Add EdgeNGram indexing on id and full_title_tesim for autocomplete. F…

…ixes #293
projectblacklight · Feb 26, 2014 · 00a16d0 · 00a16d0
1 parent 3a451b9
commit 00a16d0
Show file tree

Hide file tree

Showing 3 changed files with 36 additions and 1 deletion.
diff --git a/solr_conf/conf/schema.xml b/solr_conf/conf/schema.xml
@@ -15,6 +15,9 @@
     <field name="lng" type="tdouble" stored="true" indexed="true" multiValued="false"/>
     <field name="original_pid_tesim" type="pid_text" stored="true" indexed="true" multiValued="true"/>
 
+    <field name="full_title_tesim" type="text_en_ng" stored="true" indexed="true" multiValued="true"/>
+    <field name="id_tesim" type="text_en_ng" stored="true" indexed="true" multiValued="false"/>
+
     <!-- NOTE:  not all possible Solr field types are represented in the dynamic fields -->
 
     <!-- text (_t...) -->
@@ -198,6 +201,7 @@
       destination field is to use the dynamic field syntax.
       copyField also supports a maxChars to copy setting.  -->
 
+  <copyField source="id" dest="id_tesim" maxChars="3000"/>
   <copyField source="*_tesim" dest="all_text_timv" maxChars="3000"/>
 
   <types>
@@ -288,6 +292,28 @@
       </analyzer>
     </fieldType>
 
+    <!-- A text field with defaults appropriate for English an NGrams -->
+    <fieldType name="text_en_ng" class="solr.TextField" positionIncrementGap="100">
+      <analyzer type="index">
+        <tokenizer class="solr.ICUTokenizerFactory"/>
+        <filter class="solr.ICUFoldingFilterFactory"/>  <!-- NFKC, case folding, diacritics removed -->
+        <filter class="solr.EnglishPossessiveFilterFactory"/>
+        <!-- EnglishMinimalStemFilterFactory is less aggressive than PorterStemFilterFactory: -->
+        <filter class="solr.EnglishMinimalStemFilterFactory"/>
+        <filter class="solr.TrimFilterFactory"/>
+        <filter class="solr.EdgeNGramFilterFactory" minGramSize="3" maxGramSize="15" side="front"/>:
+      </analyzer>
+
+      <analyzer type="index">
+        <tokenizer class="solr.ICUTokenizerFactory"/>
+        <filter class="solr.ICUFoldingFilterFactory"/>  <!-- NFKC, case folding, diacritics removed -->
+        <filter class="solr.EnglishPossessiveFilterFactory"/>
+        <!-- EnglishMinimalStemFilterFactory is less aggressive than PorterStemFilterFactory: -->
+        <filter class="solr.EnglishMinimalStemFilterFactory"/>
+        <filter class="solr.TrimFilterFactory"/>
+      </analyzer>
+    </fieldType>
+
     <fieldType name="pid_text" class="solr.TextField" positionIncrementGap="100">
       <analyzer>
         <tokenizer class="solr.KeywordTokenizerFactory"/>

diff --git a/solr_conf/conf/solrconfig.xml b/solr_conf/conf/solrconfig.xml
@@ -42,7 +42,7 @@
             http://wiki.apache.org/solr/LocalParams
        -->
         <str name="qf">
-          id
+          id_tesim
           full_title_tesim
           short_title_tesim
           alternative_title_tesim

diff --git a/spec/controllers/spotlight/catalog_controller_spec.rb b/spec/controllers/spotlight/catalog_controller_spec.rb
@@ -57,6 +57,15 @@
         get :index, exhibit_id: exhibit, q: 'map'
         expect(controller.blacklight_config.show.partials.first).to eq "curation_mode_toggle"
       end
+
+      it "should have partial matches for title" do
+        get :index, exhibit_id: exhibit, q: 'ZUYDLA'
+        expect(assigns[:document_list].first.id).to eq 'dx157dh4345'
+      end
+      it "should have partial matches for id" do
+        get :index, exhibit_id: exhibit, q: 'dx157'
+        expect(assigns[:document_list].first.id).to eq 'dx157dh4345'
+      end
     end
   end