lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-3042] Abstract Spark update Strategy to make code more clean and remove duplicates (#4845)

Browse Source 
Co-authored-by: Hui An <hui.an@shopee.com>
This commit is contained in:
RexAn
2022-02-21 22:53:09 +08:00
committed by GitHub
parent bf16bc122a
commit 801fdab55c
3 changed files with 59 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

53
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/client/clustering/update/strategy/BaseSparkUpdateStrategy.java Normal file
View File

@@ -0,0 +1,53 @@
/*
* 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.
*/
package org.apache.hudi.client.clustering.update.strategy;
import org.apache.hudi.client.common.HoodieSparkEngineContext;
import org.apache.hudi.common.model.HoodieFileGroupId;
import org.apache.hudi.common.model.HoodieRecord;
import org.apache.hudi.common.model.HoodieRecordPayload;
import org.apache.hudi.table.action.cluster.strategy.UpdateStrategy;
import org.apache.spark.api.java.JavaRDD;
import java.util.List;
import java.util.Set;
/**
* Spark base update strategy, write records to the file groups which are in clustering
* need to check. Spark relate implementations should extend this base class.
*/
public abstract class BaseSparkUpdateStrategy<T extends HoodieRecordPayload<T>> extends UpdateStrategy<T, JavaRDD<HoodieRecord<T>>> {
public BaseSparkUpdateStrategy(HoodieSparkEngineContext engineContext,
Set<HoodieFileGroupId> fileGroupsInPendingClustering) {
super(engineContext, fileGroupsInPendingClustering);
}
/**
* Get records matched file group ids.
* @param inputRecords the records to write, tagged with target file id
* @return the records matched file group ids
*/
protected List<HoodieFileGroupId> getGroupIdsWithUpdate(JavaRDD<HoodieRecord<T>> inputRecords) {
return inputRecords
.filter(record -> record.getCurrentLocation() != null)
.map(record -> new HoodieFileGroupId(record.getPartitionPath(), record.getCurrentLocation().getFileId())).distinct().collect();
}
}

14
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/client/clustering/update/strategy/SparkAllowUpdateStrategy.java
View File

@@ -23,7 +23,6 @@ import org.apache.hudi.common.model.HoodieFileGroupId;
import org.apache.hudi.common.model.HoodieRecord;
import org.apache.hudi.common.model.HoodieRecordPayload;
import org.apache.hudi.common.util.collection.Pair;
import org.apache.hudi.table.action.cluster.strategy.UpdateStrategy;
import org.apache.spark.api.java.JavaRDD;
@@ -35,20 +34,13 @@ import java.util.stream.Collectors;
/**
* Allow ingestion commits during clustering job.
*/
public class SparkAllowUpdateStrategy<T extends HoodieRecordPayload<T>> extends UpdateStrategy<T, JavaRDD<HoodieRecord<T>>> {
public class SparkAllowUpdateStrategy<T extends HoodieRecordPayload<T>> extends BaseSparkUpdateStrategy<T> {
public SparkAllowUpdateStrategy(
HoodieSparkEngineContext engineContext, HashSet<HoodieFileGroupId> fileGroupsInPendingClustering) {
public SparkAllowUpdateStrategy(HoodieSparkEngineContext engineContext,
HashSet<HoodieFileGroupId> fileGroupsInPendingClustering) {
super(engineContext, fileGroupsInPendingClustering);
}
private List<HoodieFileGroupId> getGroupIdsWithUpdate(JavaRDD<HoodieRecord<T>> inputRecords) {
List<HoodieFileGroupId> fileGroupIdsWithUpdates = inputRecords
.filter(record -> record.getCurrentLocation() != null)
.map(record -> new HoodieFileGroupId(record.getPartitionPath(), record.getCurrentLocation().getFileId())).distinct().collect();
return fileGroupIdsWithUpdates;
}
@Override
public Pair<JavaRDD<HoodieRecord<T>>, Set<HoodieFileGroupId>> handleUpdate(JavaRDD<HoodieRecord<T>> taggedRecordsRDD) {
List<HoodieFileGroupId> fileGroupIdsWithRecordUpdate = getGroupIdsWithUpdate(taggedRecordsRDD);

13
hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/client/clustering/update/strategy/SparkRejectUpdateStrategy.java
View File

@@ -24,7 +24,6 @@ import org.apache.hudi.common.model.HoodieRecord;
import org.apache.hudi.common.model.HoodieRecordPayload;
import org.apache.hudi.common.util.collection.Pair;
import org.apache.hudi.exception.HoodieClusteringUpdateException;
import org.apache.hudi.table.action.cluster.strategy.UpdateStrategy;
import org.apache.log4j.LogManager;
import org.apache.log4j.Logger;
@@ -39,20 +38,14 @@ import java.util.Set;
* Update strategy based on following.
* if some file groups have update record, throw exception
*/
public class SparkRejectUpdateStrategy<T extends HoodieRecordPayload<T>> extends UpdateStrategy<T, JavaRDD<HoodieRecord<T>>> {
public class SparkRejectUpdateStrategy<T extends HoodieRecordPayload<T>> extends BaseSparkUpdateStrategy<T> {
private static final Logger LOG = LogManager.getLogger(SparkRejectUpdateStrategy.class);
public SparkRejectUpdateStrategy(HoodieSparkEngineContext engineContext, HashSet<HoodieFileGroupId> fileGroupsInPendingClustering) {
public SparkRejectUpdateStrategy(HoodieSparkEngineContext engineContext,
HashSet<HoodieFileGroupId> fileGroupsInPendingClustering) {
super(engineContext, fileGroupsInPendingClustering);
}
private List<HoodieFileGroupId> getGroupIdsWithUpdate(JavaRDD<HoodieRecord<T>> inputRecords) {
List<HoodieFileGroupId> fileGroupIdsWithUpdates = inputRecords
.filter(record -> record.getCurrentLocation() != null)
.map(record -> new HoodieFileGroupId(record.getPartitionPath(), record.getCurrentLocation().getFileId())).distinct().collect();
return fileGroupIdsWithUpdates;
}
@Override
public Pair<JavaRDD<HoodieRecord<T>>, Set<HoodieFileGroupId>> handleUpdate(JavaRDD<HoodieRecord<T>> taggedRecordsRDD) {
List<HoodieFileGroupId> fileGroupIdsWithRecordUpdate = getGroupIdsWithUpdate(taggedRecordsRDD);