SOLR-2977: add 'fake' excludes

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1222234 13f79535-47bb-0310-9956-ffa450edef68

solr/core/src/java/org/apache/solr/handler/component/QueryElevationComponent.java

@@ -20,16 +20,12 @@ package org.apache.solr.handler.component;
 import org.apache.lucene.analysis.Analyzer;
 import org.apache.lucene.analysis.TokenStream;
 import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
-import org.apache.lucene.document.Document;
-import org.apache.lucene.document.DocumentStoredFieldVisitor;
 import org.apache.lucene.index.*;
 import org.apache.lucene.index.IndexReader.AtomicReaderContext;
 import org.apache.lucene.search.*;
 import org.apache.lucene.util.Bits;
 import org.apache.lucene.util.BytesRef;
 import org.apache.lucene.util.SentinelIntSet;
-import org.apache.lucene.util.automaton.Automaton;
-import org.apache.lucene.util.automaton.CompiledAutomaton;
 import org.apache.solr.cloud.ZkController;
 import org.apache.solr.common.SolrException;
 import org.apache.solr.common.params.QueryElevationParams;
@@ -40,7 +36,8 @@ import org.apache.solr.common.util.SimpleOrderedMap;
 import org.apache.solr.core.Config;
 import org.apache.solr.core.SolrCore;
 import org.apache.solr.request.SolrQueryRequest;
-import org.apache.solr.response.transform.EditorialMarkerFactory;
+import org.apache.solr.response.transform.ElevatedMarkerFactory;
+import org.apache.solr.response.transform.ExcludedMarkerFactory;
 import org.apache.solr.schema.FieldType;
 import org.apache.solr.schema.SchemaField;
 import org.apache.solr.search.SolrIndexSearcher;
@@ -78,6 +75,8 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
   static final String FIELD_TYPE = "queryFieldType";
   static final String CONFIG_FILE = "config-file";
   static final String EXCLUDE = "exclude";
+  public static final String BOOSTED = "BOOSTED";
+  public static final String EXCLUDED = "EXCLUDED";
 
   // Runtime param -- should be in common?
 
@@ -97,15 +96,17 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
   class ElevationObj {
     final String text;
     final String analyzed;
-    final BooleanClause[] exclude;
+    final TermQuery [] exclude;//just keep the term query, b/c we will not always explicitly exclude the item based on markExcludes query time param
     final BooleanQuery include;
     final Map<BytesRef, Integer> priority;
     final Set<String> ids;
+    final Set<String> excludeIds;
 
     ElevationObj(String qstr, List<String> elevate, List<String> exclude) throws IOException {
       this.text = qstr;
       this.analyzed = getAnalyzedQuery(this.text);
       this.ids = new HashSet<String>();
+      this.excludeIds = new HashSet<String>();
 
       this.include = new BooleanQuery();
       this.include.setBoost(0);
@@ -122,10 +123,11 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
       if (exclude == null || exclude.isEmpty()) {
         this.exclude = null;
       } else {
-        this.exclude = new BooleanClause[exclude.size()];
+        this.exclude = new TermQuery[exclude.size()];
         for (int i = 0; i < exclude.size(); i++) {
-          TermQuery tq = new TermQuery(new Term(idField, idSchemaFT.readableToIndexed(exclude.get(i))));
-          this.exclude[i] = new BooleanClause(tq, BooleanClause.Occur.MUST_NOT);
+          String id = idSchemaFT.readableToIndexed(exclude.get(i));
+          excludeIds.add(id);
+          this.exclude[i] = new TermQuery(new Term(idField, id));
         }
       }
     }
@@ -155,12 +157,18 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
     idSchemaFT = sf.getType();
     idField = sf.getName();
     //register the EditorialMarkerFactory
-    EditorialMarkerFactory factory = new EditorialMarkerFactory();
+    String excludeName = initArgs.get(QueryElevationParams.EXCLUDE_MARKER_FIELD_NAME, "excluded");
+    if (excludeName == null || excludeName.equals("") == true){
+      excludeName = "excluded";
+    }
+    ExcludedMarkerFactory excludedMarkerFactory = new ExcludedMarkerFactory();
+    core.addTransformerFactory(excludeName, excludedMarkerFactory);
+    ElevatedMarkerFactory elevatedMarkerFactory = new ElevatedMarkerFactory();
     String markerName = initArgs.get(QueryElevationParams.EDITORIAL_MARKER_FIELD_NAME, "elevated");
     if (markerName == null || markerName.equals("") == true) {
       markerName = "elevated";
     }
-    core.addTransformerFactory(markerName, factory);
+    core.addTransformerFactory(markerName, elevatedMarkerFactory);
     forceElevation = initArgs.getBool(QueryElevationParams.FORCE_ELEVATION, forceElevation);
     try {
       synchronized (elevationCache) {
@@ -333,7 +341,7 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
     boolean exclusive = params.getBool(QueryElevationParams.EXCLUSIVE, false);
     // A runtime parameter can alter the config value for forceElevation
     boolean force = params.getBool(QueryElevationParams.FORCE_ELEVATION, forceElevation);
-
+    boolean markExcludes = params.getBool(QueryElevationParams.MARK_EXCLUDES, false);
     Query query = rb.getQuery();
     String qstr = rb.getQueryString();
     if (query == null || qstr == null) {
@@ -351,7 +359,7 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
     }
 
     if (booster != null) {
-      rb.req.getContext().put("BOOSTED", booster.ids);
+      rb.req.getContext().put(BOOSTED, booster.ids);
 
       // Change the query to insert forced documents
       if (exclusive == true) {
@@ -362,8 +370,17 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
         newq.add(query, BooleanClause.Occur.SHOULD);
         newq.add(booster.include, BooleanClause.Occur.SHOULD);
         if (booster.exclude != null) {
-          for (BooleanClause bq : booster.exclude) {
-            newq.add(bq);
+          if (markExcludes == false) {
+            for (TermQuery tq : booster.exclude) {
+              newq.add(new BooleanClause(tq, BooleanClause.Occur.MUST_NOT));
+            }
+          } else {
+            //we are only going to mark items as excluded, not actually exclude them.  This works
+            //with the EditorialMarkerFactory
+            rb.req.getContext().put(EXCLUDED, booster.excludeIds);
+            for (TermQuery tq : booster.exclude) {
+              newq.add(new BooleanClause(tq, BooleanClause.Occur.SHOULD));
+            }
           }
         }
         rb.setQuery(newq);

solr/core/src/java/org/apache/solr/response/transform/BaseEditorialTransformer.java

@@ -0,0 +1,79 @@
+package org.apache.solr.response.transform;
+
+
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import org.apache.lucene.document.Field;
+import org.apache.lucene.document.NumericField;
+import org.apache.solr.common.SolrDocument;
+import org.apache.solr.handler.component.QueryElevationComponent;
+import org.apache.solr.schema.FieldType;
+
+import java.util.Set;
+
+/**
+ *
+ *
+ **/
+public abstract class BaseEditorialTransformer extends TransformerWithContext {
+
+  final String name;
+  final String idFieldName;
+  final FieldType ft;
+
+  public BaseEditorialTransformer(String name, String idFieldName, FieldType ft) {
+    this.name = name;
+    this.idFieldName = idFieldName;
+    this.ft = ft;
+  }
+
+  @Override
+  public String getName()
+  {
+    return name;
+  }
+
+  @Override
+  public void transform(SolrDocument doc, int docid) {
+    //this only gets added if QueryElevationParams.MARK_EXCLUDED is true
+    Set<String> ids = getIdSet();
+    if (ids != null && ids.isEmpty() == false) {
+      String key = getKey(doc);
+      doc.setField(name, ids.contains(key));
+    } else {
+      //if we have no ids, that means we weren't marking, but the user still asked for the field to be added, so just mark everything as false
+      doc.setField(name, Boolean.FALSE);
+    }
+  }
+
+  protected abstract Set<String> getIdSet();
+
+  protected String getKey(SolrDocument doc) {
+    String key;
+    Object field = doc.get(idFieldName);
+    if (field instanceof NumericField){
+      key = ((Field)field).stringValue();
+      key = ft.readableToIndexed(key);
+    } else if (field instanceof Field){
+      key = ((Field)field).stringValue();
+    } else {
+      key = field.toString();
+    }
+    return key;
+  }
+}

solr/core/src/java/org/apache/solr/response/transform/ElevatedMarkerFactory.java

@@ -28,10 +28,10 @@ import org.apache.solr.schema.FieldType;
 import org.apache.solr.schema.SchemaField;
 
 /**
- *
+ * Used to mark whether a document has been elevated or not
  * @since solr 4.0
  */
-public class EditorialMarkerFactory extends TransformerFactory
+public class ElevatedMarkerFactory extends TransformerFactory
 {
   @Override
   public DocTransformer create(String field, SolrParams params, SolrQueryRequest req) {
@@ -41,45 +41,16 @@ public class EditorialMarkerFactory extends TransformerFactory
   }
 }
 
-class MarkTransformer extends TransformerWithContext
+class MarkTransformer extends BaseEditorialTransformer
 {
-  final String name;
-  final String idFieldName;
-  final FieldType ft;
 
-  public MarkTransformer( String name, String idFieldName, FieldType ft)
-  {
-    this.name = name;
-    this.idFieldName = idFieldName;
-    this.ft = ft;
+  MarkTransformer(String name, String idFieldName, FieldType ft) {
+    super(name, idFieldName, ft);
   }
 
   @Override
-  public String getName()
-  {
-    return name;
-  }
-
-  @Override
-  public void transform(SolrDocument doc, int docid) {
-    Set<?> ids = (Set<?>)context.req.getContext().get("BOOSTED");
-    if(ids!=null) {
-      String key;
-      Object field = doc.get(idFieldName);
-      if (field instanceof NumericField){
-        key = ((Field)field).stringValue();
-        key = ft.readableToIndexed(key);
-      } else if (field instanceof Field){
-        key = ((Field)field).stringValue();
-      } else {
-        key = field.toString();
-      }
-
-      doc.setField(name, ids.contains(key));
-    } else {
-      //if we have no ids, that means we weren't boosting, but the user still asked for the field to be added, so just mark everything as false
-      doc.setField(name, Boolean.FALSE);
-    }
+  protected Set<String> getIdSet() {
+    return (Set<String>) context.req.getContext().get("BOOSTED");
   }
 }
 

solr/core/src/java/org/apache/solr/response/transform/ExcludedMarkerFactory.java

@@ -0,0 +1,58 @@
+package org.apache.solr.response.transform;
+
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements.  See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License.  You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import java.util.Set;
+
+import org.apache.lucene.document.Field;
+import org.apache.lucene.document.NumericField;
+import org.apache.solr.common.SolrDocument;
+import org.apache.solr.common.params.SolrParams;
+import org.apache.solr.handler.component.QueryElevationComponent;
+import org.apache.solr.request.SolrQueryRequest;
+import org.apache.solr.schema.FieldType;
+import org.apache.solr.schema.SchemaField;
+
+/**
+ *
+ * @since solr 4.0
+ */
+public class ExcludedMarkerFactory extends TransformerFactory
+{
+
+  @Override
+  public DocTransformer create(String field, SolrParams params, SolrQueryRequest req) {
+    SchemaField uniqueKeyField = req.getSchema().getUniqueKeyField();
+    String idfield = uniqueKeyField.getName();
+    return new ExcludedTransformer(field,idfield, uniqueKeyField.getType());
+  }
+}
+
+class ExcludedTransformer extends BaseEditorialTransformer {
+
+  public ExcludedTransformer( String name, String idFieldName, FieldType ft)
+  {
+    super(name, idFieldName, ft);
+  }
+
+  protected Set<String> getIdSet() {
+    return (Set<String>)context.req.getContext().get(QueryElevationComponent.EXCLUDED);
+  }
+
+}
+

solr/core/src/test-files/solr/conf/elevate.xml

@@ -37,6 +37,10 @@
   <doc id="7" />
  </query>
 
+ <query text="XXXX XXXX">
+  <doc id="5" />
+  <doc id="6" exclude="true" />
+ </query>
 
 
 </elevate>

solr/core/src/test/org/apache/solr/handler/component/QueryElevationComponentTest.java

@@ -157,7 +157,7 @@ public class QueryElevationComponentTest extends SolrTestCaseJ4 {
       req.close();
 
       // Make sure the boosts loaded properly
-      assertEquals(4, map.size());
+      assertEquals(5, map.size());
       assertEquals(1, map.get("XXXX").priority.size());
       assertEquals(2, map.get("YYYY").priority.size());
       assertEquals(3, map.get("ZZZZ").priority.size());
@@ -174,7 +174,7 @@ public class QueryElevationComponentTest extends SolrTestCaseJ4 {
       comp.init(args);
       comp.inform(core);
       map = comp.getElevationMap(reader, core);
-      assertEquals(4, map.size());
+      assertEquals(5, map.size());
       assertEquals(null, map.get("XXXX"));
       assertEquals(null, map.get("YYYY"));
       assertEquals(null, map.get("ZZZZ"));
@@ -237,6 +237,52 @@ public class QueryElevationComponentTest extends SolrTestCaseJ4 {
     }
   }
 
+  @Test
+  public void testMarkExcludes() throws Exception {
+    try {
+      init("schema12.xml");
+      assertU(adoc("id", "1", "title", "XXXX XXXX", "str_s1", "a"));
+      assertU(adoc("id", "2", "title", "YYYY", "str_s1", "b"));
+      assertU(adoc("id", "3", "title", "ZZZZ", "str_s1", "c"));
+
+      assertU(adoc("id", "4", "title", "XXXX XXXX", "str_s1", "x"));
+      assertU(adoc("id", "5", "title", "YYYY YYYY", "str_s1", "y"));
+      assertU(adoc("id", "6", "title", "XXXX XXXX", "str_s1", "z"));
+      assertU(adoc("id", "7", "title", "AAAA", "str_s1", "a"));
+      assertU(commit());
+
+      assertQ("", req(CommonParams.Q, "XXXX XXXX", CommonParams.QT, "/elevate",
+          QueryElevationParams.MARK_EXCLUDES, "true",
+          CommonParams.FL, "id, score, [excluded]")
+          , "//*[@numFound='4']"
+          , "//result/doc[1]/str[@name='id'][.='5']"
+          , "//result/doc[2]/str[@name='id'][.='6']"
+          , "//result/doc[3]/str[@name='id'][.='1']"
+          , "//result/doc[4]/str[@name='id'][.='4']",
+          "//result/doc[1]/bool[@name='[excluded]'][.='false']",
+          "//result/doc[2]/bool[@name='[excluded]'][.='true']",
+          "//result/doc[3]/bool[@name='[excluded]'][.='false']",
+          "//result/doc[4]/bool[@name='[excluded]'][.='false']"
+      );
+      //ask for excluded as a field, but don't actually request the MARK_EXCLUDES
+      //thus, number 6 should not be returned, b/c it is excluded
+      assertQ("", req(CommonParams.Q, "XXXX XXXX", CommonParams.QT, "/elevate",
+          QueryElevationParams.MARK_EXCLUDES, "false",
+          CommonParams.FL, "id, score, [excluded]")
+          , "//*[@numFound='3']"
+          , "//result/doc[1]/str[@name='id'][.='5']"
+          , "//result/doc[2]/str[@name='id'][.='1']"
+          , "//result/doc[3]/str[@name='id'][.='4']",
+          "//result/doc[1]/bool[@name='[excluded]'][.='false']",
+          "//result/doc[2]/bool[@name='[excluded]'][.='false']",
+          "//result/doc[3]/bool[@name='[excluded]'][.='false']"
+      );
+
+    } finally {
+      delete();
+    }
+  }
+
   @Test
   public void testSorting() throws Exception {
     try {

solr/example/solr/conf/elevate.xml

@@ -19,8 +19,10 @@
 <!-- If this file is found in the config directory, it will only be
      loaded once at startup.  If it is found in Solr's data
      directory, it will be re-loaded every commit.
--->
 
+   See http://wiki.apache.org/solr/QueryElevationComponent for more info
+
+-->
 <elevate>
  <query text="foo bar">
   <doc id="1" />

solr/solrj/src/java/org/apache/solr/common/params/QueryElevationParams.java

@@ -34,4 +34,18 @@ public interface QueryElevationParams {
    * See http://wiki.apache.org/solr/DocTransformers
    */
   String EDITORIAL_MARKER_FIELD_NAME = "editorialMarkerFieldName";
+  /**
+   * The name of the field that excluded editorial results will be written out as when using the QueryElevationComponent, which
+   * automatically configures the EditorialMarkerFactory.  The default name is "excluded".  This is only used
+   * when {@link #MARK_EXCLUDES} is set to true at query time.
+   * <br/>
+   * See http://wiki.apache.org/solr/DocTransformers
+   */
+  String EXCLUDE_MARKER_FIELD_NAME = "excludeMarkerFieldName";
+
+  /**
+   * Instead of removing excluded items from the results, passing in this parameter allows you to get back the excluded items, but to mark them
+   * as excluded.
+   */
+  String MARK_EXCLUDES = "markExcludes";
 }