z3950_search.pl). The array elements
2254 are hash refs whose keys are name and value, and whose values are the
2255 name of a search parameter, the value of that search parameter and the URL encoded
2256 value of that parameter.
2258 The search parameter names are lccn, isbn, issn, title, author, dewey and subject.
2260 The search parameter values are obtained from the bibliographic record whose
2261 data is in a hash reference in $matchpoints, as returned by Biblio::GetBiblioData().
2263 If $matchpoints is a scalar, it is assumed to be an unnamed query descriptor, e.g.
2264 a general purpose search argument. In this case, the returned array contains only
2265 entry: the key is 'title' and the value is derived from $matchpoints.
2267 If a search parameter value is undefined or empty, it is not included in the returned
2268 array.
2270 The returned array reference may be passed directly to the template parameters.
2272 =over 2
2274 =item C<Output arg:>
2276 * $array containing hash refs as described above
2278 =item C<usage in the script:>
2280 =back
2282 $data = Biblio::GetBiblioData($bibno);
2283 $template->param ( MYLOOP => C4::Search::z3950_search_args($data) )
2285 *OR*
2287 $template->param ( MYLOOP => C4::Search::z3950_search_args($searchscalar) )
2289 =cut
2291 sub z3950_search_args {
2292 my $bibrec = shift;
2294 my $isbn_string = ref( $bibrec ) ? $bibrec->{title} : $bibrec;
2295 my $isbn = Business::ISBN->new( $isbn_string );
2297 if (defined $isbn && $isbn->is_valid)
2299 if ( ref($bibrec) ) {
2300 $bibrec->{isbn} = $isbn_string;
2301 $bibrec->{title} = undef;
2302 } else {
2303 $bibrec = { isbn => $isbn_string };
2306 else {
2307 $bibrec = { title => $bibrec } if !ref $bibrec;
2309 my $array = [];
2310 for my $field (qw/ lccn isbn issn title author dewey subject /)
2312 push @$array, { name => $field, value => $bibrec->{$field} }
2313 if defined $bibrec->{$field};
2315 return $array;
2318 =head2 GetDistinctValues($field);
2320 C<$field> is a reference to the fields array
2322 =cut
2324 sub GetDistinctValues {
2325 my ($fieldname,$string)=@_;
2326 # returns a reference to a hash of references to branches...
2327 if ($fieldname=~/\./){
2328 my ($table,$column)=split /\./, $fieldname;
2329 my $dbh = C4::Context->dbh;
2330 warn "select DISTINCT($column) as value, count(*) as cnt from $table group by lib order by $column " if $DEBUG;
2331 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 ");
2332 $sth->execute;
2333 my $elements=$sth->fetchall_arrayref({});
2334 return $elements;
2336 else {
2337 $string||= qq("");
2338 my @servers=qw<biblioserver authorityserver>;
2339 my (@zconns,@results);
2340 for ( my $i = 0 ; $i < @servers ; $i++ ) {
2341 $zconns[$i] = C4::Context->Zconn( $servers[$i], 1 );
2342 $results[$i] =
2343 $zconns[$i]->scan(
2344 ZOOM::Query::CCL2RPN->new( qq"$fieldname $string", $zconns[$i])
2347 # The big moment: asynchronously retrieve results from all servers
2348 my @elements;
2349 _ZOOM_event_loop(
2350 \@zconns,
2351 \@results,
2352 sub {
2353 my ( $i, $size ) = @_;
2354 for ( my $j = 0 ; $j < $size ; $j++ ) {
2355 my %hashscan;
2356 @hashscan{qw(value cnt)} =
2357 $results[ $i - 1 ]->display_term($j);
2358 push @elements, \%hashscan;
2362 return \@elements;
2366 =head2 _ZOOM_event_loop
2368 _ZOOM_event_loop(\@zconns, \@results, sub {
2369 my ( $i, $size ) = @_;
2370 ....
2371 } );
2373 Processes a ZOOM event loop and passes control to a closure for
2374 processing the results, and destroying the resultsets.
2376 =cut
2378 sub _ZOOM_event_loop {
2379 my ($zconns, $results, $callback) = @_;
2380 while ( ( my $i = ZOOM::event( $zconns ) ) != 0 ) {
2381 my $ev = $zconns->[ $i - 1 ]->last_event();
2382 if ( $ev == ZOOM::Event::ZEND ) {
2383 next unless $results->[ $i - 1 ];
2384 my $size = $results->[ $i - 1 ]->size();
2385 if ( $size > 0 ) {
2386 $callback->($i, $size);
2391 foreach my $result (@$results) {
2392 $result->destroy();
2396 =head2 new_record_from_zebra
2398 Given raw data from a Zebra result set, return a MARC::Record object
2400 This helper function is needed to take into account all the involved
2401 system preferences and configuration variables to properly create the
2402 MARC::Record object.
2404 If we are using GRS-1, then the raw data we get from Zebra should be USMARC
2405 data. If we are using DOM, then it has to be MARCXML.
2407 =cut
2409 sub new_record_from_zebra {
2411 my $server = shift;
2412 my $raw_data = shift;
2413 # Set the default indexing modes
2414 my $index_mode = ( $server eq 'biblioserver' )
2415 ? C4::Context->config('zebra_bib_index_mode') // 'grs1'
2416 : C4::Context->config('zebra_auth_index_mode') // 'dom';
2418 my $marc_record = eval {
2419 if ( $index_mode eq 'dom' ) {
2420 MARC::Record->new_from_xml( $raw_data, 'UTF-8' );
2421 } else {
2422 MARC::Record->new_from_usmarc( $raw_data );
2426 if ($@) {
2427 return;
2428 } else {
2429 return $marc_record;
2434 END { } # module clean-up code here (global destructor)
2437 __END__
2439 =head1 AUTHOR
2441 Koha Development Team <http://koha-community.org/>
2443 =cut