Changeset 6284

Timestamp:

06/05/15 15:05:33 (9 years ago)

Author:

Twan Goosen

Message:

Reintegrated vlo-ticket761 feature branch
Fixes #761

Location:

vlo/trunk

Files:

• . (modified) (1 prop)
• UPGRADE.txt (modified) (1 prop)
• vlo-commons/src/main/resources/VloConfig.xml (modified) (2 diffs)
• vlo-commons/src/test/java/eu/clarin/cmdi/vlo/config/DefaultVloConfigFactoryTest.java (modified) (1 diff)
• vlo-solr/src/main/solr_conf/solr/collection1/conf/schema.xml (modified) (1 diff)
• vlo-solr/src/main/solr_conf/solr/collection1/conf/solrconfig.xml (modified) (3 diffs)
• vlo-solr/src/test (copied) (copied from vlo/branches/vlo-ticket761/vlo-solr/src/test)
• vlo-web-app/src/main/java/eu/clarin/cmdi/vlo/config/VloSolrSpringConfig.java (modified) (1 diff)
• vlo-web-app/src/main/java/eu/clarin/cmdi/vlo/service/solr/impl/AbstractSolrQueryFactory.java (modified) (2 diffs)
• vlo-web-app/src/main/java/eu/clarin/cmdi/vlo/service/solr/impl/SolrDocumentQueryFactoryImpl.java (modified) (1 diff)
• vlo-web-app/src/test/java/eu/clarin/cmdi/vlo/service/solr/impl/SearchResultsDaoImplTest.java (modified) (2 diffs)
• vlo-web-app/src/test/java/eu/clarin/cmdi/vlo/service/solr/impl/SolrDocumentQueryFactoryImplTest.java (modified) (1 diff)
• vlo-web-app/src/test/java/eu/clarin/cmdi/vlo/service/solr/impl/SolrFacetQueryFactoryImplTest.java (modified) (4 diffs)
• vlo-web-app/src/test/resources/solr/collection1/conf/solrconfig.xml (modified) (5 diffs)

vlo/trunk/vlo-commons/src/main/resources/VloConfig.xml

@@ -129 +129 @@
         <technicalField>_lastSeen</technicalField>
         <technicalField>_componentProfile</technicalField>
+        <technicalField>_hierarchyWeight</technicalField>
+        <technicalField>_hasPart</technicalField>
+        <technicalField>_isPartOf</technicalField>
     </technicalFields>
     
@@ -160 +163 @@
     
     <!-- makes the score for the result ranking visible in the search results -->
-    <showResultScores>false</showResultScores>
+    <showResultScores>true</showResultScores>
 </VloConfig>

vlo/trunk/vlo-commons/src/test/java/eu/clarin/cmdi/vlo/config/DefaultVloConfigFactoryTest.java

@@ -928 +928 @@
     public void testGetTechnicalFields() {
         Set<String> result = config.getTechnicalFields();
-        assertEquals(9, result.size());
+        assertEquals(12, result.size());
     }
 

vlo/trunk/vlo-solr/src/main/solr_conf/solr/collection1/conf/schema.xml

@@ -416 +416 @@
    <field name="id" type="string" indexed="true" stored="true" required="true"/>
    <field name="collection" type="string" indexed="true" stored="true" multiValued="true" />
-   <field name="name" type="string" indexed="true" stored="true" multiValued="false"/>
+   <field name="name" type="text" indexed="true" stored="true" multiValued="false"/>
    <field name="projectName" type="string" indexed="true" stored="true" multiValued="true"/>
    <field name="continent" type="string" indexed="true" stored="true" multiValued="true"/>

vlo/trunk/vlo-solr/src/main/solr_conf/solr/collection1/conf/solrconfig.xml

@@ -135 +135 @@
     </listener>
     -->
-
+    <updateLog/>
   </updateHandler>
   
@@ -363 +363 @@
     <!-- default values for query parameters -->
      <lst name="defaults">
-       <str name="echoParams">explicit</str>
-       <!--
-       <int name="rows">10</int>
-       <str name="fl">*</str>
-       <str name="version">2.1</str>
-        -->
+        <str name="echoParams">explicit</str>
+        
+        <!-- See https://wiki.apache.org/solr/DisMaxQParserPlugin -->
+        <str name="defType">dismax</str>
+        
+        <!-- 'Fetch all documents' query -->
+        <str name="q.alt">*:*</str>
+        
+        <!-- Query fields -->
+        <str name="qf">
+            <!-- TODO: Add _languagename field for search purposes? -->
+           name^8 description^4 keywords^2 languageCode^2 country^2 organisation^2 subject^2 collection^1 modality^1 genre^1 continent^.5 id^.1
+        </str>
+        
+        <!-- Phrase fields (boost if phrase is matched) -->
+        <str name="pf">
+            name^2 description
+        </str>
+        
+        <!-- Boost fields (boost based on field value regardless of query) -->
+        <str name="bf">
+            <!-- boost by hierarchy weight (reverse order because lower numbers indicate a higher level -->
+            rord(_hierarchyWeight)
+           <!-- TODO: use bf to boost records with one or more resources? -->
+        </str>
+        
+          <!--
+          <int name="rows">10</int>
+          <str name="fl">*</str>
+          <str name="version">2.1</str>
+           -->
      </lst>
   </requestHandler>
@@ -905 +930 @@
     <queryResponseWriter name="phps" class="org.apache.solr.request.PHPSerializedResponseWriter"/>
 
-    <queryResponseWriter name="custom" class="com.example.MyResponseWriter"/>
-  -->
-
-  <!-- XSLT response writer transforms the XML output by any xslt file found
-       in Solr's conf/xslt directory.  Changes to xslt files are checked for
-       every xsltCacheLifetimeSeconds.  
-   -->
-  <queryResponseWriter name="xslt" class="org.apache.solr.response.XSLTResponseWriter">
-    <int name="xsltCacheLifetimeSeconds">5</int>
-  </queryResponseWriter>
-
-
-  <!-- example of registering a query parser
-  <queryParser name="lucene" class="org.apache.solr.search.LuceneQParserPlugin"/>
-  -->
-
-  <!-- example of registering a custom function parser 
-  <valueSourceParser name="myfunc" class="com.mycompany.MyValueSourceParser" />
-  -->
-
-  <!-- config for the admin interface -->
-  <admin>
-    <defaultQuery>solr</defaultQuery>
-
-    <!-- configure a healthcheck file for servers behind a loadbalancer
-    <healthcheck type="file">server-enabled</healthcheck>
-    -->
-  </admin>
+      <queryResponseWriter name="custom" class="com.example.MyResponseWriter"/>
+    -->
+
+    <!-- XSLT response writer transforms the XML output by any xslt file found
+        in Solr's conf/xslt directory.  Changes to xslt files are checked for
+        every xsltCacheLifetimeSeconds.  
+    -->
+    <queryResponseWriter name="xslt" class="org.apache.solr.response.XSLTResponseWriter">
+        <int name="xsltCacheLifetimeSeconds">5</int>
+    </queryResponseWriter>
+
+
+    <!-- example of registering a query parser
+    <queryParser name="lucene" class="org.apache.solr.search.LuceneQParserPlugin"/>
+    -->
+
+    <!-- example of registering a custom function parser 
+    <valueSourceParser name="myfunc" class="com.mycompany.MyValueSourceParser" />
+    -->
+
+    <!-- config for the admin interface -->
+    <admin>
+        <defaultQuery>solr</defaultQuery>
+
+        <!-- configure a healthcheck file for servers behind a loadbalancer
+        <healthcheck type="file">server-enabled</healthcheck>
+        -->
+    </admin>
 
 </config>

vlo/trunk/vlo-web-app/src/main/java/eu/clarin/cmdi/vlo/config/VloSolrSpringConfig.java

@@ -119 +119 @@
             FacetConstants.FIELD_CLARIN_PROFILE,
             FacetConstants.FIELD_COMPLETE_METADATA,
+            FacetConstants.FIELD_HIERARCHY_WEIGHT,
+            FacetConstants.FIELD_HAS_PART,
+            FacetConstants.FIELD_IS_PART_OF,
             FacetConstants.FIELD_SOLR_SCORE
     );

vlo/trunk/vlo-web-app/src/main/java/eu/clarin/cmdi/vlo/service/solr/impl/AbstractSolrQueryFactory.java

@@ -32 +32 @@
 public abstract class AbstractSolrQueryFactory {
 
-    protected static final String SOLR_SEARCH_ALL = "*:*";
+    protected static final String SOLR_SEARCH_ALL = null;
 
     protected final void addQueryFacetParameters(final SolrQuery query, QueryFacetsSelection queryFacetsSelections) {
@@ -40 +40 @@
         } else {
             // escape query content and wrap in quotes to make literal query
-            query.setQuery("\"" + ClientUtils.escapeQueryChars(queryString) + "\"");
+            query.setQuery(queryString);
         }
         final Map<String, FacetSelection> selections = queryFacetsSelections.getSelection();

vlo/trunk/vlo-web-app/src/main/java/eu/clarin/cmdi/vlo/service/solr/impl/SolrDocumentQueryFactoryImpl.java

@@ -42 +42 @@
         defaultQueryTemplate = new SolrQuery();
         defaultQueryTemplate.setFields(documentFields.toArray(new String[]{}));
+//        //TODO: qf (all fields with weights - make configurable (later)
+//        defaultQueryTemplate.setParam(DisMaxParams.QF, "name^20 description^10");
     }
 

vlo/trunk/vlo-web-app/src/test/java/eu/clarin/cmdi/vlo/service/solr/impl/SearchResultsDaoImplTest.java

@@ -105 +105 @@
         query.setStart(0);
         query.setFields(FIELD_NAME, FIELD_ID, FIELD_DESCRIPTION);
-        query.setQuery("*:*");
+        query.setQuery(null);
 
         query.setFacet(true);
@@ -133 +133 @@
         query.setStart(0);
         query.setFields(FIELD_NAME, FIELD_ID, FIELD_DESCRIPTION);
-        query.setQuery("*:*");
         {
             // all documents should match this

vlo/trunk/vlo-web-app/src/test/java/eu/clarin/cmdi/vlo/service/solr/impl/SolrDocumentQueryFactoryImplTest.java

@@ -66 +66 @@
         assertTrue(fields.contains("field2"));
         assertTrue(fields.contains("field3"));
-        assertEquals("\"query\"", query.getQuery());
+        assertEquals("query", query.getQuery());
     }
 

vlo/trunk/vlo-web-app/src/test/java/eu/clarin/cmdi/vlo/service/solr/impl/SolrFacetQueryFactoryImplTest.java

@@ -58 +58 @@
 
         // default: query selects all values 
-        assertEquals("*:*", query.getQuery());
+        assertNull(query.getQuery());
 
         // no selection -> no filter queries
@@ -82 +82 @@
 
         // default: query selects all values 
-        assertEquals("*:*", query.getQuery());
+        assertNull(query.getQuery());
 
         // Only empty selections -> no filter queries
@@ -107 +107 @@
 
         // default: query selects all values 
-        assertEquals("*:*", query.getQuery());
+        assertNull(query.getQuery());
 
         // Expecting three filter queries as three values have been selected in total
@@ -133 +133 @@
 
         assertEquals(1, query.getFilterQueries().length);
-        assertEquals("\"query\\ string\"", query.getQuery()); //space should be escaped, and query wrapped!
+        assertEquals("query string", query.getQuery());
 
         // Expecting three filter queries as three values have been selected in total

vlo/trunk/vlo-web-app/src/test/resources/solr/collection1/conf/solrconfig.xml

@@ -72 +72 @@
        If replication is in use, this should match the replication configuration. -->
 <!--   <dataDir></dataDir>-->
-
+       
   <!--  Enables JMX if and only if an existing MBeanServer is found, use this
     if you want to configure JMX through JVM parameters. Remove this to disable
@@ -359 +359 @@
      <lst name="defaults">
        <str name="echoParams">explicit</str>
+     <str name="defType">edismax</str>
+     <str name="q.alt">*:*</str>
        <!--
        <int name="rows">10</int>
@@ -401 +403 @@
      <str name="echoParams">explicit</str>
      <float name="tie">0.01</float>
-     <str name="qf">
+<!--     <str name="qf">
         text^0.5 features^1.0 name^1.2 sku^1.5 id^10.0 manu^1.1 cat^1.4
+     </str>-->
+     <str name="pf">
+        name^4 description^2
      </str>
-     <str name="pf">
-        text^0.2 features^1.1 name^1.5 manu^1.4 manu_exact^1.9
-     </str>
+     <!--
      <str name="bf">
         popularity^0.5 recip(price,1,1000,1000)^0.3
@@ -416 +419 @@
         2&lt;-1 5&lt;-2 6&lt;90%
      </str>
-     <int name="ps">100</int>
+     <int name="ps">100</int>-->
      <str name="q.alt">*:*</str>
-     <!-- example highlighter config, enable per-query with hl=true -->
+<!--      example highlighter config, enable per-query with hl=true 
      <str name="hl.fl">text features name</str>
-     <!-- for this field, we want no fragmenting, just highlighting -->
+      for this field, we want no fragmenting, just highlighting 
      <str name="f.name.hl.fragsize">0</str>
-     <!-- instructs Solr to return the field itself if no query terms are
-          found -->
+      instructs Solr to return the field itself if no query terms are
+          found 
      <str name="f.name.hl.alternateField">name</str>
-     <str name="f.text.hl.fragmenter">regex</str> <!-- defined below -->
+     <str name="f.text.hl.fragmenter">regex</str>  defined below -->
     </lst>
   </requestHandler>
@@ -907 +910 @@
        every xsltCacheLifetimeSeconds.  
    -->
-<!--  <queryResponseWriter name="xslt" class="org.apache.solr.response.XSLTResponseWriter">
+  <queryResponseWriter name="xslt" class="org.apache.solr.response.XSLTResponseWriter">
     <int name="xsltCacheLifetimeSeconds">5</int>
-  </queryResponseWriter>-->
+  </queryResponseWriter>