Apache
/
lucene
mirror of https://github.com/apache/lucene.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

SOLR-2037: hook in DocTransformer for QueryElevationComponent to mark documents that have been boosted 

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1203839 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Grant Ingersoll 2011-11-18 21:07:13 +00:00
parent 5267d23d46
commit 00f5fef162
16 changed files with 199 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
solr/core/src/java/org/apache/solr/core/SolrCore.java
View File

@ -1705,6 +1705,10 @@ public final class SolrCore implements SolrInfoMBean {
public TransformerFactory getTransformerFactory(String name) {
return transformerFactories.get(name);
}
public void addTransformerFactory(String name, TransformerFactory factory){
transformerFactories.put(name, factory);
}
/**

21
solr/core/src/java/org/apache/solr/handler/component/QueryElevationComponent.java
View File

@ -23,14 +23,10 @@ import java.io.InputStream;
import java.io.StringReader;
import java.net.MalformedURLException;
import java.net.URL;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.WeakHashMap;
import java.util.*;
import org.apache.solr.common.params.QueryElevationParams;
import org.apache.solr.response.transform.EditorialMarkerFactory;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
@ -103,6 +99,7 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
final BooleanClause[] exclude;
final BooleanQuery include;
final Map<BytesRef,Integer> priority;
final Set<String> ids;
// use singletons so hashCode/equals on Sort will just work
final FieldComparatorSource comparatorSource;
@ -111,12 +108,14 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
{
this.text = qstr;
this.analyzed = getAnalyzedQuery( this.text );
this.ids = new HashSet<String>();
this.include = new BooleanQuery();
this.include.setBoost( 0 );
this.priority = new HashMap<BytesRef, Integer>();
int max = elevate.size()+5;
for( String id : elevate ) {
ids.add(id);
TermQuery tq = new TermQuery( new Term( idField, id ) );
include.add( tq, BooleanClause.Occur.SHOULD );
this.priority.put( new BytesRef(id), max-- );
@ -161,7 +160,13 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
"QueryElevationComponent requires the schema to have a uniqueKeyField implemented using StrField" );
}
idField = sf.getName();
//register the EditorialMarkerFactory
EditorialMarkerFactory factory = new EditorialMarkerFactory();
String markerName = initArgs.get(QueryElevationParams.EDITORIAL_MARKER_FIELD_NAME, "elevated");
if (markerName == null || markerName.equals("") == true){
markerName = "elevated";
}
core.addTransformerFactory(markerName, factory);
forceElevation = initArgs.getBool( QueryElevationParams.FORCE_ELEVATION, forceElevation );
try {
synchronized( elevationCache ) {
@ -357,6 +362,8 @@ public class QueryElevationComponent extends SearchComponent implements SolrCore
}
if( booster != null ) {
rb.req.getContext().put("BOOSTED", booster.ids);
// Change the query to insert forced documents
if (exclusive == true){
//we only want these results

7
solr/core/src/java/org/apache/solr/handler/component/RealTimeGetComponent.java
View File

@ -31,6 +31,7 @@ import org.apache.solr.core.SolrCore;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.response.SolrQueryResponse;
import org.apache.solr.response.transform.DocTransformer;
import org.apache.solr.response.transform.TransformContext;
import org.apache.solr.schema.FieldType;
import org.apache.solr.schema.IndexSchema;
import org.apache.solr.schema.SchemaField;
@ -40,6 +41,7 @@ import org.apache.solr.update.DocumentBuilder;
import org.apache.solr.update.UpdateLog;
import org.apache.solr.util.RefCounted;
import javax.xml.transform.Transformer;
import java.io.IOException;
import java.net.URL;
import java.util.ArrayList;
@ -103,6 +105,11 @@ public class RealTimeGetComponent extends SearchComponent
RefCounted<SolrIndexSearcher> searcherHolder = null;
DocTransformer transformer = rsp.getReturnFields().getTransformer();
if (transformer != null) {
TransformContext context = new TransformContext();
context.req = req;
transformer.setContext(context);
}
try {
SolrIndexSearcher searcher = null;

16
solr/core/src/java/org/apache/solr/response/BinaryResponseWriter.java
View File

@ -119,17 +119,17 @@ public class BinaryResponseWriter implements BinaryQueryResponseWriter {
protected void writeResultsBody( ResultContext res, JavaBinCodec codec ) throws IOException
{
DocList ids = res.docs;
TransformContext context = new TransformContext();
context.query = res.query;
context.wantsScores = returnFields.wantsScore() && ids.hasScores();
int sz = ids.size();
codec.writeTag(JavaBinCodec.ARR, sz);
if(searcher == null) searcher = solrQueryRequest.getSearcher();
if(schema == null) schema = solrQueryRequest.getSchema();
context.searcher = searcher;
if(schema == null) schema = solrQueryRequest.getSchema();
DocTransformer transformer = returnFields.getTransformer();
TransformContext context = new TransformContext();
context.query = res.query;
context.wantsScores = returnFields.wantsScore() && ids.hasScores();
context.req = solrQueryRequest;
context.searcher = searcher;
if( transformer != null ) {
transformer.setContext( context );
}
@ -141,7 +141,7 @@ public class BinaryResponseWriter implements BinaryQueryResponseWriter {
Document doc = searcher.doc(id, fnames);
SolrDocument sdoc = getDoc(doc);
if( transformer != null ) {
transformer.transform(sdoc, id );
transformer.transform(sdoc, id);
}
codec.writeSolrDocument(sdoc);
}

11
solr/core/src/java/org/apache/solr/response/TextResponseWriter.java
View File

@ -146,8 +146,12 @@ public abstract class TextResponseWriter {
writeDouble(name, ((Double)val).doubleValue());
} else if (val instanceof Document) {
SolrDocument doc = toSolrDocument( (Document)val );
if( returnFields.getTransformer() != null ) {
returnFields.getTransformer().transform( doc, -1 );
DocTransformer transformer = returnFields.getTransformer();
if( transformer != null ) {
TransformContext context = new TransformContext();
context.req = req;
transformer.setContext(context);
transformer.transform(doc, -1);
}
writeSolrDocument(name, doc, returnFields, 0 );
} else if (val instanceof SolrDocument) {
@ -232,6 +236,7 @@ public abstract class TextResponseWriter {
TransformContext context = new TransformContext();
context.query = res.query;
context.wantsScores = fields.wantsScore() && ids.hasScores();
context.req = req;
writeStartDocumentList(name, ids.offset(), ids.size(), ids.matches(),
context.wantsScores ? new Float(ids.maxScore()) : null );
@ -248,7 +253,7 @@ public abstract class TextResponseWriter {
Document doc = context.searcher.doc(id, fnames);
SolrDocument sdoc = toSolrDocument( doc );
if( transformer != null ) {
transformer.transform( sdoc, id );
transformer.transform( sdoc, id);
}
writeSolrDocument( null, sdoc, returnFields, i );
}

25
solr/core/src/java/org/apache/solr/response/transform/DocTransformer.java
View File

@ -20,16 +20,41 @@ package org.apache.solr.response.transform;
import java.io.IOException;
import org.apache.solr.common.SolrDocument;
import org.apache.solr.request.SolrQueryRequest;
/**
* A DocTransformer can add, remove or alter a Document before it is written out to the Response. For instance, there are implementations
* that can put explanations inline with a document, add constant values and mark items as being artificially boosted (see {@link org.apache.solr.handler.component.QueryElevationComponent})
*
* <p/>
* New instance for each request
*
* @see TransformerFactory
*
*/
public abstract class DocTransformer
{
/**
*
* @return The name of the transformer
*/
public abstract String getName();
/**
* This is called before transform and sets
* @param context The {@link org.apache.solr.response.transform.TransformContext} stores information about the current state of things in Solr that may be
* useful for doing transformations.
*/
public void setContext( TransformContext context ) {}
/**
* This is where implementations do the actual work
*
*
* @param doc The document to alter
* @param docid The Lucene internal doc id
* @throws IOException
*/
public abstract void transform(SolrDocument doc, int docid) throws IOException;
@Override

3
solr/core/src/java/org/apache/solr/response/transform/DocTransformers.java
View File

@ -23,6 +23,7 @@ import java.util.Iterator;
import java.util.List;
import org.apache.solr.common.SolrDocument;
import org.apache.solr.request.SolrQueryRequest;
/**
* Transform a document before it gets sent out
@ -73,7 +74,7 @@ public class DocTransformers extends DocTransformer
@Override
public void transform(SolrDocument doc, int docid) throws IOException {
for( DocTransformer a : children ) {
a.transform( doc, docid );
a.transform( doc, docid);
}
}
}

75
solr/core/src/java/org/apache/solr/response/transform/EditorialMarkerFactory.java Normal file
View File

@ -0,0 +1,75 @@
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.solr.common.SolrDocument;
import org.apache.solr.common.params.SolrParams;
import org.apache.solr.request.SolrQueryRequest;
/**
*
* @since solr 4.0
*/
public class EditorialMarkerFactory extends TransformerFactory
{
@Override
public DocTransformer create(String field, SolrParams params, SolrQueryRequest req) {
String idfield = req.getSchema().getUniqueKeyField().getName();
return new MarkTransformer(field,idfield);
}
}
class MarkTransformer extends TransformerWithContext
{
final String name;
final String idFieldName;
public MarkTransformer( String name, String idFieldName)
{
this.name = name;
this.idFieldName = idFieldName;
}
@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 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);
}
}
}

1
solr/core/src/java/org/apache/solr/response/transform/RenameFieldsTransformer.java
View File

@ -18,6 +18,7 @@ package org.apache.solr.response.transform;
import org.apache.solr.common.SolrDocument;
import org.apache.solr.common.util.NamedList;
import org.apache.solr.request.SolrQueryRequest;
/**
* Return a field with a name that is different that what is indexed

1
solr/core/src/java/org/apache/solr/response/transform/ScoreAugmenter.java
View File

@ -17,6 +17,7 @@
package org.apache.solr.response.transform;
import org.apache.solr.common.SolrDocument;
import org.apache.solr.request.SolrQueryRequest;
/**
* Simple Augmenter that adds the docId

2
solr/core/src/java/org/apache/solr/response/transform/TransformContext.java
View File

@ -17,6 +17,7 @@
package org.apache.solr.response.transform;
import org.apache.lucene.search.Query;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.search.DocIterator;
import org.apache.solr.search.SolrIndexSearcher;
@ -32,4 +33,5 @@ public class TransformContext
public boolean wantsScores = false;
public DocIterator iterator;
public SolrIndexSearcher searcher;
public SolrQueryRequest req;
}

1
solr/core/src/java/org/apache/solr/response/transform/ValueSourceAugmenter.java
View File

@ -25,6 +25,7 @@ import org.apache.lucene.queries.function.ValueSource;
import org.apache.lucene.util.ReaderUtil;
import org.apache.solr.common.SolrDocument;
import org.apache.solr.common.SolrException;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.search.QParser;
import org.apache.solr.search.SolrIndexSearcher;

39
solr/core/src/test/org/apache/solr/handler/component/QueryElevationComponentTest.java
View File

@ -117,6 +117,45 @@ public class QueryElevationComponentTest extends SolrTestCaseJ4 {
}
@Test
public void testMarker() throws Exception {
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", CommonParams.QT, "/elevate",
CommonParams.FL, "id, score, [elevated]")
,"//*[@numFound='3']"
,"//result/doc[1]/str[@name='id'][.='1']"
,"//result/doc[2]/str[@name='id'][.='4']"
,"//result/doc[3]/str[@name='id'][.='6']",
"//result/doc[1]/bool[@name='[elevated]'][.='true']",
"//result/doc[2]/bool[@name='[elevated]'][.='false']",
"//result/doc[3]/bool[@name='[elevated]'][.='false']"
);
assertQ("", req(CommonParams.Q, "AAAA", CommonParams.QT, "/elevate",
CommonParams.FL, "id, score, [elevated]")
,"//*[@numFound='1']"
,"//result/doc[1]/str[@name='id'][.='7']",
"//result/doc[1]/bool[@name='[elevated]'][.='false']"
);
assertQ("", req(CommonParams.Q, "AAAA", CommonParams.QT, "/elevate",
CommonParams.FL, "id, score, [elev]")
,"//*[@numFound='1']"
,"//result/doc[1]/str[@name='id'][.='7']",
"not(//result/doc[1]/bool[@name='[elevated]'][.='false'])",
"not(//result/doc[1]/bool[@name='[elev]'][.='false'])" // even though we asked for elev, there is no Transformer registered w/ that, so we shouldn't get a result
);
}
@Test
public void testSorting() throws IOException
{

4
solr/example/solr/conf/solrconfig.xml
View File

@ -1631,6 +1631,10 @@
<transformer name="mytrans3" class="org.apache.solr.response.transform.ValueAugmenterFactory" >
<double name="defaultValue">5</double>
</transformer>
If you are using the QueryElevationComponent, you may wish to mark documents that get boosted. The
EditorialMarkerFactory will do exactly that:
<transformer name="qecBooster" class="org.apache.solr.response.transform.EditorialMarkerFactory" />
-->

2
solr/solrj/src/java/org/apache/solr/common/SolrDocument.java
View File

@ -275,7 +275,7 @@ public class SolrDocument implements Map<String,Object>, Iterable<Map.Entry<Stri
public Set<Entry<String, Object>> entrySet() {
return _fields.entrySet();
}
//TODO: Shouldn't the input parameter here be a String? The _fields map requires a String.
public Object get(Object key) {
return _fields.get(key);
}

7
solr/solrj/src/java/org/apache/solr/common/params/QueryElevationParams.java
View File

@ -27,4 +27,11 @@ public interface QueryElevationParams {
String ENABLE = "enableElevation";
String EXCLUSIVE = "exclusive";
String FORCE_ELEVATION = "forceElevation";
/**
* The name of the field that editorial results will be written out as when using the QueryElevationComponent, which
* automatically configures the EditorialMarkerFactory. The default name is "elevated"
* <br/>
* See http://wiki.apache.org/solr/DocTransformers
*/
String EDITORIAL_MARKER_FIELD_NAME = "editorialMarkerFieldName";
}