[GitHub] [hudi] yihua commented on a change in pull request #3857: [HUDI-2332] Add clustering and compaction in Kafka Connect Sink

[GitBox <gi...@apache.org>]

yihua commented on a change in pull request #3857:
URL: https://github.com/apache/hudi/pull/3857#discussion_r749716962



##########
File path: hudi-client/hudi-java-client/src/main/java/org/apache/hudi/client/clustering/run/strategy/JavaExecutionStrategy.java
##########
@@ -0,0 +1,245 @@
+/*
+ * 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.run.strategy;
+
+import org.apache.hudi.avro.HoodieAvroUtils;
+import org.apache.hudi.avro.model.HoodieClusteringGroup;
+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.engine.HoodieEngineContext;
+import org.apache.hudi.common.model.ClusteringOperation;
+import org.apache.hudi.common.model.HoodieFileGroupId;
+import org.apache.hudi.common.model.HoodieKey;
+import org.apache.hudi.common.model.HoodieRecord;
+import org.apache.hudi.common.model.HoodieRecordPayload;
+import org.apache.hudi.common.model.RewriteAvroPayload;
+import org.apache.hudi.common.table.HoodieTableConfig;
+import org.apache.hudi.common.table.log.HoodieMergedLogRecordScanner;
+import org.apache.hudi.common.util.Option;
+import org.apache.hudi.common.util.StringUtils;
+import org.apache.hudi.common.util.collection.Pair;
+import org.apache.hudi.config.HoodieWriteConfig;
+import org.apache.hudi.exception.HoodieClusteringException;
+import org.apache.hudi.io.IOUtils;
+import org.apache.hudi.io.storage.HoodieFileReader;
+import org.apache.hudi.io.storage.HoodieFileReaderFactory;
+import org.apache.hudi.keygen.BaseKeyGenerator;
+import org.apache.hudi.keygen.KeyGenUtils;
+import org.apache.hudi.table.BulkInsertPartitioner;
+import org.apache.hudi.table.HoodieTable;
+import org.apache.hudi.table.action.HoodieWriteMetadata;
+import org.apache.hudi.table.action.cluster.strategy.ClusteringExecutionStrategy;
+
+import org.apache.avro.Schema;
+import org.apache.avro.generic.GenericRecord;
+import org.apache.avro.generic.IndexedRecord;
+import org.apache.hadoop.fs.Path;
+import org.apache.log4j.LogManager;
+import org.apache.log4j.Logger;
+
+import java.io.IOException;
+import java.io.UncheckedIOException;
+import java.util.ArrayList;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+import java.util.stream.Collectors;
+import java.util.stream.StreamSupport;
+
+import static org.apache.hudi.common.table.log.HoodieFileSliceReader.getFileSliceReader;
+
+/**
+ * Clustering strategy for Java engine.
+ */
+public abstract class JavaExecutionStrategy<T extends HoodieRecordPayload<T>>
+    extends ClusteringExecutionStrategy<T, List<HoodieRecord<T>>, List<HoodieKey>, List<WriteStatus>> {
+
+  private static final Logger LOG = LogManager.getLogger(JavaExecutionStrategy.class);
+
+  public JavaExecutionStrategy(
+      HoodieTable table, HoodieEngineContext engineContext, HoodieWriteConfig writeConfig) {
+    super(table, engineContext, writeConfig);
+  }
+
+  @Override
+  public HoodieWriteMetadata<List<WriteStatus>> performClustering(
+      HoodieClusteringPlan clusteringPlan, Schema schema, String instantTime) {
+    // execute clustering for each group and collect WriteStatus
+    List<WriteStatus> writeStatusList = new ArrayList<>();
+    clusteringPlan.getInputGroups().forEach(
+        inputGroup -> writeStatusList.addAll(runClusteringForGroup(
+            inputGroup, clusteringPlan.getStrategy().getStrategyParams(),
+            Option.ofNullable(clusteringPlan.getPreserveHoodieMetadata()).orElse(false),
+            instantTime)));
+    HoodieWriteMetadata<List<WriteStatus>> writeMetadata = new HoodieWriteMetadata<>();
+    writeMetadata.setWriteStatuses(writeStatusList);
+    return writeMetadata;
+  }
+
+  /**
+   * Execute clustering to write inputRecords into new files as defined by rules in strategy parameters.
+   * The number of new file groups created is bounded by numOutputGroups.
+   * Note that commit is not done as part of strategy. commit is callers responsibility.
+   *
+   * @param inputRecords           List of {@link HoodieRecord}.
+   * @param numOutputGroups        Number of output file groups.
+   * @param instantTime            Clustering (replace commit) instant time.
+   * @param strategyParams         Strategy parameters containing columns to sort the data by when clustering.
+   * @param schema                 Schema of the data including metadata fields.
+   * @param fileGroupIdList        File group id corresponding to each out group.
+   * @param preserveHoodieMetadata Whether to preserve commit metadata while clustering.
+   * @return List of {@link WriteStatus}.
+   */
+  public abstract List<WriteStatus> performClusteringWithRecordList(
+      final List<HoodieRecord<T>> inputRecords, final int numOutputGroups, final String instantTime,
+      final Map<String, String> strategyParams, final Schema schema,
+      final List<HoodieFileGroupId> fileGroupIdList, final boolean preserveHoodieMetadata);
+
+  /**
+   * Create {@link BulkInsertPartitioner} based on strategy params.
+   *
+   * @param strategyParams Strategy parameters containing columns to sort the data by when clustering.
+   * @param schema         Schema of the data including metadata fields.
+   * @return empty for now.
+   */
+  protected Option<BulkInsertPartitioner<T>> getPartitioner(Map<String, String> strategyParams, Schema schema) {
+    return Option.empty();

Review comment:
       @codope @vinothchandar I see clustering in Spark follows the `LAYOUT_OPTIMIZE_ENABLE` and `PLAN_STRATEGY_SORT_COLUMNS` configs to choose Spark-specific partitioner, and the default is empty.  Does Java client need to respect these configs?




-- 
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.

To unsubscribe, e-mail: commits-unsubscribe@hudi.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org