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

LUCENE-3461: catch same doc-values field added twice in one doc 

git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1175935 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Michael McCandless 2011-09-26 16:20:10 +00:00
parent f5d625821f
commit e1d1d34e56
2 changed files with 89 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
lucene/src/java/org/apache/lucene/index/DocFieldProcessor.java
View File

@ -83,9 +83,9 @@ final class DocFieldProcessor extends DocConsumer {
// FieldInfo.storePayload.
final String fileName = IndexFileNames.segmentFileName(state.segmentName, "", IndexFileNames.FIELD_INFOS_EXTENSION);
state.fieldInfos.write(state.directory, fileName);
for (DocValuesConsumer consumers : docValues.values()) {
consumers.finish(state.numDocs);
};
for (DocValuesConsumerAndDocID consumers : docValues.values()) {
consumers.docValuesConsumer.finish(state.numDocs);
}
// close perDocConsumer during flush to ensure all files are flushed due to PerCodec CFS
IOUtils.close(perDocConsumers.values());
}
@ -297,14 +297,28 @@ final class DocFieldProcessor extends DocConsumer {
}
}
final private Map<String, DocValuesConsumer> docValues = new HashMap<String, DocValuesConsumer>();
private static class DocValuesConsumerAndDocID {
public int docID;
final DocValuesConsumer docValuesConsumer;
public DocValuesConsumerAndDocID(DocValuesConsumer docValuesConsumer) {
this.docValuesConsumer = docValuesConsumer;
}
}
final private Map<String, DocValuesConsumerAndDocID> docValues = new HashMap<String, DocValuesConsumerAndDocID>();
final private Map<Integer, PerDocConsumer> perDocConsumers = new HashMap<Integer, PerDocConsumer>();
DocValuesConsumer docValuesConsumer(DocState docState, FieldInfo fieldInfo)
throws IOException {
DocValuesConsumer docValuesConsumer = docValues.get(fieldInfo.name);
if (docValuesConsumer != null) {
return docValuesConsumer;
DocValuesConsumerAndDocID docValuesConsumerAndDocID = docValues.get(fieldInfo.name);
if (docValuesConsumerAndDocID != null) {
if (docState.docID == docValuesConsumerAndDocID.docID) {
throw new IllegalArgumentException("IndexDocValuesField \"" + fieldInfo.name + "\" appears more than once in this document (only one value is allowed, per field)");
}
assert docValuesConsumerAndDocID.docID < docState.docID;
docValuesConsumerAndDocID.docID = docState.docID;
return docValuesConsumerAndDocID.docValuesConsumer;
}
PerDocConsumer perDocConsumer = perDocConsumers.get(fieldInfo.getCodecId());
if (perDocConsumer == null) {
@ -316,6 +330,7 @@ final class DocFieldProcessor extends DocConsumer {
perDocConsumers.put(Integer.valueOf(fieldInfo.getCodecId()), perDocConsumer);
}
boolean success = false;
DocValuesConsumer docValuesConsumer = null;
try {
docValuesConsumer = perDocConsumer.addValuesField(fieldInfo);
fieldInfo.commitDocValues();
@ -325,7 +340,10 @@ final class DocFieldProcessor extends DocConsumer {
fieldInfo.revertUncommitted();
}
}
docValues.put(fieldInfo.name, docValuesConsumer);
docValuesConsumerAndDocID = new DocValuesConsumerAndDocID(docValuesConsumer);
docValuesConsumerAndDocID.docID = docState.docID;
docValues.put(fieldInfo.name, docValuesConsumerAndDocID);
return docValuesConsumer;
}
}

65
lucene/src/test/org/apache/lucene/index/values/TestDocValuesIndexing.java
View File

@ -25,9 +25,9 @@ import java.util.EnumSet;
import java.util.List;
import org.apache.lucene.analysis.MockAnalyzer;
import org.apache.lucene.document.IndexDocValuesField;
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.apache.lucene.document.IndexDocValuesField;
import org.apache.lucene.document.StringField;
import org.apache.lucene.document.TextField;
import org.apache.lucene.index.CorruptIndexException;
@ -37,6 +37,7 @@ import org.apache.lucene.index.IndexWriterConfig;
import org.apache.lucene.index.LogDocMergePolicy;
import org.apache.lucene.index.LogMergePolicy;
import org.apache.lucene.index.MultiPerDocValues;
import org.apache.lucene.index.RandomIndexWriter;
import org.apache.lucene.index.Term;
import org.apache.lucene.index.codecs.CodecProvider;
import org.apache.lucene.index.codecs.PerDocValues;
@ -45,10 +46,10 @@ import org.apache.lucene.search.*;
import org.apache.lucene.store.Directory;
import org.apache.lucene.store.LockObtainFailedException;
import org.apache.lucene.util.BytesRef;
import org.apache.lucene.util.FixedBitSet;
import org.apache.lucene.util.FloatsRef;
import org.apache.lucene.util.LongsRef;
import org.apache.lucene.util.LuceneTestCase;
import org.apache.lucene.util.FixedBitSet;
import org.apache.lucene.util._TestUtil;
import org.junit.Before;
@ -562,4 +563,64 @@ public class TestDocValuesIndexing extends LuceneTestCase {
}
return deleted;
}
public void testMultiValuedIndexDocValuesField() throws Exception {
assumeFalse("cannot work with preflex codec", CodecProvider.getDefault().getDefaultFieldCodec().equals("PreFlex"));
Directory d = newDirectory();
RandomIndexWriter w = new RandomIndexWriter(random, d);
Document doc = new Document();
IndexDocValuesField f = new IndexDocValuesField("field");
f.setInt(17);
// Index doc values are single-valued so we should not
// be able to add same field more than once:
doc.add(f);
doc.add(f);
try {
w.addDocument(doc);
fail("didn't hit expected exception");
} catch (IllegalArgumentException iae) {
// expected
}
doc = new Document();
doc.add(f);
w.addDocument(doc);
w.optimize();
IndexReader r = w.getReader();
w.close();
assertEquals(17, r.getSequentialSubReaders()[0].perDocValues().docValues("field").load().getInt(0));
r.close();
d.close();
}
public void testDifferentTypedDocValuesField() throws Exception {
assumeFalse("cannot work with preflex codec", CodecProvider.getDefault().getDefaultFieldCodec().equals("PreFlex"));
Directory d = newDirectory();
RandomIndexWriter w = new RandomIndexWriter(random, d);
Document doc = new Document();
IndexDocValuesField f = new IndexDocValuesField("field");
f.setInt(17);
// Index doc values are single-valued so we should not
// be able to add same field more than once:
doc.add(f);
IndexDocValuesField f2 = new IndexDocValuesField("field");
f2.setFloat(22.0);
doc.add(f2);
try {
w.addDocument(doc);
fail("didn't hit expected exception");
} catch (IllegalArgumentException iae) {
// expected
}
doc = new Document();
doc.add(f);
w.addDocument(doc);
w.optimize();
IndexReader r = w.getReader();
w.close();
assertEquals(17, r.getSequentialSubReaders()[0].perDocValues().docValues("field").load().getInt(0));
r.close();
d.close();
}
}