2 # Script to perform searching
3 # For documentation try 'perldoc /path/to/search'
7 # Copyright 2006 LibLime
9 # This file is part of Koha
11 # Koha is free software; you can redistribute it and/or modify it under the
12 # terms of the GNU General Public License as published by the Free Software
13 # Foundation; either version 2 of the License, or (at your option) any later
16 # Koha is distributed in the hope that it will be useful, but WITHOUT ANY
17 # WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
18 # A PARTICULAR PURPOSE. See the GNU General Public License for more details.
20 # You should have received a copy of the GNU General Public License along with
21 # Koha; if not, write to the Free Software Foundation, Inc., 59 Temple Place,
22 # Suite 330, Boston, MA 02111-1307 USA
26 search - a search script for finding records in a Koha system (Version 3)
30 This script utilizes a new search API for Koha 3. It is designed to be
31 simple to use and configure, yet capable of performing feats like stemming,
32 field weighting, relevance ranking, support for multiple query language
33 formats (CCL, CQL, PQF), full support for the bib1 attribute set, extended
34 attribute sets defined in Zebra profiles, access to the full range of Z39.50
35 and SRU query options, federated searches on Z39.50/SRU targets, etc.
37 The API as represented in this script is mostly sound, even if the individual
38 functions in Search.pm and Koha.pm need to be cleaned up. Of course, you are
41 I will attempt to describe what is happening at each part of this script.
42 -- Joshua Ferraro <jmf AT liblime DOT com>
46 This script performs two functions:
50 =item 1. interacts with Koha to retrieve and display the results of a search
52 =item 2. loads the advanced search page
56 These two functions share many of the same variables and modules, so the first
57 task is to load what they have in common and determine which template to use.
58 Once determined, proceed to only load the variables and procedures necessary
61 =head2 LOADING ADVANCED SEARCH PAGE
63 This is fairly straightforward, and I won't go into detail ;-)
65 =head2 PERFORMING A SEARCH
67 If we're performing a search, this script performs three primary
72 =item 1. builds query strings (yes, plural)
74 =item 2. perform the search and return the results array
76 =item 3. build the HTML for output to the template
80 There are several additional secondary functions performed that I will
83 =head3 1. Building Query Strings
85 There are several types of queries needed in the process of search and retrieve:
89 =item 1 $query - the fully-built query passed to zebra
91 This is the most complex query that needs to be built. The original design goal
92 was to use a custom CCL2PQF query parser to translate an incoming CCL query into
93 a multi-leaf query to pass to Zebra. It needs to be multi-leaf to allow field
94 weighting, koha-specific relevance ranking, and stemming. When I have a chance
95 I'll try to flesh out this section to better explain.
97 This query incorporates query profiles that aren't compatible with most non-Zebra
98 Z39.50 targets to acomplish the field weighting and relevance ranking.
100 =item 2 $simple_query - a simple query that doesn't contain the field weighting,
101 stemming, etc., suitable to pass off to other search targets
103 This query is just the user's query expressed in CCL CQL, or PQF for passing to a
104 non-zebra Z39.50 target (one that doesn't support the extended profile that Zebra does).
106 =item 3 $query_cgi - passed to the template / saved for future refinements of
109 This is a simple string that completely expresses the query as a CGI string that
110 can be used for future refinements of the query or as a part of a history feature.
112 =item 4 $query_desc - Human search description - what the user sees in search
115 This is a simple string that is human readable. It will contain '=', ',', etc.
119 =head3 2. Perform the Search
121 This section takes the query strings and performs searches on the named servers,
122 including the Koha Zebra server, stores the results in a deeply nested object,
123 builds 'faceted results', and returns these objects.
127 The final major section of this script takes the objects collected thusfar and
128 builds the HTML for output to the template and user.
130 =head3 Additional Notes
136 use strict; # always use
138 ## STEP 1. Load things that are used in both search page and
139 # results page and decide which template to load, operations
147 use C4::Languages qw(getAllLanguages);
149 use POSIX qw(ceil floor);
150 use C4::Branch; # GetBranches
152 # create a new CGI object
153 # FIXME: no_undef_params needs to be tested
154 use CGI qw('-no_undef_params');
157 my ($template,$borrowernumber,$cookie);
159 # decide which template to use
162 my @params = $cgi->param("limit");
163 if ((@params>=1) || ($cgi->param("q")) || ($cgi->param('multibranchlimit')) || ($cgi->param('limit-yr')) ) {
164 $template_name = 'catalogue/results.tmpl';
167 $template_name = 'catalogue/advsearch.tmpl';
168 $template_type = 'advsearch';
171 ($template, $borrowernumber, $cookie) = get_template_and_user({
172 template_name => $template_name,
175 authnotrequired => 0,
176 flagsrequired => { catalogue => 1 },
179 if (C4::Context->preference("marcflavour") eq "UNIMARC" ) {
180 $template->param('UNIMARC' => 1);
184 # Deprecated, but preserved because it's interesting :-)
185 # The same thing can be accomplished with mod_rewrite in
189 #my $uri = $cgi->url(-base => 1);
190 #my $relative_url = $cgi->url(-relative=>1);
191 #$uri.="/".$relative_url."?";
193 #my @cgi_params_list = $cgi->param();
194 #my $url_params = $cgi->Vars;
196 #for my $each_param_set (@cgi_params_list) {
197 # $uri.= join "", map "\&$each_param_set=".$_, split("\0",$url_params->{$each_param_set}) if $url_params->{$each_param_set};
199 #warn "New URI:$uri";
200 # Only re-write a URI if there are params or if it already hasn't been re-written
201 #unless (($cgi->param('r')) || (!$cgi->param()) ) {
202 # print $cgi->redirect( -uri=>$uri."&r=1",
203 # -cookie => $cookie);
208 my $branches = GetBranches();
211 for my $branch_hash (sort { $branches->{$a}->{branchname} cmp $branches->{$b}->{branchname} } keys %$branches) {
212 push @branch_loop, {value => "$branch_hash" , branchname => $branches->{$branch_hash}->{'branchname'}, };
215 my $categories = GetBranchCategories(undef,'searchdomain');
217 $template->param(branchloop => \@branch_loop, searchdomainloop => $categories);
219 # load the Type stuff
220 # load the Type stuff
221 my $itemtypes = GetItemTypes;
222 # the index parameter is different for item-level itemtypes
223 my $itype_or_itemtype = (C4::Context->preference("item-level_itypes"))?'itype':'itemtype';
227 my $imgdir = getitemtypeimagesrc('intranet');
228 my $advanced_search_types = C4::Context->preference("AdvancedSearchTypes");
230 if (!$advanced_search_types or $advanced_search_types eq 'itemtypes') { foreach my $thisitemtype ( sort {$itemtypes->{$a}->{'description'} cmp $itemtypes->{$b}->{'description'} } keys %$itemtypes ) {
231 my %row =( number=>$cnt++,
232 imageurl=> $itemtypes->{$thisitemtype}->{'imageurl'}?($imgdir."/".$itemtypes->{$thisitemtype}->{'imageurl'}):"",
233 ccl => $itype_or_itemtype,
234 code => $thisitemtype,
235 selected => $selected,
236 description => $itemtypes->{$thisitemtype}->{'description'},
239 $selected = 0 if ($selected) ;
240 push @itemtypesloop, \%row;
242 $template->param(itemtypeloop => \@itemtypesloop);
244 my $advsearchtypes = GetAuthorisedValues($advanced_search_types);
245 for my $thisitemtype (@$advsearchtypes) {
248 imageurl=> $imgdir."/".$thisitemtype->{'imageurl'},
249 ccl => $advanced_search_types,
250 code => $thisitemtype->{authorised_value},
251 selected => $selected,
252 description => $thisitemtype->{'lib'},
255 push @itemtypesloop, \%row;
257 $template->param(itemtypeloop => \@itemtypesloop);
260 # The following should only be loaded if we're bringing up the advanced search template
261 if ( $template_type eq 'advsearch' ) {
263 # load the servers (used for searching -- to do federated searching, etc.)
264 my $primary_servers_loop;# = displayPrimaryServers();
265 $template->param(outer_servers_loop => $primary_servers_loop,);
267 my $secondary_servers_loop;# = displaySecondaryServers();
268 $template->param(outer_sup_servers_loop => $secondary_servers_loop,);
270 # determine what to display next to the search boxes (ie, boolean option
271 # shouldn't appear on the first one, scan indexes should, adding a new
272 # box should only appear on the last, etc.
273 my @search_boxes_array;
274 my $search_boxes_count = C4::Context->preference("OPACAdvSearchInputCount") | 3; # FIXME: should be a syspref
275 for (my $i=1;$i<=$search_boxes_count;$i++) {
276 # if it's the first one, don't display boolean option, but show scan indexes
278 push @search_boxes_array,
284 # if it's the last one, show the 'add field' box
285 elsif ($i==$search_boxes_count) {
286 push @search_boxes_array,
293 push @search_boxes_array,
300 $template->param(uc(C4::Context->preference("marcflavour")) => 1,
301 search_boxes_loop => \@search_boxes_array);
303 # load the language limits (for search)
304 my $languages_limit_loop = getAllLanguages();
305 $template->param(search_languages_loop => $languages_limit_loop,);
307 # use the global setting by default
308 if ( C4::Context->preference("expandedSearchOption") == 1) {
309 $template->param( expanded_options => C4::Context->preference("expandedSearchOption") );
311 # but let the user override it
312 if ( ($cgi->param('expanded_options') == 0) || ($cgi->param('expanded_options') == 1 ) ) {
313 $template->param( expanded_options => $cgi->param('expanded_options'));
316 output_html_with_http_headers $cgi, $cookie, $template->output;
320 ### OK, if we're this far, we're performing a search, not just loading the advanced search page
322 # Fetch the paramater list as a hash in scalar context:
323 # * returns paramater list as tied hash ref
324 # * we can edit the values by changing the key
325 # * multivalued CGI paramaters are returned as a packaged string separated by "\0" (null)
326 my $params = $cgi->Vars;
328 # Params that can have more than one value
329 # sort by is used to sort the query
330 # in theory can have more than one but generally there's just one
332 my $default_sort_by = C4::Context->preference('defaultSortField')."_".C4::Context->preference('defaultSortOrder')
333 if (C4::Context->preference('defaultSortField') && C4::Context->preference('defaultSortOrder'));
335 @sort_by = split("\0",$params->{'sort_by'}) if $params->{'sort_by'};
336 $sort_by[0] = $default_sort_by unless $sort_by[0];
337 foreach my $sort (@sort_by) {
338 $template->param($sort => 1);
340 $template->param('sort_by' => $sort_by[0]);
342 # Use the servers defined, or just search our local catalog(default)
344 @servers = split("\0",$params->{'server'}) if $params->{'server'};
346 #FIXME: this should be handled using Context.pm
347 @servers = ("biblioserver");
348 # @servers = C4::Context->config("biblioserver");
350 # operators include boolean and proximity operators and are used
351 # to evaluate multiple operands
353 @operators = split("\0",$params->{'op'}) if $params->{'op'};
355 # indexes are query qualifiers, like 'title', 'author', etc. They
356 # can be single or multiple parameters separated by comma: kw,right-Truncation
358 @indexes = split("\0",$params->{'idx'});
360 # an operand can be a single term, a phrase, or a complete ccl query
362 @operands = split("\0",$params->{'q'}) if $params->{'q'};
364 # limits are use to limit to results to a pre-defined category such as branch or language
366 @limits = split("\0",$params->{'limit'}) if $params->{'limit'};
368 if($params->{'multibranchlimit'}) {
369 push @limits, join(" or ", map { "branch: $_ "} @{GetBranchesInCategory($params->{'multibranchlimit'})}) ;
373 foreach my $limit(@limits) {
374 if ($limit =~/available/) {
378 $template->param(available => $available);
380 # append year limits if they exist
383 if ($params->{'limit-yr'}) {
384 if ($params->{'limit-yr'} =~ /\d{4}-\d{4}/) {
385 my ($yr1,$yr2) = split(/-/, $params->{'limit-yr'});
386 $limit_yr = "yr,st-numeric,ge=$yr1 and yr,st-numeric,le=$yr2";
387 $limit_yr_value = "$yr1-$yr2";
389 elsif ($params->{'limit-yr'} =~ /\d{4}/) {
390 $limit_yr = "yr,st-numeric=$params->{'limit-yr'}";
391 $limit_yr_value = $params->{'limit-yr'};
393 push @limits,$limit_yr;
394 #FIXME: Should return a error to the user, incorect date format specified
397 # Params that can only have one value
398 my $scan = $params->{'scan'};
399 my $count = C4::Context->preference('numSearchResults') || 20;
400 my $results_per_page = $params->{'count'} || $count;
401 my $offset = $params->{'offset'} || 0;
402 my $page = $cgi->param('page') || 1;
403 #my $offset = ($page-1)*$results_per_page;
405 my $expanded_facet = $params->{'expand'};
407 # Define some global variables
408 my ( $error,$query,$simple_query,$query_cgi,$query_desc,$limit,$limit_cgi,$limit_desc,$stopwords_removed,$query_type);
412 ## I. BUILD THE QUERY
413 ( $error,$query,$simple_query,$query_cgi,$query_desc,$limit,$limit_cgi,$limit_desc,$stopwords_removed,$query_type) = buildQuery(\@operators,\@operands,\@indexes,\@limits,\@sort_by,$scan);
415 ## parse the query_cgi string and put it into a form suitable for <input>s
419 for my $this_cgi ( split('&',$query_cgi) ) {
420 next unless $this_cgi;
421 $this_cgi =~ m/(.*=)(.*)/;
423 my $input_value = $2;
424 $input_name =~ s/=$//;
425 push @query_inputs, { input_name => $input_name, input_value => $input_value };
426 $scan_index = $input_value unless $scan_index;
428 $template->param ( QUERY_INPUTS => \@query_inputs,
429 scan_index => $scan_index );
431 ## parse the limit_cgi string and put it into a form suitable for <input>s
433 for my $this_cgi ( split('&',$limit_cgi) ) {
434 next unless $this_cgi;
435 # handle special case limit-yr
436 if ($this_cgi =~ /yr,st-numeric/) {
437 push @limit_inputs, { input_name => 'limit-yr', input_value => $limit_yr_value };
440 $this_cgi =~ m/(.*=)(.*)/;
442 my $input_value = $2;
443 $input_name =~ s/=$//;
444 push @limit_inputs, { input_name => $input_name, input_value => $input_value };
446 $template->param ( LIMIT_INPUTS => \@limit_inputs );
448 ## II. DO THE SEARCH AND GET THE RESULTS
449 my $total; # the total results for the whole set
450 my $facets; # this object stores the faceted results that display on the left-hand of the results page
454 if (C4::Context->preference('NoZebra')) {
455 $query=~s/yr(:|=)\s*([\d]{1,4})-([\d]{1,4})/(yr>=$2 and yr<=$3)/g;
456 $simple_query=~s/yr\s*(:|=)([\d]{1,4})-([\d]{1,4})/(yr>=$2 and yr<=$3)/g;
459 ($error, $results_hashref, $facets) = NZgetRecords($query,$simple_query,\@sort_by,\@servers,$results_per_page,$offset,$expanded_facet,$branches,$query_type,$scan);
463 ($error, $results_hashref, $facets) = getRecords($query,$simple_query,\@sort_by,\@servers,$results_per_page,$offset,$expanded_facet,$branches,$query_type,$scan);
467 $template->param(query_error => $error.$@);
469 output_html_with_http_headers $cgi, $cookie, $template->output;
473 # FIXME: This belongs in tools/ not in the primary search results page
474 my $op=$cgi->param("operation");
475 if ($op eq "bulkedit"){
476 my ($countchanged,$listunchanged)=
477 ModBiblios($results_hashref->{'biblioserver'}->{"RECORDS"},
478 $params->{"tagsubfield"},
479 $params->{"inputvalue"},
480 $params->{"targetvalue"},
483 $template->param(bulkeditresults=>1,
484 tagsubfield=>$params->{"tagsubfield"},
485 inputvalue=>$params->{"inputvalue"},
486 targetvalue=>$params->{"targetvalue"},
487 countchanged=>$countchanged,
488 countunchanged=>scalar(@$listunchanged),
489 listunchanged=>$listunchanged);
491 if (C4::Context->userenv->{'flags'}==1 ||(C4::Context->userenv->{'flags'} & ( 2**9 ) )){
492 #Edit Catalogue Permissions
493 my $editable_subfields = GetManagedTagSubfields();
494 # change '--' to '—' to avoid escaping issues
495 for (my $i = 0; $i <= $#{$editable_subfields}; $i++) {
496 $editable_subfields->[$i]->{subfielddesc} =~ s/--/—/g;
497 $editable_subfields->[$i]->{tagdesc} =~ s/--/—/g;
499 $template->param(bulkedit => 1);
500 $template->param(tagsubfields=>$editable_subfields);
504 # At this point, each server has given us a result set
505 # now we build that set for template display
506 my @sup_results_array;
507 for (my $i=0;$i<@servers;$i++) {
508 my $server = $servers[$i];
509 if ($server =~/biblioserver/) { # this is the local bibliographic server
510 $hits = $results_hashref->{$server}->{"hits"};
511 my $page = $cgi->param('page') || 0;
512 my @newresults = searchResults( $query_desc,$hits,$results_per_page,$offset,@{$results_hashref->{$server}->{"RECORDS"}});
513 $total = $total + $results_hashref->{$server}->{"hits"};
514 ## If there's just one result, redirect to the detail page
516 my $biblionumber=@newresults[0]->{biblionumber};
517 if (C4::Context->preference('IntranetBiblioDefaultView') eq 'isbd') {
518 print $cgi->redirect("/cgi-bin/koha/catalogue/ISBDdetail.pl?biblionumber=$biblionumber");
519 } elsif (C4::Context->preference('IntranetBiblioDefaultView') eq 'marc') {
520 print $cgi->redirect("/cgi-bin/koha/catalogue/MARCdetail.pl?biblionumber=$biblionumber");
522 print $cgi->redirect("/cgi-bin/koha/catalogue/detail.pl?biblionumber=$biblionumber");
527 $template->param(total => $hits);
528 my $limit_cgi_not_availablity = $limit_cgi;
529 $limit_cgi_not_availablity =~ s/&limit=available//g;
530 $template->param(limit_cgi_not_availablity => $limit_cgi_not_availablity);
531 $template->param(limit_cgi => $limit_cgi);
532 $template->param(query_cgi => $query_cgi);
533 $template->param(query_desc => $query_desc);
534 $template->param(limit_desc => $limit_desc);
535 if ($query_desc || $limit_desc) {
536 $template->param(searchdesc => 1);
538 $template->param(stopwords_removed => "@$stopwords_removed") if $stopwords_removed;
539 $template->param(results_per_page => $results_per_page);
540 $template->param(SEARCH_RESULTS => \@newresults);
542 ## FIXME: add a global function for this, it's better than the current global one
543 ## Build the page numbers on the bottom of the page
545 # total number of pages there will be
546 my $pages = ceil($hits / $results_per_page);
547 # default page number
548 my $current_page_number = 1;
549 $current_page_number = ($offset / $results_per_page + 1) if $offset;
550 my $previous_page_offset = $offset - $results_per_page unless ($offset - $results_per_page <0);
551 my $next_page_offset = $offset + $results_per_page;
552 # If we're within the first 10 pages, keep it simple
553 #warn "current page:".$current_page_number;
554 if ($current_page_number < 10) {
555 # just show the first 10 pages
556 # Loop through the pages
557 my $pages_to_show = 10;
558 $pages_to_show = $pages if $pages<10;
559 for (my $i=1; $i<=$pages_to_show;$i++) {
560 # the offset for this page
561 my $this_offset = (($i*$results_per_page)-$results_per_page);
562 # the page number for this page
563 my $this_page_number = $i;
564 # it should only be highlighted if it's the current page
565 my $highlight = 1 if ($this_page_number == $current_page_number);
566 # put it in the array
567 push @page_numbers, { offset => $this_offset, pg => $this_page_number, highlight => $highlight, sort_by => join " ",@sort_by };
572 # now, show twenty pages, with the current one smack in the middle
574 for (my $i=$current_page_number; $i<=($current_page_number + 20 );$i++) {
575 my $this_offset = ((($i-9)*$results_per_page)-$results_per_page);
576 my $this_page_number = $i-9;
577 my $highlight = 1 if ($this_page_number == $current_page_number);
578 if ($this_page_number <= $pages) {
579 push @page_numbers, { offset => $this_offset, pg => $this_page_number, highlight => $highlight, sort_by => join " ",@sort_by };
584 # FIXME: no previous_page_offset when pages < 2
585 $template->param( PAGE_NUMBERS => \@page_numbers,
586 previous_page_offset => $previous_page_offset) unless $pages < 2;
587 $template->param( next_page_offset => $next_page_offset) unless $pages eq $current_page_number;
591 $template->param(searchdesc => 1,query_desc => $query_desc,limit_desc => $limit_desc);
593 } # end of the if local
595 # asynchronously search the authority server
596 elsif ($server =~/authorityserver/) { # this is the local authority server
597 my @inner_sup_results_array;
598 for my $sup_record ( @{$results_hashref->{$server}->{"RECORDS"}} ) {
599 my $marc_record_object = MARC::Record->new_from_usmarc($sup_record);
600 # warn "Authority Found: ".$marc_record_object->as_formatted();
601 push @inner_sup_results_array, {
602 'title' => $marc_record_object->field(100)->subfield('a'),
603 'link' => "&idx=an&q=".$marc_record_object->field('001')->as_string(),
606 my $servername = $server;
607 push @sup_results_array, { servername => $servername,
608 inner_sup_results_loop => \@inner_sup_results_array} if @inner_sup_results_array;
610 # FIXME: can add support for other targets as needed here
611 $template->param( outer_sup_results_loop => \@sup_results_array);
612 } #/end of the for loop
613 #$template->param(FEDERATED_RESULTS => \@results_array);
617 #classlist => $classlist,
620 facets_loop => $facets,
622 search_error => $error,
625 if ($query_desc || $limit_desc) {
626 $template->param(searchdesc => 1);
629 ## Now let's find out if we have any supplemental data to show the user
630 # and in the meantime, save the current query for statistical purposes, etc.
631 my $koha_spsuggest; # a flag to tell if we've got suggestions coming from Koha
632 my @koha_spsuggest; # place we store the suggestions to be returned to the template as LOOP
633 my $phrases = $query_desc;
636 if ( C4::Context->preference("kohaspsuggest") ) {
637 my ($suggest_host, $suggest_dbname, $suggest_user, $suggest_pwd) = split(':', C4::Context->preference("kohaspsuggest"));
639 my $koha_spsuggest_dbh;
640 # FIXME: this needs to be moved to Context.pm
642 $koha_spsuggest_dbh=DBI->connect("DBI:mysql:$suggest_dbname:$suggest_host","$suggest_user","$suggest_pwd");
645 warn "can't connect to spsuggest db";
648 my $koha_spsuggest_insert = "INSERT INTO phrase_log(phr_phrase,phr_resultcount,phr_ip) VALUES(?,?,?)";
649 my $koha_spsuggest_query = "SELECT display FROM distincts WHERE strcmp(soundex(suggestion), soundex(?)) = 0 order by soundex(suggestion) limit 0,5";
650 my $koha_spsuggest_sth = $koha_spsuggest_dbh->prepare($koha_spsuggest_query);
651 $koha_spsuggest_sth->execute($phrases);
652 while (my $spsuggestion = $koha_spsuggest_sth->fetchrow_array) {
653 $spsuggestion =~ s/(:|\/)//g;
655 $line{spsuggestion} = $spsuggestion;
656 push @koha_spsuggest,\%line;
660 # Now save the current query
661 $koha_spsuggest_sth=$koha_spsuggest_dbh->prepare($koha_spsuggest_insert);
662 #$koha_spsuggest_sth->execute($phrases,$results_per_page,$ipaddress);
663 $koha_spsuggest_sth->finish;
665 $template->param( koha_spsuggest => $koha_spsuggest ) unless $hits;
666 $template->param( SPELL_SUGGEST => \@koha_spsuggest,
671 warn "Kohaspsuggest failure:".$@;
675 # VI. BUILD THE TEMPLATE
676 output_html_with_http_headers $cgi, $cookie, $template->output;