1
0

[HUDI-4249] Fixing in-memory HoodieData implementation to operate lazily (#5855)

This commit is contained in:
Alexey Kudinkin
2022-07-16 16:26:48 -07:00
committed by GitHub
parent 80368a049d
commit 4bda6afe0b
35 changed files with 868 additions and 555 deletions

View File

@@ -20,7 +20,7 @@ package org.apache.hudi.client;
import org.apache.hudi.client.common.HoodieJavaEngineContext;
import org.apache.hudi.client.embedded.EmbeddedTimelineService;
import org.apache.hudi.common.data.HoodieList;
import org.apache.hudi.common.data.HoodieListData;
import org.apache.hudi.common.engine.HoodieEngineContext;
import org.apache.hudi.common.model.HoodieCommitMetadata;
import org.apache.hudi.common.model.HoodieKey;
@@ -67,8 +67,7 @@ public class HoodieJavaWriteClient<T extends HoodieRecordPayload> extends
// Create a Hoodie table which encapsulated the commits and files visible
HoodieJavaTable<T> table = HoodieJavaTable.create(config, (HoodieJavaEngineContext) context);
Timer.Context indexTimer = metrics.getIndexCtx();
List<HoodieRecord<T>> recordsWithLocation = HoodieList.getList(
getIndex().tagLocation(HoodieList.of(hoodieRecords), context, table));
List<HoodieRecord<T>> recordsWithLocation = getIndex().tagLocation(HoodieListData.eager(hoodieRecords), context, table).collectAsList();
metrics.updateIndexMetrics(LOOKUP_STR, metrics.getDurationInMs(indexTimer == null ? 0L : indexTimer.stop()));
return recordsWithLocation.stream().filter(v1 -> !v1.isCurrentLocationKnown()).collect(Collectors.toList());
}

View File

@@ -25,7 +25,7 @@ import org.apache.hudi.avro.model.HoodieClusteringPlan;
import org.apache.hudi.client.WriteStatus;
import org.apache.hudi.client.common.JavaTaskContextSupplier;
import org.apache.hudi.common.data.HoodieData;
import org.apache.hudi.common.data.HoodieList;
import org.apache.hudi.common.data.HoodieListData;
import org.apache.hudi.common.engine.HoodieEngineContext;
import org.apache.hudi.common.model.ClusteringOperation;
import org.apache.hudi.common.model.HoodieAvroRecord;
@@ -94,7 +94,7 @@ public abstract class JavaExecutionStrategy<T extends HoodieRecordPayload<T>>
Option.ofNullable(clusteringPlan.getPreserveHoodieMetadata()).orElse(false),
instantTime)));
HoodieWriteMetadata<HoodieData<WriteStatus>> writeMetadata = new HoodieWriteMetadata<>();
writeMetadata.setWriteStatuses(HoodieList.of(writeStatusList));
writeMetadata.setWriteStatuses(HoodieListData.eager(writeStatusList));
return writeMetadata;
}

View File

@@ -24,7 +24,7 @@ import org.apache.hudi.common.config.SerializableConfiguration;
import org.apache.hudi.common.data.HoodieAccumulator;
import org.apache.hudi.common.data.HoodieAtomicLongAccumulator;
import org.apache.hudi.common.data.HoodieData;
import org.apache.hudi.common.data.HoodieList;
import org.apache.hudi.common.data.HoodieListData;
import org.apache.hudi.common.engine.EngineProperty;
import org.apache.hudi.common.engine.HoodieEngineContext;
import org.apache.hudi.common.engine.TaskContextSupplier;
@@ -74,12 +74,12 @@ public class HoodieJavaEngineContext extends HoodieEngineContext {
@Override
public <T> HoodieData<T> emptyHoodieData() {
return HoodieList.of(Collections.emptyList());
return HoodieListData.eager(Collections.emptyList());
}
@Override
public <T> HoodieData<T> parallelize(List<T> data, int parallelism) {
return HoodieList.of(data);
return HoodieListData.eager(data);
}
@Override

View File

@@ -23,7 +23,7 @@ import org.apache.hudi.ApiMaturityLevel;
import org.apache.hudi.PublicAPIMethod;
import org.apache.hudi.client.WriteStatus;
import org.apache.hudi.common.data.HoodieData;
import org.apache.hudi.common.data.HoodieList;
import org.apache.hudi.common.data.HoodieListData;
import org.apache.hudi.common.engine.HoodieEngineContext;
import org.apache.hudi.common.model.HoodieRecord;
import org.apache.hudi.common.model.HoodieRecordPayload;
@@ -58,8 +58,8 @@ public abstract class JavaHoodieIndex<T extends HoodieRecordPayload> extends Hoo
public <R> HoodieData<HoodieRecord<R>> tagLocation(
HoodieData<HoodieRecord<R>> records, HoodieEngineContext context,
HoodieTable hoodieTable) throws HoodieIndexException {
List<HoodieRecord<T>> hoodieRecords = tagLocation(HoodieList.getList(records.map(record -> (HoodieRecord<T>) record)), context, hoodieTable);
return HoodieList.of(hoodieRecords.stream().map(r -> (HoodieRecord<R>) r).collect(Collectors.toList()));
List<HoodieRecord<T>> hoodieRecords = tagLocation(records.map(record -> (HoodieRecord<T>) record).collectAsList(), context, hoodieTable);
return HoodieListData.eager(hoodieRecords.stream().map(r -> (HoodieRecord<R>) r).collect(Collectors.toList()));
}
@Override
@@ -67,6 +67,6 @@ public abstract class JavaHoodieIndex<T extends HoodieRecordPayload> extends Hoo
public HoodieData<WriteStatus> updateLocation(
HoodieData<WriteStatus> writeStatuses, HoodieEngineContext context,
HoodieTable hoodieTable) throws HoodieIndexException {
return HoodieList.of(updateLocation(HoodieList.getList(writeStatuses), context, hoodieTable));
return HoodieListData.eager(updateLocation(writeStatuses.collectAsList(), context, hoodieTable));
}
}

View File

@@ -36,8 +36,6 @@ import org.apache.hudi.table.action.HoodieWriteMetadata;
import java.util.List;
import static org.apache.hudi.common.data.HoodieList.getList;
public abstract class HoodieJavaTable<T extends HoodieRecordPayload>
extends HoodieTable<T, List<HoodieRecord<T>>, List<HoodieKey>, List<WriteStatus>> {
protected HoodieJavaTable(HoodieWriteConfig config, HoodieEngineContext context, HoodieTableMetaClient metaClient) {
@@ -67,7 +65,7 @@ public abstract class HoodieJavaTable<T extends HoodieRecordPayload>
public static HoodieWriteMetadata<List<WriteStatus>> convertMetadata(
HoodieWriteMetadata<HoodieData<WriteStatus>> metadata) {
return metadata.clone(getList(metadata.getWriteStatuses()));
return metadata.clone(metadata.getWriteStatuses().collectAsList());
}
@Override

View File

@@ -19,7 +19,7 @@
package org.apache.hudi.table.action.commit;
import org.apache.hudi.client.WriteStatus;
import org.apache.hudi.common.data.HoodieList;
import org.apache.hudi.common.data.HoodieListData;
import org.apache.hudi.common.engine.HoodieEngineContext;
import org.apache.hudi.common.model.HoodieBaseFile;
import org.apache.hudi.common.model.HoodieCommitMetadata;
@@ -130,8 +130,7 @@ public abstract class BaseJavaCommitActionExecutor<T extends HoodieRecordPayload
protected List<WriteStatus> updateIndex(List<WriteStatus> writeStatuses, HoodieWriteMetadata<List<WriteStatus>> result) {
Instant indexStartTime = Instant.now();
// Update the index back
List<WriteStatus> statuses = HoodieList.getList(
table.getIndex().updateLocation(HoodieList.of(writeStatuses), context, table));
List<WriteStatus> statuses = table.getIndex().updateLocation(HoodieListData.eager(writeStatuses), context, table).collectAsList();
result.setIndexUpdateDuration(Duration.between(indexStartTime, Instant.now()));
result.setWriteStatuses(statuses);
return statuses;
@@ -339,8 +338,7 @@ public abstract class BaseJavaCommitActionExecutor<T extends HoodieRecordPayload
public void updateIndexAndCommitIfNeeded(List<WriteStatus> writeStatuses, HoodieWriteMetadata result) {
Instant indexStartTime = Instant.now();
// Update the index back
List<WriteStatus> statuses = HoodieList.getList(
table.getIndex().updateLocation(HoodieList.of(writeStatuses), context, table));
List<WriteStatus> statuses = table.getIndex().updateLocation(HoodieListData.eager(writeStatuses), context, table).collectAsList();
result.setIndexUpdateDuration(Duration.between(indexStartTime, Instant.now()));
result.setWriteStatuses(statuses);
result.setPartitionToReplaceFileIds(getPartitionToReplacedFileIds(result));

View File

@@ -19,7 +19,7 @@
package org.apache.hudi.table.action.commit;
import org.apache.hudi.client.WriteStatus;
import org.apache.hudi.common.data.HoodieList;
import org.apache.hudi.common.data.HoodieListData;
import org.apache.hudi.common.engine.HoodieEngineContext;
import org.apache.hudi.common.model.EmptyHoodieRecordPayload;
import org.apache.hudi.common.model.HoodieAvroRecord;
@@ -99,8 +99,7 @@ public class JavaDeleteHelper<R> extends
dedupedKeys.stream().map(key -> new HoodieAvroRecord<>(key, new EmptyHoodieRecordPayload())).collect(Collectors.toList());
Instant beginTag = Instant.now();
// perform index look up to get existing location of records
List<HoodieRecord<EmptyHoodieRecordPayload>> taggedRecords = HoodieList.getList(
table.getIndex().tagLocation(HoodieList.of(dedupedRecords), context, table));
List<HoodieRecord<EmptyHoodieRecordPayload>> taggedRecords = table.getIndex().tagLocation(HoodieListData.eager(dedupedRecords), context, table).collectAsList();
Duration tagLocationDuration = Duration.between(beginTag, Instant.now());
// filter out non existent keys/records

View File

@@ -19,7 +19,7 @@
package org.apache.hudi.table.action.commit;
import org.apache.hudi.client.WriteStatus;
import org.apache.hudi.common.data.HoodieList;
import org.apache.hudi.common.data.HoodieListData;
import org.apache.hudi.common.engine.HoodieEngineContext;
import org.apache.hudi.common.model.HoodieAvroRecord;
import org.apache.hudi.common.model.HoodieKey;
@@ -50,8 +50,7 @@ public class JavaWriteHelper<T extends HoodieRecordPayload,R> extends BaseWriteH
@Override
protected List<HoodieRecord<T>> tag(List<HoodieRecord<T>> dedupedRecords, HoodieEngineContext context, HoodieTable<T, List<HoodieRecord<T>>, List<HoodieKey>, List<WriteStatus>> table) {
return HoodieList.getList(
table.getIndex().tagLocation(HoodieList.of(dedupedRecords), context, table));
return table.getIndex().tagLocation(HoodieListData.eager(dedupedRecords), context, table).collectAsList();
}
@Override