From d099dc13bb4b9aa991caa0feaf24bed59c3c4844 Mon Sep 17 00:00:00 2001
From: Nick Clemens <nick@bywatersolutions.com>
Date: Fri, 6 Oct 2017 15:37:38 +0000
Subject: [PATCH] Bug 18374: (follow-up) Add support and tests for quoted
 strings

To test:
1 - prove t/db_dependent/Koha_SearchEngine_Elasticsearch_Search.t
2 - do some searches in staff client and test results

Signed-off-by: Julian Maurice <julian.maurice@biblibre.com>

Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
---
 .../Elasticsearch/QueryBuilder.pm             | 26 ++++++++++---
 .../Koha_SearchEngine_Elasticsearch_Search.t  | 38 ++++++++++++++++++-
 2 files changed, 57 insertions(+), 7 deletions(-)

diff --git a/Koha/SearchEngine/Elasticsearch/QueryBuilder.pm b/Koha/SearchEngine/Elasticsearch/QueryBuilder.pm
index 5635c85064..d7a83e272e 100644
--- a/Koha/SearchEngine/Elasticsearch/QueryBuilder.pm
+++ b/Koha/SearchEngine/Elasticsearch/QueryBuilder.pm
@@ -788,7 +788,7 @@ sub _sort_field {
     my $query = $self->_truncate_terms($query);
 
 Given a string query this function appends '*' wildcard  to all terms except
-operands.
+operands and double quoted strings.
 
 =cut
 
@@ -796,11 +796,25 @@ sub _truncate_terms {
     my ( $self, $query ) = @_;
     my @stops = qw/and or not/;
     my @new_terms;
-    my @split_query = split /[\(\s\)]/, $query;
-    foreach my $term (@split_query) {
-        next if ( $term eq '' || $term eq ' ' );
-        $term .= "*" unless ( ( grep { lc($term) =~ /^$_$/ } @stops ) || ( $term =~ /\*$/ ) );
-        push @new_terms, $term;
+    my @quote_split = split /(["])([^"]+)\1/, $query; 
+    #Above splits the string based on matching pairs of double quotes
+    #In practice we get ('','"','donald duck',' ','"','the mouse',' and pete') 
+    #given the string '"donald duck" "the mouse" and pete'
+    #so we ignore empties, quote the ones after a '"' and split the rest on spaces
+    for (my $i=0; $i < @quote_split; $i++ ) {
+        next if ( $quote_split[$i] eq '' || $quote_split[$i] eq ' ' );
+        if ( $quote_split[$i] eq '"' ){
+            $i++;
+            $quote_split[$i] = '"'.$quote_split[$i].'"';
+            push @new_terms, $quote_split[$i]
+        } else {
+            my @space_split = split /[\(\s\)]/, $quote_split[$i];
+            foreach my $term (@space_split) {
+                next if ( $term eq '' || $term eq ' ' );
+                $term .= "*" unless ( ( grep { lc($term) =~ /^$_$/ } @stops ) || ( $term =~ /\*$/ ) );
+                push @new_terms, $term;
+            }
+        }
     }
     $query = join ' ', @new_terms;
     return $query;
diff --git a/t/db_dependent/Koha_SearchEngine_Elasticsearch_Search.t b/t/db_dependent/Koha_SearchEngine_Elasticsearch_Search.t
index f547fcbc17..b54a5f0401 100644
--- a/t/db_dependent/Koha_SearchEngine_Elasticsearch_Search.t
+++ b/t/db_dependent/Koha_SearchEngine_Elasticsearch_Search.t
@@ -81,7 +81,7 @@ subtest 'json2marc' => sub {
 };
 
 subtest 'build_query tests' => sub {
-    plan tests => 10;
+    plan tests => 15;
 
     t::lib::Mocks::mock_preference('DisplayLibraryFacets','both');
     my $query = $builder->build_query();
@@ -135,5 +135,41 @@ subtest 'build_query tests' => sub {
         "(donald* duck*)",
         "query with '*' is unaltered when QueryAutoTruncate is enabled"
     );
+
+    ( undef, $query ) = $builder->build_query_compat( undef, ['donald duck and the mouse'] );
+    is(
+        $query->{query}{query_string}{query},
+        "(donald* duck* and the* mouse*)",
+        "individual words are all truncated and stopwords ignored"
+    );
+
+    ( undef, $query ) = $builder->build_query_compat( undef, ['*'] );
+    is(
+        $query->{query}{query_string}{query},
+        "(*)",
+        "query of just '*' is unaltered when QueryAutoTruncate is enabled"
+    );
+
+    ( undef, $query ) = $builder->build_query_compat( undef, ['"donald duck"'] );
+    is(
+        $query->{query}{query_string}{query},
+        '("donald duck")',
+        "query with quotes is unaltered when QueryAutoTruncate is enabled"
+    );
+
+
+    ( undef, $query ) = $builder->build_query_compat( undef, ['"donald duck" and "the mouse"'] );
+    is(
+        $query->{query}{query_string}{query},
+        '("donald duck" and "the mouse")',
+        "all quoted strings are unaltered if more than one in query"
+    );
+
+    ( undef, $query ) = $builder->build_query_compat( undef, ['barcode:123456'] );
+    is(
+        $query->{query}{query_string}{query},
+        '(barcode:123456*)',
+        "query of specific field is truncated"
+    );
 };
 
-- 
2.39.5