lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

Introducing HoodieLogFormat V2 with versioning support

Browse Source 
- HoodieLogFormat V2 has support for LogFormat evolution through versioning
			- LogVersion is associated with a LogBlock not a LogFile
			- Based on a version for a LogBlock, approporiate code path is executed
		- Implemented LazyReading of Hoodie Log Blocks with Memory / IO tradeoff
		- Implemented Reverse pointer to be able to traverse the log in reverse
		- Introduce new MAGIC for backwards compatibility with logs without versions
This commit is contained in:
Nishith Agarwal
2018-02-15 11:01:25 -08:00
committed by vinoth chandar
parent dfd1979c51
commit 5405a6287b
32 changed files with 2066 additions and 677 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
hoodie-client/src/main/java/com/uber/hoodie/io/HoodieAppendHandle.java
View File

@@ -159,11 +159,14 @@ public class HoodieAppendHandle<T extends HoodieRecordPayload> extends HoodieIOH
return Optional.empty();
}
// TODO (NA) - Perform a schema check of current input record with the last schema on log file
// to make sure we don't append records with older (shorter) schema than already appended
public void doAppend() {
int maxBlockSize = config.getLogFileDataBlockMaxSize(); int numberOfRecords = 0;
Map<HoodieLogBlock.LogMetadataType, String> metadata = Maps.newHashMap();
metadata.put(HoodieLogBlock.LogMetadataType.INSTANT_TIME, commitTime);
Map<HoodieLogBlock.HeaderMetadataType, String> header = Maps.newHashMap();
header.put(HoodieLogBlock.HeaderMetadataType.INSTANT_TIME, commitTime);
header.put(HoodieLogBlock.HeaderMetadataType.SCHEMA, schema.toString());
while (recordItr.hasNext()) {
HoodieRecord record = recordItr.next();
// update the new location of the record, so we know where to find it next
@@ -178,7 +181,7 @@ public class HoodieAppendHandle<T extends HoodieRecordPayload> extends HoodieIOH
// Recompute averageRecordSize before writing a new block and update existing value with avg of new and old
logger.info("AvgRecordSize => " + averageRecordSize);
averageRecordSize = (averageRecordSize + SizeEstimator.estimate(record))/2;
doAppend(metadata);
doAppend(header);
numberOfRecords = 0;
}
Optional<IndexedRecord> indexedRecord = getIndexedRecord(record);
@@ -189,18 +192,18 @@ public class HoodieAppendHandle<T extends HoodieRecordPayload> extends HoodieIOH
}
numberOfRecords++;
}
doAppend(metadata);
doAppend(header);
}
private void doAppend(Map<HoodieLogBlock.LogMetadataType, String> metadata) {
private void doAppend(Map<HoodieLogBlock.HeaderMetadataType, String> header) {
try {
if (recordList.size() > 0) {
writer = writer.appendBlock(new HoodieAvroDataBlock(recordList, schema, metadata));
writer = writer.appendBlock(new HoodieAvroDataBlock(recordList, header));
recordList.clear();
}
if (keysToDelete.size() > 0) {
writer = writer.appendBlock(
new HoodieDeleteBlock(keysToDelete.stream().toArray(String[]::new), metadata));
new HoodieDeleteBlock(keysToDelete.stream().toArray(String[]::new), header));
keysToDelete.clear();
}
} catch (Exception e) {

8
hoodie-client/src/main/java/com/uber/hoodie/io/HoodieCommitArchiveLog.java
View File

@@ -18,6 +18,7 @@ package com.uber.hoodie.io;
import com.fasterxml.jackson.databind.DeserializationFeature;
import com.fasterxml.jackson.databind.ObjectMapper;
import com.google.common.collect.Maps;
import com.google.common.collect.Sets;
import com.uber.hoodie.avro.model.HoodieArchivedMetaEntry;
import com.uber.hoodie.avro.model.HoodieCleanMetadata;
@@ -30,6 +31,7 @@ import com.uber.hoodie.common.table.HoodieTableMetaClient;
import com.uber.hoodie.common.table.HoodieTimeline;
import com.uber.hoodie.common.table.log.HoodieLogFormat;
import com.uber.hoodie.common.table.log.block.HoodieAvroDataBlock;
import com.uber.hoodie.common.table.log.block.HoodieLogBlock;
import com.uber.hoodie.common.table.timeline.HoodieArchivedTimeline;
import com.uber.hoodie.common.table.timeline.HoodieInstant;
import com.uber.hoodie.common.util.AvroUtils;
@@ -39,6 +41,7 @@ import com.uber.hoodie.exception.HoodieException;
import com.uber.hoodie.exception.HoodieIOException;
import com.uber.hoodie.table.HoodieTable;
import org.apache.avro.Schema;
import org.apache.avro.file.DataFileStream;
import org.apache.avro.generic.IndexedRecord;
import org.apache.hadoop.fs.Path;
import org.apache.log4j.LogManager;
@@ -47,6 +50,7 @@ import org.apache.log4j.Logger;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import java.util.Optional;
import java.util.stream.Collectors;
import java.util.stream.Stream;
@@ -190,7 +194,9 @@ public class HoodieCommitArchiveLog {
for (HoodieInstant hoodieInstant : instants) {
records.add(convertToAvroRecord(commitTimeline, hoodieInstant));
}
HoodieAvroDataBlock block = new HoodieAvroDataBlock(records, wrapperSchema);
Map<HoodieLogBlock.HeaderMetadataType, String> header = Maps.newHashMap();
header.put(HoodieLogBlock.HeaderMetadataType.SCHEMA, wrapperSchema.toString());
HoodieAvroDataBlock block = new HoodieAvroDataBlock(records, header);
this.writer = writer.appendBlock(block);
} catch (Exception e) {
throw new HoodieCommitException("Failed to archive commits", e);

3
hoodie-client/src/main/java/com/uber/hoodie/io/compact/HoodieRealtimeTableCompactor.java
View File

@@ -154,7 +154,8 @@ public class HoodieRealtimeTableCompactor implements HoodieCompactor {
HoodieCompactedLogRecordScanner scanner = new HoodieCompactedLogRecordScanner(fs,
metaClient.getBasePath(),
operation.getDeltaFilePaths(), readerSchema, maxInstantTime, config.getMaxMemorySizePerCompactionInBytes());
operation.getDeltaFilePaths(), readerSchema, maxInstantTime, config.getMaxMemorySizePerCompactionInBytes(),
config.getCompactionLazyBlockReadEnabled(), config.getCompactionReverseLogReadEnabled());
if (!scanner.iterator().hasNext()) {
return Lists.newArrayList();
}