SOLR-3076: block join parent and child queries

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1513290 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Yonik Seeley 2013-08-12 22:52:57 +00:00
parent d4c20941c9
commit f28864213f
29 changed files with 1614 additions and 73 deletions

View File

@ -25,5 +25,6 @@
<orderEntry type="module" module-name="analysis-common" />
<orderEntry type="module" module-name="lucene-core" />
<orderEntry type="module" module-name="queryparser" />
<orderEntry type="module" module-name="join" />
</component>
</module>

View File

@ -27,5 +27,6 @@
<orderEntry type="module" scope="TEST" module-name="suggest" />
<orderEntry type="module" scope="TEST" module-name="spatial" />
<orderEntry type="module" scope="TEST" module-name="misc" />
<orderEntry type="module" scope="TEST" module-name="join" />
</component>
</module>

View File

@ -93,6 +93,11 @@
<artifactId>lucene-misc</artifactId>
<version>${project.version}</version>
</dependency>
<dependency>
<groupId>org.apache.lucene</groupId>
<artifactId>lucene-join</artifactId>
<version>${project.version}</version>
</dependency>
<dependency>
<groupId>org.apache.lucene</groupId>
<artifactId>lucene-queryparser</artifactId>

View File

@ -198,6 +198,12 @@
<artifactId>jetty-util</artifactId>
<scope>test</scope>
</dependency>
<dependency>
<groupId>dom4j</groupId>
<artifactId>dom4j</artifactId>
<version>1.6.1</version>
<scope>test</scope>
</dependency>
</dependencies>
<build>
<sourceDirectory/>

View File

@ -133,6 +133,17 @@
<property name="queryparser-javadocs.uptodate" value="true"/>
</target>
<property name="join.jar" value="${common.dir}/build/join/lucene-join-${version}.jar"/>
<target name="check-join-uptodate" unless="join.uptodate">
<module-uptodate name="join" jarfile="${join.jar}" property="join.uptodate"/>
</target>
<target name="jar-join" unless="join.uptodate" depends="check-join-uptodate">
<ant dir="${common.dir}/join" target="jar-core" inheritAll="false">
<propertyset refid="uptodate.and.compiled.properties"/>
</ant>
<property name="join.uptodate" value="true"/>
</target>
<property name="analyzers-common.jar" value="${common.dir}/build/analysis/common/lucene-analyzers-common-${version}.jar"/>
<target name="check-analyzers-common-uptodate" unless="analyzers-common.uptodate">
<module-uptodate name="analysis/common" jarfile="${analyzers-common.jar}" property="analyzers-common.uptodate"/>

View File

@ -182,6 +182,7 @@ Upgrading from Solr 4.3.0
in 5.0. If you are still using these field types, you should migrate your
fields to TrieIntField.
Detailed Change List
----------------------
@ -261,6 +262,16 @@ New Features
* SOLR-4943: Add a new system wide info admin handler that exposes the system info
that could previously only be retrieved using a SolrCore. (Mark Miller)
* SOLR-3076: Block joins. Documents and their sub-documents must be indexed
as a block.
{!parent which=<allParents>}<someChildren> takes in a query that matches child
documents and results in matches on their parents.
{!child of=<allParents>}<someParents> takes in a query that matches some parent
documents and results in matches on their children.
(Mikhail Khludnev, Vadim Kirilchuk, Alan Woodward, Tom Burton-West, Mike McCandless,
hossman, yonik)
Bug Fixes
----------------------

View File

@ -82,6 +82,7 @@
<pathelement location="${grouping.jar}"/>
<pathelement location="${queries.jar}"/>
<pathelement location="${queryparser.jar}"/>
<pathelement location="${join.jar}"/>
</path>
<path id="solr.base.classpath">
@ -141,7 +142,7 @@
<target name="prep-lucene-jars"
depends="jar-lucene-core, jar-analyzers-phonetic, jar-analyzers-kuromoji, jar-codecs, jar-suggest, jar-highlighter, jar-memory,
jar-misc, jar-spatial, jar-grouping, jar-queries, jar-queryparser">
jar-misc, jar-spatial, jar-grouping, jar-queries, jar-queryparser, jar-join">
<property name="solr.deps.compiled" value="true"/>
</target>

View File

@ -41,6 +41,7 @@
<dependency org="org.restlet.jee" name="org.restlet" rev="2.1.1" conf="compile->*"/>
<dependency org="org.restlet.jee" name="org.restlet.ext.servlet" rev="2.1.1" conf="compile->*"/>
<dependency org="joda-time" name="joda-time" rev="2.2" conf="compile->*"/>
<dependency org="dom4j" name="dom4j" rev="1.6.1" transitive="false"/>
<dependency org="javax.servlet" name="javax.servlet-api" rev="3.0.1" conf="test->*"/>
<dependency org="org.easymock" name="easymock" rev="3.0" conf="test->*"/>

View File

@ -16,50 +16,51 @@ package org.apache.solr.handler.loader;
* limitations under the License.
*/
import com.google.common.collect.Lists;
import org.apache.commons.io.IOUtils;
import org.apache.solr.common.SolrException;
import org.apache.solr.common.SolrInputDocument;
import org.apache.solr.common.params.CommonParams;
import org.apache.solr.common.params.ModifiableSolrParams;
import org.apache.solr.update.processor.UpdateRequestProcessor;
import org.apache.solr.update.AddUpdateCommand;
import org.apache.solr.update.CommitUpdateCommand;
import org.apache.solr.update.RollbackUpdateCommand;
import org.apache.solr.update.DeleteUpdateCommand;
import org.apache.solr.util.xslt.TransformerProvider;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.response.SolrQueryResponse;
import org.apache.solr.common.params.SolrParams;
import org.apache.solr.common.params.UpdateParams;
import org.apache.solr.common.util.ContentStream;
import org.apache.solr.common.util.ContentStreamBase;
import org.apache.solr.common.util.StrUtils;
import org.apache.solr.common.util.XMLErrorLogger;
import org.apache.solr.common.SolrException;
import org.apache.solr.common.SolrInputDocument;
import org.apache.solr.common.params.CommonParams;
import org.apache.solr.common.params.SolrParams;
import org.apache.solr.common.params.UpdateParams;
import org.apache.solr.core.SolrConfig;
import org.apache.solr.handler.RequestHandlerUtils;
import org.apache.solr.handler.UpdateRequestHandler;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.response.SolrQueryResponse;
import org.apache.solr.update.AddUpdateCommand;
import org.apache.solr.update.CommitUpdateCommand;
import org.apache.solr.update.DeleteUpdateCommand;
import org.apache.solr.update.RollbackUpdateCommand;
import org.apache.solr.update.processor.UpdateRequestProcessor;
import org.apache.solr.util.EmptyEntityResolver;
import org.apache.commons.io.IOUtils;
import org.apache.solr.util.xslt.TransformerProvider;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.xml.sax.InputSource;
import org.xml.sax.XMLReader;
import javax.xml.stream.XMLStreamReader;
import javax.xml.stream.XMLStreamException;
import javax.xml.parsers.SAXParserFactory;
import javax.xml.stream.FactoryConfigurationError;
import javax.xml.stream.XMLStreamConstants;
import javax.xml.stream.XMLInputFactory;
import javax.xml.stream.XMLStreamConstants;
import javax.xml.stream.XMLStreamException;
import javax.xml.stream.XMLStreamReader;
import javax.xml.transform.Transformer;
import javax.xml.transform.TransformerException;
import javax.xml.transform.dom.DOMResult;
import javax.xml.transform.dom.DOMSource;
import javax.xml.transform.sax.SAXSource;
import javax.xml.parsers.SAXParserFactory;
import java.io.ByteArrayInputStream;
import java.io.InputStream;
import java.io.IOException;
import java.io.InputStream;
import java.util.ArrayList;
import java.util.Collection;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
@ -381,6 +382,7 @@ public class XMLLoader extends ContentStreamLoader {
float boost = 1.0f;
boolean isNull = false;
String update = null;
Collection<SolrInputDocument> subDocs = null;
Map<String, Map<String, Object>> updateMap = null;
boolean complete = false;
while (!complete) {
@ -395,9 +397,14 @@ public class XMLLoader extends ContentStreamLoader {
case XMLStreamConstants.END_ELEMENT:
if ("doc".equals(parser.getLocalName())) {
if (subDocs != null && !subDocs.isEmpty()) {
doc.addChildDocuments(subDocs);
subDocs = null;
}
complete = true;
break;
} else if ("field".equals(parser.getLocalName())) {
// should I warn in some text has been found too
Object v = isNull ? null : text.toString();
if (update != null) {
if (updateMap == null) updateMap = new HashMap<String, Map<String, Object>>();
@ -425,12 +432,20 @@ public class XMLLoader extends ContentStreamLoader {
}
doc.addField(name, v, boost);
boost = 1.0f;
// field is over
name = null;
}
break;
case XMLStreamConstants.START_ELEMENT:
text.setLength(0);
String localName = parser.getLocalName();
if ("doc".equals(localName)) {
if (subDocs == null)
subDocs = Lists.newArrayList();
subDocs.add(readDoc(parser));
}
else {
if (!"field".equals(localName)) {
log.warn("unexpected XML tag doc/" + localName);
throw new SolrException(SolrException.ErrorCode.BAD_REQUEST,
@ -455,6 +470,7 @@ public class XMLLoader extends ContentStreamLoader {
log.warn("Unknown attribute doc/field/@" + attrName);
}
}
}
break;
}
}

View File

@ -20,6 +20,8 @@ import org.apache.solr.common.params.SolrParams;
import org.apache.solr.common.util.NamedList;
import org.apache.solr.core.SolrInfoMBean;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.search.join.BlockJoinChildQParserPlugin;
import org.apache.solr.search.join.BlockJoinParentQParserPlugin;
import org.apache.solr.util.plugin.NamedListInitializedPlugin;
import java.net.URL;
@ -47,7 +49,9 @@ public abstract class QParserPlugin implements NamedListInitializedPlugin, SolrI
JoinQParserPlugin.NAME, JoinQParserPlugin.class,
SurroundQParserPlugin.NAME, SurroundQParserPlugin.class,
SwitchQParserPlugin.NAME, SwitchQParserPlugin.class,
MaxScoreQParserPlugin.NAME, MaxScoreQParserPlugin.class
MaxScoreQParserPlugin.NAME, MaxScoreQParserPlugin.class,
BlockJoinParentQParserPlugin.NAME, BlockJoinParentQParserPlugin.class,
BlockJoinChildQParserPlugin.NAME, BlockJoinChildQParserPlugin.class
};
/** return a {@link QParser} */

View File

@ -0,0 +1,46 @@
/*
* 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.
*/
package org.apache.solr.search.join;
import org.apache.lucene.util.OpenBitSet;
class BitSetSlice {
private final OpenBitSet obs;
private final int off;
private final int len;
BitSetSlice(OpenBitSet obs, int off, int len) {
this.obs = obs;
this.off = off;
this.len = len;
}
public boolean get(int pos) {
return obs.get(pos + off);
}
public int prevSetBit(int pos) {
int result = obs.prevSetBit(pos + off) - off;
return (result < 0) ? -1 : result;
}
public int nextSetBit(int pos) {
int result = obs.nextSetBit(pos + off) - off;
return (result < 0 || result >= len) ? -1 : result;
}
}

View File

@ -0,0 +1,41 @@
/*
* 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.
*/
package org.apache.solr.search.join;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.join.ToChildBlockJoinQuery;
import org.apache.solr.common.params.SolrParams;
import org.apache.solr.request.SolrQueryRequest;
public class BlockJoinChildQParser extends BlockJoinParentQParser {
public BlockJoinChildQParser(String qstr, SolrParams localParams, SolrParams params, SolrQueryRequest req) {
super(qstr, localParams, params, req);
}
protected Query createQuery(Query parentListQuery, Query query) {
return new ToChildBlockJoinQuery(query, getFilter(parentListQuery), false);
}
@Override
protected String getParentFilterLocalParamName() {
return "of";
}
}

View File

@ -0,0 +1,36 @@
/*
* 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.
*/
package org.apache.solr.search.join;
import org.apache.solr.common.params.SolrParams;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.search.QParser;
/**
* Usage: {!child of="PARENT:true"}PARENT_PRICE:10
*
**/
public class BlockJoinChildQParserPlugin extends BlockJoinParentQParserPlugin {
public static String NAME = "child";
@Override
protected QParser createBJQParser(String qstr, SolrParams localParams, SolrParams params, SolrQueryRequest req) {
return new BlockJoinChildQParser(qstr, localParams, params, req);
}
}

View File

@ -0,0 +1,98 @@
/*
* 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.
*/
package org.apache.solr.search.join;
import org.apache.lucene.search.CachingWrapperFilter;
import org.apache.lucene.search.ConstantScoreQuery;
import org.apache.lucene.search.Filter;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.QueryWrapperFilter;
import org.apache.lucene.search.join.ScoreMode;
import org.apache.lucene.search.join.ToParentBlockJoinQuery;
import org.apache.solr.common.params.SolrParams;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.search.QParser;
import org.apache.solr.search.QueryParsing;
import org.apache.solr.search.SolrCache;
import org.apache.solr.search.SolrConstantScoreQuery;
import org.apache.solr.search.SyntaxError;
class BlockJoinParentQParser extends QParser {
/** implementation detail subject to change */
public String CACHE_NAME="perSegFilter";
protected String getParentFilterLocalParamName() {
return "which";
}
BlockJoinParentQParser(String qstr, SolrParams localParams, SolrParams params, SolrQueryRequest req) {
super(qstr, localParams, params, req);
}
@Override
public Query parse() throws SyntaxError {
String filter = localParams.get(getParentFilterLocalParamName());
QParser parentParser = subQuery(filter, null);
Query parentQ = parentParser.getQuery();
String queryText = localParams.get(QueryParsing.V);
// there is no child query, return parent filter from cache
if (queryText == null || queryText.length()==0) {
SolrConstantScoreQuery wrapped = new SolrConstantScoreQuery(getFilter(parentQ));
wrapped.setCache(false);
return wrapped;
}
QParser childrenParser = subQuery(queryText, null);
Query childrenQuery = childrenParser.getQuery();
return createQuery(parentQ, childrenQuery);
}
protected Query createQuery(Query parentList, Query query) {
return new ToParentBlockJoinQuery(query, getFilter(parentList), ScoreMode.None);
}
protected Filter getFilter(Query parentList) {
SolrCache parentCache = req.getSearcher().getCache(CACHE_NAME);
// lazily retrieve from solr cache
Filter filter = null;
if (parentCache != null) {
filter = (Filter) parentCache.get(parentList);
}
Filter result;
if (filter == null) {
result = createParentFilter(parentList);
if (parentCache != null) {
parentCache.put(parentList, result);
}
} else {
result = filter;
}
return result;
}
protected Filter createParentFilter(Query parentQ) {
return new CachingWrapperFilter(new QueryWrapperFilter(parentQ)) {
};
}
}

View File

@ -0,0 +1,47 @@
/*
* 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.
*/
package org.apache.solr.search.join;
import org.apache.solr.common.params.SolrParams;
import org.apache.solr.common.util.NamedList;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.search.QParser;
import org.apache.solr.search.QParserPlugin;
/**
* Usage: {!parent which="PARENT:true"}CHILD_PRICE:10
*
**/
public class BlockJoinParentQParserPlugin extends QParserPlugin {
public static String NAME = "parent";
@Override
public QParser createParser(String qstr, SolrParams localParams, SolrParams params, SolrQueryRequest req) {
QParser parser = createBJQParser(qstr, localParams, params, req);
return parser;
}
protected QParser createBJQParser(String qstr, SolrParams localParams, SolrParams params, SolrQueryRequest req) {
return new BlockJoinParentQParser(qstr, localParams, params, req);
}
@Override
public void init(NamedList args) {
}
}

View File

@ -0,0 +1,122 @@
/*
* 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.
*/
package org.apache.solr.search.join;
import org.apache.lucene.index.AtomicReaderContext;
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.index.Term;
import org.apache.lucene.search.Explanation;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.Scorer;
import org.apache.lucene.search.Weight;
import org.apache.lucene.util.Bits;
import java.io.IOException;
import java.util.Set;
public class IgnoreAcceptDocsQuery extends Query {
private final Query q;
public IgnoreAcceptDocsQuery(Query q) {
this.q = q;
}
@Override
public void setBoost(float b) {
q.setBoost(b);
}
@Override
public float getBoost() {
return q.getBoost();
}
@Override
public String toString() {
return q.toString();
}
@Override
public Weight createWeight(IndexSearcher searcher) throws IOException {
Weight inner = q.createWeight(searcher);
return new IADWeight(inner);
}
private class IADWeight extends Weight {
Weight w;
IADWeight(Weight delegate) {
this.w = delegate;
}
@Override
public Explanation explain(AtomicReaderContext context, int doc) throws IOException {
return w.explain(context, doc);
}
@Override
public Query getQuery() {
return q;
}
@Override
public float getValueForNormalization() throws IOException {
return w.getValueForNormalization();
}
@Override
public void normalize(float norm, float topLevelBoost) {
w.normalize(norm, topLevelBoost);
}
@Override
public Scorer scorer(AtomicReaderContext context, boolean scoreDocsInOrder, boolean topScorer, Bits acceptDocs) throws IOException {
return w.scorer(context, scoreDocsInOrder, topScorer, null);
}
}
@Override
public Query rewrite(IndexReader reader) throws IOException {
Query n = q.rewrite(reader);
if (q == n) return this;
return new IgnoreAcceptDocsQuery(n);
}
@Override
public void extractTerms(Set<Term> terms) {
q.extractTerms(terms);
}
@Override
public int hashCode() {
return q.hashCode()*31;
}
@Override
public boolean equals(Object o) {
if (!(o instanceof IgnoreAcceptDocsQuery)) return false;
IgnoreAcceptDocsQuery other = (IgnoreAcceptDocsQuery)o;
return q.equals(other.q);
}
@Override
public String toString(String field) {
return "IgnoreAcceptDocs(" + q + ")";
}
}

View File

@ -18,6 +18,7 @@
package org.apache.solr.update;
import org.apache.lucene.document.Document;
import org.apache.lucene.index.IndexDocument;
import org.apache.lucene.index.Term;
import org.apache.lucene.util.BytesRef;
import org.apache.solr.common.SolrException;
@ -27,10 +28,15 @@ import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.schema.IndexSchema;
import org.apache.solr.schema.SchemaField;
import java.util.ArrayList;
import java.util.Collections;
import java.util.Iterator;
import java.util.List;
/**
*
*/
public class AddUpdateCommand extends UpdateCommand {
public class AddUpdateCommand extends UpdateCommand implements Iterable<IndexDocument> {
// optional id in "internal" indexed form... if it is needed and not supplied,
// it will be obtained from the doc.
private BytesRef indexedId;
@ -144,6 +150,62 @@ public class AddUpdateCommand extends UpdateCommand {
return id;
}
public boolean isBlock() {
return solrDoc.hasChildDocuments();
}
@Override
public Iterator<IndexDocument> iterator() {
return new Iterator<IndexDocument>() {
Iterator<SolrInputDocument> iter;
{
List<SolrInputDocument> all = flatten(solrDoc);
SchemaField uniq = req.getSchema().getUniqueKeyField();
String idField = getHashableId();
for (SolrInputDocument sdoc : all) {
sdoc.setField("_root_", idField); // should this be a string or the same type as the ID?
// TODO: if possible concurrent modification exception (if SolrInputDocument not cloned and is being forwarded to replicas)
// then we could add this field to the generated lucene document instead.
}
iter = all.iterator();
}
@Override
public boolean hasNext() {
return iter.hasNext();
}
@Override
public IndexDocument next() {
return DocumentBuilder.toDocument(iter.next(), req.getSchema());
}
@Override
public void remove() {
throw new UnsupportedOperationException();
}
};
}
private List<SolrInputDocument> flatten(SolrInputDocument root) {
List<SolrInputDocument> unwrappedDocs = new ArrayList<SolrInputDocument>();
recUnwrapp(unwrappedDocs, root);
Collections.reverse(unwrappedDocs);
return unwrappedDocs;
}
private void recUnwrapp(List<SolrInputDocument> unwrappedDocs, SolrInputDocument currentDoc) {
unwrappedDocs.add(currentDoc);
for (SolrInputDocument child : currentDoc.getChildDocuments()) {
recUnwrapp(unwrappedDocs, child);
}
}
@Override
public String toString() {
StringBuilder sb = new StringBuilder(super.toString());
@ -153,4 +215,6 @@ public class AddUpdateCommand extends UpdateCommand {
sb.append('}');
return sb.toString();
}
}
}

View File

@ -20,16 +20,6 @@
package org.apache.solr.update;
import java.io.IOException;
import java.net.URL;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.Future;
import java.util.concurrent.atomic.AtomicLong;
import org.apache.lucene.document.Document;
import org.apache.lucene.index.DirectoryReader;
import org.apache.lucene.index.IndexReader;
@ -56,12 +46,22 @@ import org.apache.solr.schema.IndexSchema;
import org.apache.solr.schema.SchemaField;
import org.apache.solr.search.FunctionRangeQuery;
import org.apache.solr.search.QParser;
import org.apache.solr.search.SyntaxError;
import org.apache.solr.search.QueryUtils;
import org.apache.solr.search.SolrIndexSearcher;
import org.apache.solr.search.SyntaxError;
import org.apache.solr.search.function.ValueSourceRangeFilter;
import org.apache.solr.util.RefCounted;
import java.io.IOException;
import java.net.URL;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.Future;
import java.util.concurrent.atomic.AtomicLong;
/**
* <code>DirectUpdateHandler2</code> implements an UpdateHandler where documents are added
* directly to the main Lucene index as opposed to adding to a separate smaller index.
@ -199,19 +199,23 @@ public class DirectUpdateHandler2 extends UpdateHandler implements SolrCoreState
// normal update
Term updateTerm;
Term idTerm = new Term(idField.getName(), cmd.getIndexedId());
Term idTerm = new Term(cmd.isBlock() ? "_root_" : idField.getName(), cmd.getIndexedId());
boolean del = false;
if (cmd.updateTerm == null) {
updateTerm = idTerm;
} else {
// this is only used by the dedup update processor
del = true;
updateTerm = cmd.updateTerm;
}
if (cmd.isBlock()) {
writer.updateDocuments(updateTerm, cmd, schema.getAnalyzer());
} else {
Document luceneDocument = cmd.getLuceneDocument();
// SolrCore.verbose("updateDocument",updateTerm,luceneDocument,writer);
writer.updateDocument(updateTerm, luceneDocument,
schema.getAnalyzer());
writer.updateDocument(updateTerm, luceneDocument, schema.getAnalyzer());
}
// SolrCore.verbose("updateDocument",updateTerm,"DONE");
if (del) { // ensure id remains unique
@ -234,7 +238,12 @@ public class DirectUpdateHandler2 extends UpdateHandler implements SolrCoreState
} else {
// allow duplicates
if (cmd.isBlock()) {
writer.addDocuments(cmd, schema.getAnalyzer());
} else {
writer.addDocument(cmd.getLuceneDocument(), schema.getAnalyzer());
}
if (ulog != null) ulog.add(cmd);
}

View File

@ -441,6 +441,8 @@
<fields>
<field name="id" type="int" indexed="true" stored="true" multiValued="false" required="false"/>
<field name="_root_" type="int" indexed="true" stored="true" multiValued="false" required="false"/>
<field name="signatureField" type="string" indexed="true" stored="false"/>
<field name="uuid" type="uuid" stored="true" />
<field name="name" type="nametext" indexed="true" stored="true"/>

View File

@ -523,12 +523,14 @@
<field name="uniq3" type="string" indexed="true" stored="true"/>
<field name="nouniq" type="string" indexed="true" stored="true" multiValued="true"/>
<!-- for versioning -->
<field name="_version_" type="long" indexed="true" stored="true"/>
<field name="copyfield_source" type="string" indexed="true" stored="true" multiValued="true"/>
<!-- for versioning -->
<field name="_version_" type="long" indexed="true" stored="true"/>
<!-- points to the root document of a block of nested documents -->
<field name="_root_" type="string" indexed="true" stored="true"/>
<dynamicField name="*_coordinate" type="tdouble" indexed="true" stored="false"/>
@ -560,6 +562,7 @@
<dynamicField name="*_l" type="long" indexed="true" stored="true"/>
<dynamicField name="*_t" type="text" indexed="true" stored="true"/>
<dynamicField name="*_tt" type="text" indexed="true" stored="true"/>
<dynamicField name="*_ws" type="nametext" indexed="true" stored="true"/>
<dynamicField name="*_b" type="boolean" indexed="true" stored="true"/>
<dynamicField name="*_f" type="float" indexed="true" stored="true"/>
<dynamicField name="*_d" type="double" indexed="true" stored="true"/>

View File

@ -120,6 +120,12 @@
initialSize="512"
autowarmCount="0"/>
<cache name="perSegFilter"
class="solr.search.LRUCache"
size="10"
initialSize="0"
autowarmCount="10" />
<!-- If true, stored fields that are not requested will be loaded lazily.
-->
<enableLazyFieldLoading>true</enableLazyFieldLoading>
@ -553,3 +559,4 @@
</updateRequestProcessorChain>
</config>

View File

@ -121,6 +121,8 @@ public class FullSolrCloudDistribCmdsTest extends AbstractFullDistribZkTestBase
results = query(cloudClient);
assertEquals(2, results.getResults().getNumFound());
docId = testIndexQueryDeleteHierarchical(docId);
testIndexingWithSuss();
// TODO: testOptimisticUpdate(results);
@ -235,6 +237,75 @@ public class FullSolrCloudDistribCmdsTest extends AbstractFullDistribZkTestBase
assertEquals(0, query(cloudClient).getResults().getNumFound());
}
private long testIndexQueryDeleteHierarchical(long docId) throws Exception {
//index
int topDocsNum = atLeast(10);
int childsNum = atLeast(10);
for (int i = 0; i < topDocsNum; ++i) {
UpdateRequest uReq = new UpdateRequest();
SolrInputDocument topDocument = new SolrInputDocument();
topDocument.addField("id", docId++);
topDocument.addField("type_s", "parent");
topDocument.addField(i + "parent_f1_s", "v1");
topDocument.addField(i + "parent_f2_s", "v2");
for (int index = 0; index < childsNum; ++index) {
docId = addChildren("child", topDocument, index, false, docId);
}
uReq.add(topDocument);
uReq.process(cloudClient);
uReq.process(controlClient);
}
commit();
checkShardConsistency();
assertDocCounts(VERBOSE);
//query
// parents
SolrQuery query = new SolrQuery("type_s:parent");
QueryResponse results = cloudClient.query(query);
assertEquals(topDocsNum, results.getResults().getNumFound());
//childs
query = new SolrQuery("type_s:child");
results = cloudClient.query(query);
assertEquals(topDocsNum * childsNum, results.getResults().getNumFound());
//grandchilds
query = new SolrQuery("type_s:grand");
results = cloudClient.query(query);
//each topDoc has t childs where each child has x = 0 + 2 + 4 + ..(t-1)*2 grands
//x = 2 * (1 + 2 + 3 +.. (t-1)) => arithmetic summ of t-1
//x = 2 * ((t-1) * t / 2) = t * (t - 1)
assertEquals(topDocsNum * childsNum * (childsNum - 1), results.getResults().getNumFound());
//delete
del("*:*");
commit();
return docId;
}
private long addChildren(String prefix, SolrInputDocument topDocument, int childIndex, boolean lastLevel, long docId) {
SolrInputDocument childDocument = new SolrInputDocument();
childDocument.addField("id", docId++);
childDocument.addField("type_s", prefix);
for (int index = 0; index < childIndex; ++index) {
childDocument.addField(childIndex + prefix + index + "_s", childIndex + "value"+ index);
}
if (!lastLevel) {
for (int i = 0; i < childIndex * 2; ++i) {
docId = addChildren("grand", childDocument, i, true, docId);
}
}
topDocument.addChildDocument(childDocument);
return docId;
}
private void testIndexingWithSuss() throws Exception {
ConcurrentUpdateSolrServer suss = new ConcurrentUpdateSolrServer(
((HttpSolrServer) clients.get(0)).getBaseURL(), 3, 1);

View File

@ -298,6 +298,13 @@ public class QueryEqualityTest extends SolrTestCaseJ4 {
}
}
public void testBlockJoin() throws Exception {
assertQueryEquals("parent", "{!parent which=foo_s:parent}dude",
"{!parent which=foo_s:parent}dude");
assertQueryEquals("child", "{!child of=foo_s:parent}dude",
"{!child of=foo_s:parent}dude");
}
public void testQuerySurround() throws Exception {
assertQueryEquals("surround", "{!surround}and(apache,solr)",
"and(apache,solr)", "apache AND solr");

View File

@ -0,0 +1,274 @@
/*
* 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.
*/
package org.apache.solr.search.join;
import org.apache.solr.SolrTestCaseJ4;
import org.apache.solr.common.util.NamedList;
import org.apache.solr.request.SolrQueryRequest;
import org.apache.solr.search.QParser;
import org.apache.solr.search.SolrCache;
import org.apache.solr.search.SyntaxError;
import org.junit.BeforeClass;
import org.junit.Test;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
import java.util.List;
import java.util.ListIterator;
public class BJQParserTest extends SolrTestCaseJ4 {
private static final String[] klm = new String[] {"k", "l", "m"};
private static final List<String> xyz = Arrays.asList("x", "y", "z");
private static final String[] abcdef = new String[] {"a", "b", "c", "d", "e", "f"};
@BeforeClass
public static void beforeClass() throws Exception {
initCore("solrconfig.xml", "schema15.xml");
createIndex();
}
public static void createIndex() throws IOException, Exception {
int i = 0;
List<List<String[]>> blocks = createBlocks();
for (List<String[]> block : blocks) {
for (String[] doc : block) {
String[] idDoc = Arrays.copyOf(doc,doc.length+2);
idDoc[doc.length]="id";
idDoc[doc.length+1]=Integer.toString(i);
assertU(add(doc(idDoc)));
i++;
}
if (random().nextBoolean()) {
assertU(commit());
// force empty segment (actually, this will no longer create an empty segment, only a new segments_n)
if (random().nextBoolean()) {
assertU(commit());
}
}
}
assertU(commit());
assertQ(req("q", "*:*"), "//*[@numFound='" + i + "']");
/*
* dump docs well System.out.println(h.query(req("q","*:*",
* "sort","_docid_ asc", "fl",
* "parent_s,child_s,parentchild_s,grand_s,grand_child_s,grand_parentchild_s"
* , "wt","csv", "rows","1000"))); /
*/
}
private static int id=0;
private static List<List<String[]>> createBlocks() {
List<List<String[]>> blocks = new ArrayList<List<String[]>>();
for (String parent : abcdef) {
List<String[]> block = createChildrenBlock(parent);
block.add(new String[] {"parent_s", parent});
blocks.add(block);
}
Collections.shuffle(blocks, random());
return blocks;
}
private static List<String[]> createChildrenBlock(String parent) {
List<String[]> block = new ArrayList<String[]>();
for (String child : klm) {
block
.add(new String[] {"child_s", child, "parentchild_s", parent + child});
}
Collections.shuffle(block, random());
addGrandChildren(block);
return block;
}
private static void addGrandChildren(List<String[]> block) {
List<String> grandChildren = new ArrayList<String>(xyz);
// add grandchildren after children
for (ListIterator<String[]> iter = block.listIterator(); iter.hasNext();) {
String[] child = iter.next();
String child_s = child[1];
String parentchild_s = child[3];
int grandChildPos = 0;
boolean lastLoopButStillHasGrCh = !iter.hasNext()
&& !grandChildren.isEmpty();
while (!grandChildren.isEmpty()
&& ((grandChildPos = random().nextInt(grandChildren.size() * 2)) < grandChildren
.size() || lastLoopButStillHasGrCh)) {
grandChildPos = grandChildPos >= grandChildren.size() ? 0
: grandChildPos;
iter.add(new String[] {"grand_s", grandChildren.remove(grandChildPos),
"grand_child_s", child_s, "grand_parentchild_s", parentchild_s});
}
}
// and reverse after that
Collections.reverse(block);
}
@Test
public void testFull() throws IOException, Exception {
String childb = "{!parent which=\"parent_s:[* TO *]\"}child_s:l";
assertQ(req("q", childb), sixParents);
}
private static final String sixParents[] = new String[] {
"//*[@numFound='6']", "//doc/arr[@name=\"parent_s\"]/str='a'",
"//doc/arr[@name=\"parent_s\"]/str='b'",
"//doc/arr[@name=\"parent_s\"]/str='c'",
"//doc/arr[@name=\"parent_s\"]/str='d'",
"//doc/arr[@name=\"parent_s\"]/str='e'",
"//doc/arr[@name=\"parent_s\"]/str='f'"};
@Test
public void testJustParentsFilter() throws IOException {
assertQ(req("q", "{!parent which=\"parent_s:[* TO *]\"}"), sixParents);
}
private final static String beParents[] = new String[] {"//*[@numFound='2']",
"//doc/arr[@name=\"parent_s\"]/str='b'",
"//doc/arr[@name=\"parent_s\"]/str='e'"};
@Test
public void testIntersectBqBjq() {
assertQ(
req("q", "+parent_s:(e b) +_query_:\"{!parent which=$pq v=$chq}\"",
"chq", "child_s:l", "pq", "parent_s:[* TO *]"), beParents);
assertQ(
req("fq", "{!parent which=$pq v=$chq}\"", "q", "parent_s:(e b)", "chq",
"child_s:l", "pq", "parent_s:[* TO *]"), beParents);
assertQ(
req("q", "*:*", "fq", "{!parent which=$pq v=$chq}\"", "fq",
"parent_s:(e b)", "chq", "child_s:l", "pq", "parent_s:[* TO *]"),
beParents);
}
@Test
public void testFq() {
assertQ(
req("q", "{!parent which=$pq v=$chq}", "fq", "parent_s:(e b)", "chq",
"child_s:l", "pq", "parent_s:[* TO *]"// ,"debugQuery","on"
), beParents);
boolean qfq = random().nextBoolean();
assertQ(
req(qfq ? "q" : "fq", "parent_s:(a e b)", (!qfq) ? "q" : "fq",
"{!parent which=$pq v=$chq}", "chq", "parentchild_s:(bm ek cl)",
"pq", "parent_s:[* TO *]"), beParents);
}
@Test
public void testIntersectParentBqChildBq() throws IOException {
assertQ(
req("q", "+parent_s:(a e b) +_query_:\"{!parent which=$pq v=$chq}\"",
"chq", "parentchild_s:(bm ek cl)", "pq", "parent_s:[* TO *]"),
beParents);
}
@Test
public void testGrandChildren() throws IOException {
assertQ(
req("q", "{!parent which=$parentfilter v=$children}", "children",
"{!parent which=$childrenfilter v=$grandchildren}",
"grandchildren", "grand_s:" + "x", "parentfilter",
"parent_s:[* TO *]", "childrenfilter", "child_s:[* TO *]"),
sixParents);
// int loops = atLeast(1);
String grandChildren = xyz.get(random().nextInt(xyz.size()));
assertQ(
req("q", "+parent_s:(a e b) +_query_:\"{!parent which=$pq v=$chq}\"",
"chq", "{!parent which=$childfilter v=$grandchq}", "grandchq",
"+grand_s:" + grandChildren + " +grand_parentchild_s:(b* e* c*)",
"pq", "parent_s:[* TO *]", "childfilter", "child_s:[* TO *]"),
beParents);
}
@Test
public void testChildrenParser() {
assertQ(
req("q", "{!child of=\"parent_s:[* TO *]\"}parent_s:a", "fq",
"NOT grand_s:[* TO *]"), "//*[@numFound='3']",
"//doc/arr[@name=\"child_s\"]/str='k'",
"//doc/arr[@name=\"child_s\"]/str='l'",
"//doc/arr[@name=\"child_s\"]/str='m'");
assertQ(
req("q", "{!child of=\"parent_s:[* TO *]\"}parent_s:b", "fq",
"-parentchild_s:bm", "fq", "-grand_s:*"), "//*[@numFound='2']",
"//doc/arr[@name=\"child_s\"]/str='k'",
"//doc/arr[@name=\"child_s\"]/str='l'");
}
@Test
public void testCacheHit() throws IOException {
SolrCache parentFilterCache = (SolrCache) h.getCore().getInfoRegistry()
.get("perSegFilter");
SolrCache filterCache = (SolrCache) h.getCore().getInfoRegistry()
.get("filterCache");
NamedList parentsBefore = parentFilterCache.getStatistics();
NamedList filtersBefore = filterCache.getStatistics();
// it should be weird enough to be uniq
String parentFilter = "parent_s:([a TO c] [d TO f])";
assertQ("search by parent filter",
req("q", "{!parent which=\"" + parentFilter + "\"}"),
"//*[@numFound='6']");
assertQ("filter by parent filter",
req("q", "*:*", "fq", "{!parent which=\"" + parentFilter + "\"}"),
"//*[@numFound='6']");
assertEquals("didn't hit fqCache yet ", 0L,
delta("hits", filterCache.getStatistics(), filtersBefore));
assertQ(
"filter by join",
req("q", "*:*", "fq", "{!parent which=\"" + parentFilter
+ "\"}child_s:l"), "//*[@numFound='6']");
assertEquals("in cache mode every request lookups", 3,
delta("lookups", parentFilterCache.getStatistics(), parentsBefore));
assertEquals("last two lookups causes hits", 2,
delta("hits", parentFilterCache.getStatistics(), parentsBefore));
assertEquals("the first lookup gets insert", 1,
delta("inserts", parentFilterCache.getStatistics(), parentsBefore));
assertEquals("true join query is cached in fqCache", 1L,
delta("lookups", filterCache.getStatistics(), filtersBefore));
}
private long delta(String key, NamedList a, NamedList b) {
return (Long) a.get(key) - (Long) b.get(key);
}
@Test
public void nullInit() {
new BlockJoinParentQParserPlugin().init(null);
}
}

View File

@ -0,0 +1,601 @@
package org.apache.solr.update;
import org.apache.commons.io.output.ByteArrayOutputStream;
import org.apache.lucene.index.Term;
import org.apache.lucene.search.TermQuery;
import org.apache.lucene.search.TermRangeFilter;
import org.apache.lucene.search.TopDocs;
import org.apache.lucene.search.join.ScoreMode;
import org.apache.lucene.search.join.ToParentBlockJoinQuery;
import org.apache.solr.SolrTestCaseJ4;
import org.apache.solr.client.solrj.request.RequestWriter;
import org.apache.solr.client.solrj.request.UpdateRequest;
import org.apache.solr.common.SolrInputDocument;
import org.apache.solr.common.util.JavaBinCodec;
import org.apache.solr.handler.loader.XMLLoader;
import org.apache.solr.search.SolrIndexSearcher;
import org.apache.solr.util.DefaultSolrThreadFactory;
import org.apache.solr.util.RefCounted;
import org.dom4j.Document;
import org.dom4j.DocumentHelper;
import org.dom4j.Element;
import org.junit.After;
import org.junit.AfterClass;
import org.junit.Before;
import org.junit.BeforeClass;
import org.junit.Test;
import org.xml.sax.SAXException;
import javax.xml.stream.XMLInputFactory;
import javax.xml.stream.XMLStreamException;
import javax.xml.stream.XMLStreamReader;
import java.io.ByteArrayInputStream;
import java.io.IOException;
import java.io.InputStream;
import java.io.OutputStream;
import java.io.StringReader;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collection;
import java.util.Collections;
import java.util.List;
import java.util.concurrent.Callable;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;
import java.util.concurrent.atomic.AtomicInteger;
import javax.xml.stream.XMLStreamException;
import javax.xml.stream.XMLStreamReader;
/**
* 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.
*/
public class AddBlockUpdateTest extends SolrTestCaseJ4 {
private static final String child = "child_s";
private static final String parent = "parent_s";
private static final String type = "type_s";
private static ExecutorService exe;
private static AtomicInteger counter = new AtomicInteger();
private static boolean cachedMode;
private static XMLInputFactory inputFactory;
private RefCounted<SolrIndexSearcher> searcherRef;
private SolrIndexSearcher _searcher;
@BeforeClass
public static void beforeClass() throws Exception {
String oldCacheNamePropValue = System
.getProperty("blockJoinParentFilterCache");
System.setProperty("blockJoinParentFilterCache", (cachedMode = random()
.nextBoolean()) ? "blockJoinParentFilterCache" : "don't cache");
if (oldCacheNamePropValue != null) {
System.setProperty("blockJoinParentFilterCache", oldCacheNamePropValue);
}
inputFactory = XMLInputFactory.newInstance();
exe = // Executors.newSingleThreadExecutor();
rarely() ? Executors.newFixedThreadPool(atLeast(2), new DefaultSolrThreadFactory("AddBlockUpdateTest")) : Executors
.newCachedThreadPool(new DefaultSolrThreadFactory("AddBlockUpdateTest"));
initCore("solrconfig.xml", "schema15.xml");
}
@Before
public void prepare() {
// assertU("<rollback/>");
assertU(delQ("*:*"));
assertU(commit("expungeDeletes", "true"));
}
private SolrIndexSearcher getSearcher() {
if (_searcher == null) {
searcherRef = h.getCore().getSearcher();
_searcher = searcherRef.get();
}
return _searcher;
}
@After
public void cleanup() {
if (searcherRef != null || _searcher != null) {
searcherRef.decref();
searcherRef = null;
_searcher = null;
}
}
@AfterClass
public static void afterClass() throws Exception {
inputFactory = null;
exe.shutdownNow();
}
@Test
public void testBasics() throws Exception {
List<Document> blocks = new ArrayList<Document>(Arrays.asList(
block("abcD"),
block("efgH"),
merge(block("ijkL"), block("mnoP")),
merge(block("qrsT"), block("uvwX")),
block("Y"),
block("Z")));
Collections.shuffle(blocks);
log.trace("{}", blocks);
for (Future<Void> f : exe.invokeAll(callables(blocks))) {
f.get(); // exceptions?
}
assertU(commit());
final SolrIndexSearcher searcher = getSearcher();
// final String resp = h.query(req("q","*:*", "sort","_docid_ asc", "rows",
// "10000"));
// log.trace(resp);
int parentsNum = "DHLPTXYZ".length();
assertQ(req(parent + ":[* TO *]"), "//*[@numFound='" + parentsNum + "']");
assertQ(req(child + ":[* TO *]"), "//*[@numFound='"
+ (('z' - 'a' + 1) - parentsNum) + "']");
assertQ(req("*:*"), "//*[@numFound='" + ('z' - 'a' + 1) + "']");
assertSingleParentOf(searcher, one("abc"), "D");
assertSingleParentOf(searcher, one("efg"), "H");
assertSingleParentOf(searcher, one("ijk"), "L");
assertSingleParentOf(searcher, one("mno"), "P");
assertSingleParentOf(searcher, one("qrs"), "T");
assertSingleParentOf(searcher, one("uvw"), "X");
}
/***
@Test
public void testSmallBlockDirect() throws Exception {
final AddBlockUpdateCommand cmd = new AddBlockUpdateCommand(req("*:*"));
final List<SolrInputDocument> docs = Arrays.asList(new SolrInputDocument() {
{
addField("id", id());
addField(child, "a");
}
}, new SolrInputDocument() {
{
addField("id", id());
addField(parent, "B");
}
});
cmd.setDocs(docs);
assertEquals(2, h.getCore().getUpdateHandler().addBlock(cmd));
assertU(commit());
final SolrIndexSearcher searcher = getSearcher();
assertQ(req("*:*"), "//*[@numFound='2']");
assertSingleParentOf(searcher, one("a"), "B");
}
@Test
public void testEmptyDirect() throws Exception {
final AddBlockUpdateCommand cmd = new AddBlockUpdateCommand(req("*:*"));
// let's add empty one
cmd.setDocs(Collections.<SolrInputDocument> emptyList());
assertEquals(0,
((DirectUpdateHandler2) h.getCore().getUpdateHandler()).addBlock(cmd));
assertU(commit());
assertQ(req("*:*"), "//*[@numFound='0']");
}
***/
@Test
public void testExceptionThrown() throws Exception {
final String abcD = block("abcD").asXML();
log.info(abcD);
assertBlockU(abcD);
Document docToFail = DocumentHelper.createDocument();
Element root = docToFail.addElement("add");
Element doc1 = root.addElement("doc");
attachField(doc1, "id", id());
attachField(doc1, parent, "Y");
attachField(doc1, "sample_i", "notanumber");
Element subDoc1 = doc1.addElement("doc");
attachField(subDoc1, "id", id());
attachField(subDoc1, child, "x");
Element doc2 = root.addElement("doc");
attachField(doc2, "id", id());
attachField(doc2, parent, "W");
assertFailedBlockU(docToFail.asXML());
assertBlockU(block("efgH").asXML());
assertBlockU(commit());
final SolrIndexSearcher searcher = getSearcher();
assertQ(req("q","*:*","indent","true", "fl","id,parent_s,child_s"), "//*[@numFound='" + "abcDefgH".length() + "']");
assertSingleParentOf(searcher, one("abc"), "D");
assertSingleParentOf(searcher, one("efg"), "H");
assertQ(req(child + ":x"), "//*[@numFound='0']");
assertQ(req(parent + ":Y"), "//*[@numFound='0']");
assertQ(req(parent + ":W"), "//*[@numFound='0']");
}
@SuppressWarnings("serial")
@Test
public void testSolrJXML() throws IOException {
UpdateRequest req = new UpdateRequest();
List<SolrInputDocument> docs = new ArrayList<SolrInputDocument>();
SolrInputDocument document1 = new SolrInputDocument() {
{
final String id = id();
addField("id", id);
addField("parent_s", "X");
ArrayList<SolrInputDocument> ch1 = new ArrayList<SolrInputDocument>(
Arrays.asList(new SolrInputDocument() {
{
addField("id", id());
addField("child_s", "y");
}
}, new SolrInputDocument() {
{
addField("id", id());
addField("child_s", "z");
}
}));
Collections.shuffle(ch1, random());
addChildDocuments(ch1);
}
};
SolrInputDocument document2 = new SolrInputDocument() {
{
final String id = id();
addField("id", id);
addField("parent_s", "A");
addChildDocument(new SolrInputDocument() {
{
addField("id", id());
addField("child_s", "b");
}
});
addChildDocument(new SolrInputDocument() {
{
addField("id", id());
addField("child_s", "c");
}
});
}
};
docs.add(document1);
docs.add(document2);
Collections.shuffle(docs, random());
req.add(docs);
RequestWriter requestWriter = new RequestWriter();
OutputStream os = new ByteArrayOutputStream();
requestWriter.write(req, os);
assertBlockU(os.toString());
assertU(commit());
final SolrIndexSearcher searcher = getSearcher();
assertSingleParentOf(searcher, one("yz"), "X");
assertSingleParentOf(searcher, one("bc"), "A");
}
//This is the same as testSolrJXML above but uses the XMLLoader
// to illustrate the structure of the XML documents
@Test
public void testXML() throws IOException, XMLStreamException {
UpdateRequest req = new UpdateRequest();
List<SolrInputDocument> docs = new ArrayList<SolrInputDocument>();
String xml_doc1 =
"<doc >" +
" <field name=\"id\">1</field>" +
" <field name=\"parent_s\">X</field>" +
"<doc> " +
" <field name=\"id\" >2</field>" +
" <field name=\"child_s\">y</field>" +
"</doc>"+
"<doc> " +
" <field name=\"id\" >3</field>" +
" <field name=\"child_s\">z</field>" +
"</doc>"+
"</doc>";
String xml_doc2 =
"<doc >" +
" <field name=\"id\">4</field>" +
" <field name=\"parent_s\">A</field>" +
"<doc> " +
" <field name=\"id\" >5</field>" +
" <field name=\"child_s\">b</field>" +
"</doc>"+
"<doc> " +
" <field name=\"id\" >6</field>" +
" <field name=\"child_s\">c</field>" +
"</doc>"+
"</doc>";
XMLStreamReader parser =
inputFactory.createXMLStreamReader( new StringReader( xml_doc1 ) );
parser.next(); // read the START document...
//null for the processor is all right here
XMLLoader loader = new XMLLoader();
SolrInputDocument document1 = loader.readDoc( parser );
XMLStreamReader parser2 =
inputFactory.createXMLStreamReader( new StringReader( xml_doc2 ) );
parser2.next(); // read the START document...
//null for the processor is all right here
//XMLLoader loader = new XMLLoader();
SolrInputDocument document2 = loader.readDoc( parser2 );
docs.add(document1);
docs.add(document2);
Collections.shuffle(docs, random());
req.add(docs);
RequestWriter requestWriter = new RequestWriter();
OutputStream os = new ByteArrayOutputStream();
requestWriter.write(req, os);
assertBlockU(os.toString());
assertU(commit());
final SolrIndexSearcher searcher = getSearcher();
assertSingleParentOf(searcher, one("yz"), "X");
assertSingleParentOf(searcher, one("bc"), "A");
}
@Test
public void testJavaBinCodec() throws IOException { //actually this test must be in other test class
SolrInputDocument topDocument = new SolrInputDocument();
topDocument.addField("parent_f1", "v1");
topDocument.addField("parent_f2", "v2");
int childsNum = atLeast(10);
for (int index = 0; index < childsNum; ++index) {
addChildren("child", topDocument, index, false);
}
ByteArrayOutputStream os = new ByteArrayOutputStream();
new JavaBinCodec().marshal(topDocument, os);
byte[] buffer = os.toByteArray();
//now read the Object back
InputStream is = new ByteArrayInputStream(buffer);
SolrInputDocument result = (SolrInputDocument) new JavaBinCodec().unmarshal(is);
assertEquals(2, result.size());
assertEquals("v1", result.getFieldValue("parent_f1"));
assertEquals("v2", result.getFieldValue("parent_f2"));
List<SolrInputDocument> resultChilds = result.getChildDocuments();
assertEquals(childsNum, resultChilds.size());
for (int childIndex = 0; childIndex < childsNum; ++childIndex) {
SolrInputDocument child = resultChilds.get(childIndex);
for (int fieldNum = 0; fieldNum < childIndex; ++fieldNum) {
assertEquals(childIndex + "value" + fieldNum, child.getFieldValue(childIndex + "child" + fieldNum));
}
List<SolrInputDocument> grandChilds = child.getChildDocuments();
assertEquals(childIndex * 2, grandChilds.size());
for (int grandIndex = 0; grandIndex < childIndex * 2; ++grandIndex) {
SolrInputDocument grandChild = grandChilds.get(grandIndex);
assertFalse(grandChild.hasChildDocuments());
for (int fieldNum = 0; fieldNum < grandIndex; ++fieldNum) {
assertEquals(grandIndex + "value" + fieldNum, grandChild.getFieldValue(grandIndex + "grand" + fieldNum));
}
}
}
}
private void addChildren(String prefix, SolrInputDocument topDocument, int childIndex, boolean lastLevel) {
SolrInputDocument childDocument = new SolrInputDocument();
for (int index = 0; index < childIndex; ++index) {
childDocument.addField(childIndex + prefix + index, childIndex + "value"+ index);
}
if (!lastLevel) {
for (int i = 0; i < childIndex * 2; ++i) {
addChildren("grand", childDocument, i, true);
}
}
topDocument.addChildDocument(childDocument);
}
/**
* on the given abcD it generates one parent doc, taking D from the tail and
* two subdocs relaitons ab and c uniq ids are supplied also
*
* <pre>
* {@code
* <add>
* <doc>
* <field name="parent_s">D</field>
* <doc>
* <field name="child_s">a</field>
* <field name="type_s">1</field>
* </doc>
* <doc>
* <field name="child_s">b</field>
* <field name="type_s">1</field>
* </doc>
* <doc>
* <field name="child_s">c</field>
* <field name="type_s">2</field>
* </doc>
* </doc>
* </add>
* }
* </pre>
* */
private Document block(String string) {
Document document = DocumentHelper.createDocument();
Element root = document.addElement("add");
Element doc = root.addElement("doc");
if (string.length() > 0) {
// last character is a top parent
attachField(doc, parent,
String.valueOf(string.charAt(string.length() - 1)));
attachField(doc, "id", id());
// add subdocs
int type = 1;
for (int i = 0; i < string.length() - 1; i += 2) {
String relation = string.substring(i,
Math.min(i + 2, string.length() - 1));
attachSubDocs(doc, relation, type);
type++;
}
}
return document;
}
private void attachSubDocs(Element parent, String relation, int typeValue) {
for (int j = 0; j < relation.length(); j++) {
Element document = parent.addElement("doc");
attachField(document, child, String.valueOf(relation.charAt(j)));
attachField(document, "id", id());
attachField(document, type, String.valueOf(typeValue));
}
}
/**
* Merges two documents like
*
* <pre>
* {@code <add>...</add> + <add>...</add> = <add>... + ...</add>}
* </pre>
*
* @param doc1
* first document
* @param doc2
* second document
* @return merged document
*/
private Document merge(Document doc1, Document doc2) {
List<Element> list = doc2.getRootElement().elements();
for (Element element : list) {
doc1.getRootElement().add(element.detach());
}
return doc1;
}
private void attachField(Element root, String fieldName, String value) {
Element field = root.addElement("field");
field.addAttribute("name", fieldName);
field.addText(value);
}
private static String id() {
return "" + counter.incrementAndGet();
}
private String one(String string) {
return "" + string.charAt(random().nextInt(string.length()));
}
protected void assertSingleParentOf(final SolrIndexSearcher searcher,
final String childTerm, String parentExp) throws IOException {
final TopDocs docs = searcher.search(join(childTerm), 10);
assertEquals(1, docs.totalHits);
final String pAct = searcher.doc(docs.scoreDocs[0].doc).get(parent);
assertEquals(parentExp, pAct);
}
protected ToParentBlockJoinQuery join(final String childTerm) {
return new ToParentBlockJoinQuery(
new TermQuery(new Term(child, childTerm)), new TermRangeFilter(parent,
null, null, false, false), ScoreMode.None);
}
private Collection<? extends Callable<Void>> callables(List<Document> blocks) {
final List<Callable<Void>> rez = new ArrayList<Callable<Void>>();
for (Document block : blocks) {
final String msg = block.asXML();
if (msg.length() > 0) {
rez.add(new Callable<Void>() {
@Override
public Void call() {
assertBlockU(msg);
return null;
}
});
if (rarely()) {
rez.add(new Callable<Void>() {
@Override
public Void call() {
assertBlockU(commit());
return null;
}
});
}
}
}
return rez;
}
private void assertBlockU(final String msg) {
assertBlockU(msg, "0");
}
private void assertFailedBlockU(final String msg) {
try {
assertBlockU(msg, "1");
fail("expecting fail");
} catch (Exception e) {
// gotcha
}
}
private void assertBlockU(final String msg, String expected) {
try {
String res = h.checkUpdateStatus(msg, expected);
if (res != null) {
fail("update was not successful: " + res + " expected: " + expected);
}
} catch (SAXException e) {
throw new RuntimeException("Invalid XML", e);
}
}
}

View File

@ -133,6 +133,11 @@ public class ClientUtils
}
}
}
for (SolrInputDocument childDocument : doc.getChildDocuments()) {
writeXML(childDocument, writer);
}
writer.write("</doc>");
}

View File

@ -18,9 +18,12 @@
package org.apache.solr.common;
import java.io.Serializable;
import java.util.ArrayList;
import java.util.Collection;
import java.util.Collections;
import java.util.Iterator;
import java.util.LinkedHashMap;
import java.util.List;
import java.util.Map;
import java.util.Set;
@ -36,13 +39,16 @@ public class SolrInputDocument implements Map<String,SolrInputField>, Iterable<S
{
private final Map<String,SolrInputField> _fields;
private float _documentBoost = 1.0f;
private List<SolrInputDocument> _childDocuments;
public SolrInputDocument() {
_fields = new LinkedHashMap<String,SolrInputField>();
_childDocuments = new ArrayList<SolrInputDocument>();
}
public SolrInputDocument(Map<String,SolrInputField> fields) {
_fields = fields;
_childDocuments = new ArrayList<SolrInputDocument>();
}
/**
@ -54,6 +60,9 @@ public class SolrInputDocument implements Map<String,SolrInputField>, Iterable<S
if( _fields != null ) {
_fields.clear();
}
if (_childDocuments != null) {
_childDocuments.clear();
}
}
///////////////////////////////////////////////////////////////////
@ -189,7 +198,7 @@ public class SolrInputDocument implements Map<String,SolrInputField>, Iterable<S
@Override
public String toString()
{
return "SolrInputDocument" + _fields.values();
return "SolrInputDocument(fields: " + _fields.values() + ", childs: " + _childDocuments + ")";
}
public SolrInputDocument deepCopy() {
@ -199,6 +208,12 @@ public class SolrInputDocument implements Map<String,SolrInputField>, Iterable<S
clone._fields.put(fieldEntry.getKey(), fieldEntry.getValue().deepCopy());
}
clone._documentBoost = _documentBoost;
clone._childDocuments = new ArrayList<SolrInputDocument>(_childDocuments.size());
for (SolrInputDocument child : _childDocuments) {
clone._childDocuments.add(child.deepCopy());
}
return clone;
}
@ -260,4 +275,23 @@ public class SolrInputDocument implements Map<String,SolrInputField>, Iterable<S
public Collection<SolrInputField> values() {
return _fields.values();
}
public void addChildDocument(SolrInputDocument child) {
_childDocuments.add(child);
}
public void addChildDocuments(Collection<SolrInputDocument> childs) {
for (SolrInputDocument child : childs) {
addChildDocument(child);
}
}
public List<SolrInputDocument> getChildDocuments() {
return _childDocuments;
}
public boolean hasChildDocuments() {
boolean isEmpty = (_childDocuments == null || _childDocuments.isEmpty());
return !isEmpty;
}
}

View File

@ -62,6 +62,7 @@ public class JavaBinCodec {
END = 15,
SOLRINPUTDOC = 16,
SOLRINPUTDOC_CHILDS = 17,
// types that combine tag + length (or other info) in a single byte
TAG_AND_LEN = (byte) (1 << 5),
@ -358,6 +359,8 @@ public class JavaBinCodec {
public SolrInputDocument readSolrInputDocument(DataInputInputStream dis) throws IOException {
int sz = readVInt(dis);
dis.readByte(); // skip childDocuments tag
int childsSize = readVInt(dis);
float docBoost = (Float)readVal(dis);
SolrInputDocument sdoc = new SolrInputDocument();
sdoc.setDocumentBoost(docBoost);
@ -374,11 +377,17 @@ public class JavaBinCodec {
Object fieldVal = readVal(dis);
sdoc.setField(fieldName, fieldVal, boost);
}
for (int i = 0; i < childsSize; i++) {
dis.readByte(); // skip solrinputdoc tag
SolrInputDocument child = readSolrInputDocument(dis);
sdoc.addChildDocument(child);
}
return sdoc;
}
public void writeSolrInputDocument(SolrInputDocument sdoc) throws IOException {
writeTag(SOLRINPUTDOC, sdoc.size());
writeTag(SOLRINPUTDOC_CHILDS, sdoc.getChildDocuments().size());
writeFloat(sdoc.getDocumentBoost());
for (SolrInputField inputField : sdoc.values()) {
if (inputField.getBoost() != 1.0f) {
@ -387,6 +396,9 @@ public class JavaBinCodec {
writeExternString(inputField.getName());
writeVal(inputField.getValue());
}
for (SolrInputDocument child : sdoc.getChildDocuments()) {
writeSolrInputDocument(child);
}
}

View File

@ -72,6 +72,7 @@ import java.io.File;
import java.io.IOException;
import java.io.StringWriter;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Comparator;
import java.util.HashMap;
import java.util.HashSet;
@ -969,6 +970,10 @@ public abstract class SolrTestCaseJ4 extends LuceneTestCase {
return sd;
}
public static List<SolrInputDocument> sdocs(SolrInputDocument... docs) {
return Arrays.asList(docs);
}
/** Converts "test JSON" and returns standard JSON.
* Currently this only consists of changing unescaped single quotes to double quotes,
* and escaped single quotes to single quotes.
@ -1527,7 +1532,7 @@ public abstract class SolrTestCaseJ4 extends LuceneTestCase {
}
/** Return a Map from field value to a list of document ids */
Map<Comparable, List<Comparable>> invertField(Map<Comparable, Doc> model, String field) {
public Map<Comparable, List<Comparable>> invertField(Map<Comparable, Doc> model, String field) {
Map<Comparable, List<Comparable>> value_to_id = new HashMap<Comparable, List<Comparable>>();
// invert field