jpountz commented on a change in pull request #1234: Add compression for Binary
doc value fields
URL: https://github.com/apache/lucene-solr/pull/1234#discussion_r377621003
##########
File path:
lucene/core/src/java/org/apache/lucene/codecs/lucene80/Lucene80DocValuesConsumer.java
##########
@@ -353,67 +360,193 @@ private void writeBlock(long[] values, int length, long
gcd, ByteBuffersDataOutp
}
}
- @Override
- public void addBinaryField(FieldInfo field, DocValuesProducer
valuesProducer) throws IOException {
- meta.writeInt(field.number);
- meta.writeByte(Lucene80DocValuesFormat.BINARY);
-
- BinaryDocValues values = valuesProducer.getBinary(field);
- long start = data.getFilePointer();
- meta.writeLong(start); // dataOffset
- int numDocsWithField = 0;
- int minLength = Integer.MAX_VALUE;
- int maxLength = 0;
- for (int doc = values.nextDoc(); doc != DocIdSetIterator.NO_MORE_DOCS; doc
= values.nextDoc()) {
- numDocsWithField++;
- BytesRef v = values.binaryValue();
- int length = v.length;
- data.writeBytes(v.bytes, v.offset, v.length);
- minLength = Math.min(length, minLength);
- maxLength = Math.max(length, maxLength);
+ class CompressedBinaryBlockWriter implements Closeable {
+ FastCompressionHashTable ht = new LZ4.FastCompressionHashTable();
+ int uncompressedBlockLength = 0;
+ int maxUncompressedBlockLength = 0;
+ int numDocsInCurrentBlock = 0;
+ int[] docLengths = new
int[Lucene80DocValuesFormat.BINARY_DOCS_PER_COMPRESSED_BLOCK];
+ byte[] block = new byte [1024 * 16];
+ int totalChunks = 0;
+ long maxPointer = 0;
+ long blockAddressesStart = -1;
+
+ private IndexOutput tempBinaryOffsets;
+
+
+ public CompressedBinaryBlockWriter() throws IOException {
+ tempBinaryOffsets =
state.directory.createTempOutput(state.segmentInfo.name, "binary_pointers",
state.context);
+ boolean success = false;
+ try {
+ CodecUtil.writeHeader(tempBinaryOffsets,
Lucene80DocValuesFormat.META_CODEC + "FilePointers",
Lucene80DocValuesFormat.VERSION_CURRENT);
+ success = true;
+ } finally {
+ if (success == false) {
+ IOUtils.closeWhileHandlingException(this); //self-close because
constructor caller can't
+ }
+ }
}
- assert numDocsWithField <= maxDoc;
- meta.writeLong(data.getFilePointer() - start); // dataLength
- if (numDocsWithField == 0) {
- meta.writeLong(-2); // docsWithFieldOffset
- meta.writeLong(0L); // docsWithFieldLength
- meta.writeShort((short) -1); // jumpTableEntryCount
- meta.writeByte((byte) -1); // denseRankPower
- } else if (numDocsWithField == maxDoc) {
- meta.writeLong(-1); // docsWithFieldOffset
- meta.writeLong(0L); // docsWithFieldLength
- meta.writeShort((short) -1); // jumpTableEntryCount
- meta.writeByte((byte) -1); // denseRankPower
- } else {
- long offset = data.getFilePointer();
- meta.writeLong(offset); // docsWithFieldOffset
- values = valuesProducer.getBinary(field);
- final short jumpTableEntryCount = IndexedDISI.writeBitSet(values, data,
IndexedDISI.DEFAULT_DENSE_RANK_POWER);
- meta.writeLong(data.getFilePointer() - offset); // docsWithFieldLength
- meta.writeShort(jumpTableEntryCount);
- meta.writeByte(IndexedDISI.DEFAULT_DENSE_RANK_POWER);
+ void addDoc(int doc, BytesRef v) throws IOException {
+ if (blockAddressesStart < 0) {
+ blockAddressesStart = data.getFilePointer();
+ }
+ docLengths[numDocsInCurrentBlock] = v.length;
+ block = ArrayUtil.grow(block, uncompressedBlockLength + v.length);
+ System.arraycopy(v.bytes, v.offset, block, uncompressedBlockLength,
v.length);
+ uncompressedBlockLength += v.length;
+ numDocsInCurrentBlock++;
+ if (numDocsInCurrentBlock ==
Lucene80DocValuesFormat.BINARY_DOCS_PER_COMPRESSED_BLOCK) {
+ flushData();
+ }
}
- meta.writeInt(numDocsWithField);
- meta.writeInt(minLength);
- meta.writeInt(maxLength);
- if (maxLength > minLength) {
- start = data.getFilePointer();
- meta.writeLong(start);
+ private void flushData() throws IOException {
+ if (numDocsInCurrentBlock > 0) {
+ // Write offset to this block to temporary offsets file
+ totalChunks++;
+ long thisBlockStartPointer = data.getFilePointer();
+
+ // Optimisation - check if all lengths are same
+ boolean allLengthsSame = true && numDocsInCurrentBlock >0 ;
+ for (int i = 0; i <
Lucene80DocValuesFormat.BINARY_DOCS_PER_COMPRESSED_BLOCK && allLengthsSame;
i++) {
+ if (i > 0 && docLengths[i] != docLengths[i-1]) {
+ allLengthsSame = false;
+ }
+ }
+ if (allLengthsSame) {
+ // Only write one value shifted. Steal a bit to indicate all other
lengths are the same
+ int onlyOneLength = (docLengths[0] <<1) | 1;
+ data.writeVInt(onlyOneLength);
+ } else {
+ for (int i = 0; i <
Lucene80DocValuesFormat.BINARY_DOCS_PER_COMPRESSED_BLOCK; i++) {
+ if(i == 0) {
+ // Write first value shifted and steal a bit to indicate other
lengths are to follow
+ int multipleLengths = (docLengths[0] <<1);
+ data.writeVInt(multipleLengths);
+ } else {
+ data.writeVInt(docLengths[i]);
+ }
+ }
+ }
+ maxUncompressedBlockLength = Math.max(maxUncompressedBlockLength,
uncompressedBlockLength);
+ LZ4.compress(block, 0, uncompressedBlockLength, data, ht);
+ numDocsInCurrentBlock = 0;
+ // Ensure initialized with zeroes because full array is always written
+ Arrays.fill(docLengths, 0);
+ uncompressedBlockLength = 0;
+ maxPointer = data.getFilePointer();
+ tempBinaryOffsets.writeVLong(maxPointer - thisBlockStartPointer);
+ }
+ }
+
+ void writeMetaData() throws IOException {
+ if (blockAddressesStart < 0 ) {
+ return;
+ }
+
+ long startDMW = data.getFilePointer();
+ meta.writeLong(startDMW);
+
+ meta.writeVInt(totalChunks);
+ meta.writeVInt(Lucene80DocValuesFormat.BINARY_DOCS_PER_COMPRESSED_BLOCK);
+ meta.writeVInt(maxUncompressedBlockLength);
meta.writeVInt(DIRECT_MONOTONIC_BLOCK_SHIFT);
+
+
+ CodecUtil.writeFooter(tempBinaryOffsets);
+ IOUtils.close(tempBinaryOffsets);
+ //write the compressed block offsets info to the meta file by reading
from temp file
+ try (ChecksumIndexInput filePointersIn =
state.directory.openChecksumInput(tempBinaryOffsets.getName(),
IOContext.READONCE)) {
+ CodecUtil.checkHeader(filePointersIn,
Lucene80DocValuesFormat.META_CODEC + "FilePointers",
Lucene80DocValuesFormat.VERSION_CURRENT,
+ Lucene80DocValuesFormat.VERSION_CURRENT);
+ Throwable priorE = null;
+ try {
+ final DirectMonotonicWriter filePointers =
DirectMonotonicWriter.getInstance(meta, data, totalChunks,
DIRECT_MONOTONIC_BLOCK_SHIFT);
+ long fp = blockAddressesStart;
+ for (int i = 0; i < totalChunks; ++i) {
+ filePointers.add(fp);
+ fp += filePointersIn.readVLong();
+ }
+ if (maxPointer < fp) {
+ throw new CorruptIndexException("File pointers don't add up
("+fp+" vs expected "+maxPointer+")", filePointersIn);
+ }
+ filePointers.finish();
+ } catch (Throwable e) {
+ priorE = e;
+ } finally {
+ CodecUtil.checkFooter(filePointersIn, priorE);
+ }
+ }
+ // Write the length of the DMW block in the data
+ meta.writeLong(data.getFilePointer() - startDMW);
+ }
- final DirectMonotonicWriter writer =
DirectMonotonicWriter.getInstance(meta, data, numDocsWithField + 1,
DIRECT_MONOTONIC_BLOCK_SHIFT);
- long addr = 0;
- writer.add(addr);
- values = valuesProducer.getBinary(field);
+ @Override
+ public void close() throws IOException {
+ if (tempBinaryOffsets != null) {
+ IOUtils.close(tempBinaryOffsets);
+ state.directory.deleteFile(tempBinaryOffsets.getName());
+ tempBinaryOffsets = null;
+ }
+ }
+
+ }
+
+
+ @Override
+ public void addBinaryField(FieldInfo field, DocValuesProducer
valuesProducer) throws IOException {
+ meta.writeInt(field.number);
+ meta.writeByte(Lucene80DocValuesFormat.BINARY);
+
+ try (CompressedBinaryBlockWriter blockWriter = new
CompressedBinaryBlockWriter()){
+ BinaryDocValues values = valuesProducer.getBinary(field);
+ long start = data.getFilePointer();
+ meta.writeLong(start); // dataOffset
Review comment:
I like the idea but would prefer doing it in a separate PR.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
