satishkotha commented on a change in pull request #2275: URL: https://github.com/apache/hudi/pull/2275#discussion_r547482024
########## File path: hudi-client/hudi-client-common/src/main/java/org/apache/hudi/table/action/clustering/update/UpdateStrategy.java ########## @@ -0,0 +1,41 @@ +/* + * 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.table.action.clustering.update; + +import org.apache.hudi.common.model.HoodieFileGroupId; +import org.apache.hudi.common.table.timeline.HoodieInstant; +import org.apache.hudi.common.util.collection.Pair; +import org.apache.hudi.table.WorkloadProfile; + +import java.util.List; + +/** + * When file groups in clustering, write records to these file group need to check. + */ +public interface UpdateStrategy { + + /** + * check the update records to the file group in clustering. + * @param fileGroupsInPendingClustering + * @param workloadProfile workloadProfile have the records update info, + * just like BaseSparkCommitActionExecutor.getUpsertPartitioner use it. + */ + void apply(List<Pair<HoodieFileGroupId, HoodieInstant>> fileGroupsInPendingClustering, WorkloadProfile workloadProfile); Review comment: I may have mentioned this earlier, consider removing WorkloadProfile and send fileGroupsWithUpdates. I think sending taggedRecords as additional parameter will also be useful. In future, we may want to update tagged records location to a different fileId. ########## File path: hudi-client/hudi-client-common/src/main/java/org/apache/hudi/table/action/clustering/update/RejectUpdateStrategy.java ########## @@ -0,0 +1,61 @@ +/* + * 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.table.action.clustering.update; + +import org.apache.hudi.common.model.HoodieFileGroupId; +import org.apache.hudi.common.table.timeline.HoodieInstant; +import org.apache.hudi.common.util.collection.Pair; +import org.apache.hudi.exception.HoodieClusteringUpdateException; +import org.apache.hudi.table.WorkloadProfile; +import org.apache.hudi.table.WorkloadStat; +import org.apache.log4j.LogManager; +import org.apache.log4j.Logger; + +import java.util.List; +import java.util.Map; +import java.util.Set; +import java.util.stream.Collectors; + +public class RejectUpdateStrategy implements UpdateStrategy { + private static final Logger LOG = LogManager.getLogger(RejectUpdateStrategy.class); + + @Override + public void apply(List<Pair<HoodieFileGroupId, HoodieInstant>> fileGroupsInPendingClustering, WorkloadProfile workloadProfile) { + Set<Pair<String, String>> partitionPathAndFileIds = fileGroupsInPendingClustering.stream() Review comment: Why not collect Set of HoodieFileGroupId? I'd avoid Pair as much as possible because its not that easy to read. ########## File path: hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/table/action/commit/UpsertPartitioner.java ########## @@ -140,11 +140,21 @@ private void assignInserts(WorkloadProfile profile, HoodieEngineContext context) Map<String, List<SmallFile>> partitionSmallFilesMap = getSmallFilesForPartitions(new ArrayList<String>(partitionPaths), context); + // get the in pending clustering fileId for each partition path Review comment: Maybe this can also be moved into UpdateStrategy? i.e., UpdateStrategy#filterSmallFiles(List<SmallFile> smallFiles) removes small files in pending clustering. ########## File path: hudi-client/hudi-spark-client/src/main/java/org/apache/hudi/table/action/commit/UpsertPartitioner.java ########## @@ -140,11 +140,21 @@ private void assignInserts(WorkloadProfile profile, HoodieEngineContext context) Map<String, List<SmallFile>> partitionSmallFilesMap = getSmallFilesForPartitions(new ArrayList<String>(partitionPaths), context); + // get the in pending clustering fileId for each partition path + Map<String, List<String>> partitionPathToInPendingClusteringFileId = Review comment: looks like Map<String, Set<String>> makes more sense because we only do contains check? ########## File path: hudi-client/hudi-client-common/src/main/java/org/apache/hudi/config/HoodieClusteringConfig.java ########## @@ -0,0 +1,70 @@ +/* + * 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.config; + +import org.apache.hudi.common.config.DefaultHoodieConfig; +import org.apache.hudi.table.action.clustering.update.RejectUpdateStrategy; +import org.apache.hudi.table.action.clustering.update.UpdateStrategy; + +import java.io.File; +import java.io.FileReader; +import java.io.IOException; +import java.util.Properties; + +public class HoodieClusteringConfig extends DefaultHoodieConfig { Review comment: can you please add javadoc for class and properties in class. Looks like we will need to resolve conflicts with clustering PR ########## File path: hudi-client/hudi-client-common/src/main/java/org/apache/hudi/table/action/clustering/update/RejectUpdateStrategy.java ########## @@ -0,0 +1,61 @@ +/* + * 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.table.action.clustering.update; + +import org.apache.hudi.common.model.HoodieFileGroupId; +import org.apache.hudi.common.table.timeline.HoodieInstant; +import org.apache.hudi.common.util.collection.Pair; +import org.apache.hudi.exception.HoodieClusteringUpdateException; +import org.apache.hudi.table.WorkloadProfile; +import org.apache.hudi.table.WorkloadStat; +import org.apache.log4j.LogManager; +import org.apache.log4j.Logger; + +import java.util.List; +import java.util.Map; +import java.util.Set; +import java.util.stream.Collectors; + +public class RejectUpdateStrategy implements UpdateStrategy { + private static final Logger LOG = LogManager.getLogger(RejectUpdateStrategy.class); + Review comment: Should we add a constructor to take engineContext (similar to clustering strategy). For some of the future strategies, we need engineContext. So it'd be helpful if we start with that. ---------------------------------------------------------------- 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: [email protected]
