org.apache.lucene.codecs.memory.DirectDocValuesConsumer Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of lucene-codecs Show documentation
Show all versions of lucene-codecs Show documentation
Apache Lucene (module: codecs)
package org.apache.lucene.codecs.memory;
/*
* 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.io.IOException;
import java.util.Iterator;
import org.apache.lucene.codecs.CodecUtil;
import org.apache.lucene.codecs.DocValuesConsumer;
import org.apache.lucene.index.FieldInfo;
import org.apache.lucene.index.IndexFileNames;
import org.apache.lucene.index.SegmentWriteState;
import org.apache.lucene.store.IndexOutput;
import org.apache.lucene.util.BytesRef;
import org.apache.lucene.util.IOUtils;
import static org.apache.lucene.codecs.memory.DirectDocValuesProducer.VERSION_CURRENT;
import static org.apache.lucene.codecs.memory.DirectDocValuesProducer.BYTES;
import static org.apache.lucene.codecs.memory.DirectDocValuesProducer.SORTED;
import static org.apache.lucene.codecs.memory.DirectDocValuesProducer.SORTED_SET;
import static org.apache.lucene.codecs.memory.DirectDocValuesProducer.NUMBER;
/**
* Writer for {@link DirectDocValuesFormat}
*/
class DirectDocValuesConsumer extends DocValuesConsumer {
final IndexOutput data, meta;
final int maxDoc;
DirectDocValuesConsumer(SegmentWriteState state, String dataCodec, String dataExtension, String metaCodec, String metaExtension) throws IOException {
maxDoc = state.segmentInfo.getDocCount();
boolean success = false;
try {
String dataName = IndexFileNames.segmentFileName(state.segmentInfo.name, state.segmentSuffix, dataExtension);
data = state.directory.createOutput(dataName, state.context);
CodecUtil.writeHeader(data, dataCodec, VERSION_CURRENT);
String metaName = IndexFileNames.segmentFileName(state.segmentInfo.name, state.segmentSuffix, metaExtension);
meta = state.directory.createOutput(metaName, state.context);
CodecUtil.writeHeader(meta, metaCodec, VERSION_CURRENT);
success = true;
} finally {
if (!success) {
IOUtils.closeWhileHandlingException(this);
}
}
}
@Override
public void addNumericField(FieldInfo field, Iterable values) throws IOException {
meta.writeVInt(field.number);
meta.writeByte(NUMBER);
addNumericFieldValues(field, values);
}
private void addNumericFieldValues(FieldInfo field, Iterable values) throws IOException {
meta.writeLong(data.getFilePointer());
long minValue = Long.MAX_VALUE;
long maxValue = Long.MIN_VALUE;
boolean missing = false;
long count = 0;
for (Number nv : values) {
if (nv != null) {
long v = nv.longValue();
minValue = Math.min(minValue, v);
maxValue = Math.max(maxValue, v);
} else {
missing = true;
}
count++;
if (count >= DirectDocValuesFormat.MAX_SORTED_SET_ORDS) {
throw new IllegalArgumentException("DocValuesField \"" + field.name + "\" is too large, must be <= " + DirectDocValuesFormat.MAX_SORTED_SET_ORDS + " values/total ords");
}
}
meta.writeInt((int) count);
if (missing) {
long start = data.getFilePointer();
writeMissingBitset(values);
meta.writeLong(start);
meta.writeLong(data.getFilePointer() - start);
} else {
meta.writeLong(-1L);
}
byte byteWidth;
if (minValue >= Byte.MIN_VALUE && maxValue <= Byte.MAX_VALUE) {
byteWidth = 1;
} else if (minValue >= Short.MIN_VALUE && maxValue <= Short.MAX_VALUE) {
byteWidth = 2;
} else if (minValue >= Integer.MIN_VALUE && maxValue <= Integer.MAX_VALUE) {
byteWidth = 4;
} else {
byteWidth = 8;
}
meta.writeByte(byteWidth);
for (Number nv : values) {
long v;
if (nv != null) {
v = nv.longValue();
} else {
v = 0;
}
switch(byteWidth) {
case 1:
data.writeByte((byte) v);
break;
case 2:
data.writeShort((short) v);
break;
case 4:
data.writeInt((int) v);
break;
case 8:
data.writeLong(v);
break;
}
}
}
@Override
public void close() throws IOException {
boolean success = false;
try {
if (meta != null) {
meta.writeVInt(-1); // write EOF marker
}
success = true;
} finally {
if (success) {
IOUtils.close(data, meta);
} else {
IOUtils.closeWhileHandlingException(data, meta);
}
}
}
@Override
public void addBinaryField(FieldInfo field, final Iterable values) throws IOException {
meta.writeVInt(field.number);
meta.writeByte(BYTES);
addBinaryFieldValues(field, values);
}
private void addBinaryFieldValues(FieldInfo field, final Iterable values) throws IOException {
// write the byte[] data
final long startFP = data.getFilePointer();
boolean missing = false;
long totalBytes = 0;
int count = 0;
for(BytesRef v : values) {
if (v != null) {
data.writeBytes(v.bytes, v.offset, v.length);
totalBytes += v.length;
if (totalBytes > DirectDocValuesFormat.MAX_TOTAL_BYTES_LENGTH) {
throw new IllegalArgumentException("DocValuesField \"" + field.name + "\" is too large, cannot have more than DirectDocValuesFormat.MAX_TOTAL_BYTES_LENGTH (" + DirectDocValuesFormat.MAX_TOTAL_BYTES_LENGTH + ") bytes");
}
} else {
missing = true;
}
count++;
}
meta.writeLong(startFP);
meta.writeInt((int) totalBytes);
meta.writeInt(count);
if (missing) {
long start = data.getFilePointer();
writeMissingBitset(values);
meta.writeLong(start);
meta.writeLong(data.getFilePointer() - start);
} else {
meta.writeLong(-1L);
}
int addr = 0;
for (BytesRef v : values) {
data.writeInt(addr);
if (v != null) {
addr += v.length;
}
}
data.writeInt(addr);
}
// TODO: in some cases representing missing with minValue-1 wouldn't take up additional space and so on,
// but this is very simple, and algorithms only check this for values of 0 anyway (doesnt slow down normal decode)
void writeMissingBitset(Iterable values) throws IOException {
long bits = 0;
int count = 0;
for (Object v : values) {
if (count == 64) {
data.writeLong(bits);
count = 0;
bits = 0;
}
if (v != null) {
bits |= 1L << (count & 0x3f);
}
count++;
}
if (count > 0) {
data.writeLong(bits);
}
}
@Override
public void addSortedField(FieldInfo field, Iterable values, Iterable docToOrd) throws IOException {
meta.writeVInt(field.number);
meta.writeByte(SORTED);
// write the ordinals as numerics
addNumericFieldValues(field, docToOrd);
// write the values as binary
addBinaryFieldValues(field, values);
}
// note: this might not be the most efficient... but its fairly simple
@Override
public void addSortedSetField(FieldInfo field, Iterable values, final Iterable docToOrdCount, final Iterable ords) throws IOException {
meta.writeVInt(field.number);
meta.writeByte(SORTED_SET);
// First write docToOrdCounts, except we "aggregate" the
// counts so they turn into addresses, and add a final
// value = the total aggregate:
addNumericFieldValues(field, new Iterable() {
// Just aggregates the count values so they become
// "addresses", and adds one more value in the end
// (the final sum):
@Override
public Iterator iterator() {
final Iterator iter = docToOrdCount.iterator();
return new Iterator() {
long sum;
boolean ended;
@Override
public boolean hasNext() {
return iter.hasNext() || !ended;
}
@Override
public Number next() {
long toReturn = sum;
if (iter.hasNext()) {
Number n = iter.next();
if (n != null) {
sum += n.longValue();
}
} else if (!ended) {
ended = true;
} else {
assert false;
}
return toReturn;
}
@Override
public void remove() {
throw new UnsupportedOperationException();
}
};
}
});
// Write ordinals for all docs, appended into one big
// numerics:
addNumericFieldValues(field, ords);
// write the values as binary
addBinaryFieldValues(field, values);
}
}