danny0405 commented on code in PR #12269: URL: https://github.com/apache/hudi/pull/12269#discussion_r1849498644
########## hudi-client/hudi-spark-client/src/test/java/org/apache/hudi/client/functional/TestRLIRecordGeneration.java: ########## @@ -0,0 +1,214 @@ +/* + * 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.functional; + +import org.apache.hudi.client.SparkRDDWriteClient; +import org.apache.hudi.client.WriteStatus; +import org.apache.hudi.client.common.HoodieSparkEngineContext; +import org.apache.hudi.common.model.EmptyHoodieRecordPayload; +import org.apache.hudi.common.model.HoodieFailedWritesCleaningPolicy; +import org.apache.hudi.common.model.HoodieRecord; +import org.apache.hudi.common.model.HoodieTableType; +import org.apache.hudi.common.model.HoodieWriteStat; +import org.apache.hudi.common.util.Option; +import org.apache.hudi.config.HoodieCompactionConfig; +import org.apache.hudi.config.HoodieWriteConfig; +import org.apache.hudi.exception.HoodieException; +import org.apache.hudi.metadata.BaseFileUtils; +import org.apache.hudi.metadata.HoodieMetadataPayload; +import org.apache.hudi.testutils.HoodieClientTestBase; + +import org.junit.jupiter.params.ParameterizedTest; +import org.junit.jupiter.params.provider.EnumSource; + +import java.io.IOException; +import java.util.ArrayList; +import java.util.Collections; +import java.util.HashMap; +import java.util.Iterator; +import java.util.List; +import java.util.Map; +import java.util.stream.Collectors; + +import static org.apache.hudi.testutils.Assertions.assertNoWriteErrors; +import static org.junit.jupiter.api.Assertions.assertEquals; +import static org.junit.jupiter.api.Assertions.assertTrue; + +public class TestRLIRecordGeneration extends HoodieClientTestBase { + + @ParameterizedTest + @EnumSource(value = HoodieTableType.class, names = {"COPY_ON_WRITE", "MERGE_ON_READ"}) + public void testGeneratingRLIRecordsFromBaseFile(HoodieTableType tableType) throws IOException { + cleanupClients(); + initMetaClient(tableType); + cleanupTimelineService(); + initTimelineService(); + + HoodieSparkEngineContext engineContext = new HoodieSparkEngineContext(jsc); + HoodieWriteConfig writeConfig = tableType == HoodieTableType.COPY_ON_WRITE ? getConfigBuilder(HoodieFailedWritesCleaningPolicy.EAGER).build() + : getConfigBuilder(HoodieFailedWritesCleaningPolicy.EAGER) + .withCompactionConfig(HoodieCompactionConfig.newBuilder().withMaxNumDeltaCommitsBeforeCompaction(2) + .withInlineCompaction(true) + .compactionSmallFileSize(0).build()).build(); + + try (SparkRDDWriteClient client = new SparkRDDWriteClient(engineContext, writeConfig)) { + // Insert + String commitTime = client.createNewInstantTime(); + List<HoodieRecord> records1 = dataGen.generateInserts(commitTime, 100); + client.startCommitWithTime(commitTime); + List<WriteStatus> writeStatuses1 = client.insert(jsc.parallelize(records1, 1), commitTime).collect(); + assertNoWriteErrors(writeStatuses1); + + // assert RLI records for a base file from 1st commit + String finalCommitTime = commitTime; + Map<String, String> recordKeyToPartitionMapping1 = new HashMap<>(); + Map<String, String> fileIdToFileNameMapping1 = new HashMap<>(); + writeStatuses1.forEach(writeStatus -> { + assertEquals(writeStatus.getStat().getNumDeletes(), 0); + // Fetch record keys for all + try { + String writeStatFileId = writeStatus.getFileId(); + if (!fileIdToFileNameMapping1.containsKey(writeStatFileId)) { + fileIdToFileNameMapping1.put(writeStatFileId, writeStatus.getStat().getPath().substring(writeStatus.getStat().getPath().lastIndexOf("/") + 1)); + } + + Iterator<HoodieRecord> rliRecordsItr = BaseFileUtils.generateRLIMetadataHoodieRecordsForBaseFile(metaClient.getBasePath().toString(), + writeStatus.getStat(), writeConfig.getWritesFileIdEncoding(), finalCommitTime, metaClient.getStorage()); + while (rliRecordsItr.hasNext()) { + HoodieRecord rliRecord = rliRecordsItr.next(); + String key = rliRecord.getRecordKey(); + String partition = ((HoodieMetadataPayload) rliRecord.getData()).getRecordGlobalLocation().getPartitionPath(); + recordKeyToPartitionMapping1.put(key, partition); + } + } catch (IOException e) { + throw new HoodieException("Should not have failed ", e); + } + }); + + Map<String, String> expectedRecordToPartitionMapping1 = new HashMap<>(); + records1.forEach(record -> expectedRecordToPartitionMapping1.put(record.getRecordKey(), record.getPartitionPath())); + + assertEquals(expectedRecordToPartitionMapping1, recordKeyToPartitionMapping1); + + // lets update some records and assert RLI records. + commitTime = client.createNewInstantTime(); + client.startCommitWithTime(commitTime); + String finalCommitTime2 = commitTime; + List<HoodieRecord> deletes2 = dataGen.generateUniqueDeleteRecords(commitTime, 30); + List<HoodieRecord> updates2 = dataGen.generateUniqueUpdates(commitTime, 30); + List<HoodieRecord> inserts2 = dataGen.generateInserts(commitTime, 30); + List<HoodieRecord> records2 = new ArrayList<>(); + records2.addAll(inserts2); + records2.addAll(updates2); + records2.addAll(deletes2); + + List<WriteStatus> writeStatuses2 = client.upsert(jsc.parallelize(records2, 1), commitTime).collect(); + assertNoWriteErrors(writeStatuses2); + + if (tableType == HoodieTableType.COPY_ON_WRITE) { + List<String> expectedRLIInserts = inserts2.stream().map(record -> record.getKey().getRecordKey()).collect(Collectors.toList()); + List<String> expectedRLIDeletes = deletes2.stream().map(record -> record.getKey().getRecordKey()).collect(Collectors.toList()); + generateRliRecordsAndAssert(writeStatuses2, fileIdToFileNameMapping1, finalCommitTime2, writeConfig, expectedRLIInserts, expectedRLIDeletes); + } else { + // trigger 2nd commit followed by compaction. + commitTime = client.createNewInstantTime(); + client.startCommitWithTime(commitTime); + String finalCommitTime3 = commitTime; + List<HoodieRecord> deletes3 = dataGen.generateUniqueDeleteRecords(commitTime, 30); + List<HoodieRecord> updates3 = dataGen.generateUniqueUpdates(commitTime, 30); + List<HoodieRecord> inserts3 = dataGen.generateInserts(commitTime, 30); + List<HoodieRecord> records3 = new ArrayList<>(); + records3.addAll(inserts3); + records3.addAll(updates3); + records3.addAll(deletes3); + + List<WriteStatus> writeStatuses3 = client.upsert(jsc.parallelize(records3, 1), commitTime).collect(); + assertNoWriteErrors(writeStatuses3); + + // trigger compaction + Option<String> compactionInstantOpt = client.scheduleCompaction(Option.empty()); Review Comment: Can we test the base and log file read too, without compaction. -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
