nsivabalan commented on code in PR #14180: URL: https://github.com/apache/hudi/pull/14180#discussion_r2487784802
########## hudi-spark-datasource/hudi-spark/src/test/scala/org/apache/hudi/functional/TestGlobalRecordLevelIndex.scala: ########## @@ -0,0 +1,775 @@ +/* + * 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.functional + +import org.apache.hudi.DataSourceWriteOptions +import org.apache.hudi.DataSourceWriteOptions._ +import org.apache.hudi.client.SparkRDDWriteClient +import org.apache.hudi.client.common.HoodieSparkEngineContext +import org.apache.hudi.client.transaction.PreferWriterConflictResolutionStrategy +import org.apache.hudi.common.config.HoodieMetadataConfig +import org.apache.hudi.common.model._ +import org.apache.hudi.common.table.{HoodieTableConfig, HoodieTableMetaClient} +import org.apache.hudi.common.table.timeline.{HoodieInstant, HoodieTimeline} +import org.apache.hudi.common.testutils.{HoodieTestDataGenerator, InProcessTimeGenerator} +import org.apache.hudi.common.testutils.HoodieTestDataGenerator.recordsToStrings +import org.apache.hudi.config._ +import org.apache.hudi.exception.HoodieWriteConflictException +import org.apache.hudi.index.HoodieIndex +import org.apache.hudi.index.record.HoodieRecordIndex +import org.apache.hudi.metadata.{HoodieBackedTableMetadata, HoodieTableMetadataUtil, MetadataPartitionType} +import org.apache.hudi.util.JavaConversions + +import org.apache.spark.sql._ +import org.apache.spark.sql.catalyst.expressions.GenericRowWithSchema +import org.apache.spark.sql.functions.lit +import org.apache.spark.sql.types.{StringType, StructField, StructType} +import org.junit.jupiter.api._ +import org.junit.jupiter.api.Assertions.{assertEquals, assertFalse, assertTrue} +import org.junit.jupiter.params.ParameterizedTest +import org.junit.jupiter.params.provider.{Arguments, CsvSource, EnumSource, MethodSource} +import org.junit.jupiter.params.provider.Arguments.arguments + +import java.util +import java.util.{Collections, Properties} +import java.util.concurrent.Executors + +import scala.collection.JavaConverters._ +import scala.concurrent.{Await, ExecutionContext, Future} +import scala.concurrent.duration.Duration +import scala.language.postfixOps +import scala.util.Using + +@Tag("functional-b") +class TestGlobalRecordLevelIndex extends RecordLevelIndexTestBase { + + @ParameterizedTest + @EnumSource(classOf[HoodieTableType]) + def testRLIInitialization(tableType: HoodieTableType): Unit = { + val hudiOpts = commonOpts + (DataSourceWriteOptions.TABLE_TYPE.key -> tableType.name()) + doWriteAndValidateDataAndRecordIndex(hudiOpts, + operation = DataSourceWriteOptions.INSERT_OPERATION_OPT_VAL, + saveMode = SaveMode.Overwrite) + metaClient = HoodieTableMetaClient.reload(metaClient) + assertFalse(HoodieRecordIndex.isPartitioned(metaClient.getIndexMetadata.get().getIndexDefinitions.get(HoodieTableMetadataUtil.PARTITION_NAME_RECORD_INDEX))) + } + + @Test + def testRLIInitializationForMorGlobalIndex(): Unit = { + val tableType = HoodieTableType.MERGE_ON_READ + val hudiOpts = commonOpts + (DataSourceWriteOptions.TABLE_TYPE.key -> tableType.name()) + + (HoodieMetadataConfig.RECORD_INDEX_MIN_FILE_GROUP_COUNT_PROP.key -> "1") + + (HoodieMetadataConfig.RECORD_INDEX_MAX_FILE_GROUP_COUNT_PROP.key -> "1") + + (HoodieIndexConfig.INDEX_TYPE.key -> HoodieIndex.IndexType.GLOBAL_RECORD_LEVEL_INDEX.name()) + + (HoodieIndexConfig.RECORD_INDEX_UPDATE_PARTITION_PATH_ENABLE.key -> "true") - + HoodieMetadataConfig.GLOBAL_RECORD_LEVEL_INDEX_ENABLE_PROP.key + + val dataGen1 = HoodieTestDataGenerator.createTestGeneratorFirstPartition() + val dataGen2 = HoodieTestDataGenerator.createTestGeneratorSecondPartition() + + // batch1 inserts (5 records) + val instantTime1 = getNewInstantTime() + val latestBatch = recordsToStrings(dataGen1.generateInserts(instantTime1, 5)).asScala.toSeq + var operation = INSERT_OPERATION_OPT_VAL + val latestBatchDf = spark.read.json(spark.sparkContext.parallelize(latestBatch, 1)) + latestBatchDf.cache() + latestBatchDf.write.format("hudi") + .options(hudiOpts) + .mode(SaveMode.Overwrite) + .save(basePath) + val deletedDf1 = calculateMergedDf(latestBatchDf, operation, true) + deletedDf1.cache() + + // batch2. upsert. update few records to 2nd partition from partition1 and insert a few to partition2. + val instantTime2 = getNewInstantTime() + + val latestBatch2_1 = recordsToStrings(dataGen1.generateUniqueUpdates(instantTime2, 3)).asScala.toSeq + val latestBatchDf2_1 = spark.read.json(spark.sparkContext.parallelize(latestBatch2_1, 1)) + val latestBatchDf2_2 = latestBatchDf2_1.withColumn("partition", lit(HoodieTestDataGenerator.DEFAULT_SECOND_PARTITION_PATH)) + .withColumn("partition_path", lit(HoodieTestDataGenerator.DEFAULT_SECOND_PARTITION_PATH)) + val latestBatch2_3 = recordsToStrings(dataGen2.generateInserts(instantTime2, 2)).asScala.toSeq + val latestBatchDf2_3 = spark.read.json(spark.sparkContext.parallelize(latestBatch2_3, 1)) + val latestBatchDf2Final = latestBatchDf2_3.union(latestBatchDf2_2) + latestBatchDf2Final.cache() + latestBatchDf2Final.write.format("hudi") + .options(hudiOpts) + .mode(SaveMode.Append) + .save(basePath) + operation = UPSERT_OPERATION_OPT_VAL + val deletedDf2 = calculateMergedDf(latestBatchDf2Final, operation, true) + deletedDf2.cache() + + val hudiOpts2 = commonOpts + (DataSourceWriteOptions.TABLE_TYPE.key -> tableType.name()) + + (HoodieMetadataConfig.RECORD_INDEX_MIN_FILE_GROUP_COUNT_PROP.key -> "1") + + (HoodieMetadataConfig.RECORD_INDEX_MAX_FILE_GROUP_COUNT_PROP.key -> "1") + + (HoodieIndexConfig.INDEX_TYPE.key -> HoodieIndex.IndexType.GLOBAL_RECORD_LEVEL_INDEX.name()) + + (HoodieIndexConfig.RECORD_INDEX_UPDATE_PARTITION_PATH_ENABLE.key -> "true") + + (HoodieMetadataConfig.GLOBAL_RECORD_LEVEL_INDEX_ENABLE_PROP.key -> "true") + + val instantTime3 = getNewInstantTime() + // batch3. update 2 records from newly inserted records from commit 2 to partition2 + val latestBatch3 = recordsToStrings(dataGen2.generateUniqueUpdates(instantTime3, 2)).asScala.toSeq + val latestBatchDf3 = spark.read.json(spark.sparkContext.parallelize(latestBatch3, 1)) + latestBatchDf3.cache() + latestBatchDf3.write.format("hudi") + .options(hudiOpts2) + .mode(SaveMode.Append) + .save(basePath) + val deletedDf3 = calculateMergedDf(latestBatchDf3, operation, true) + deletedDf3.cache() + metaClient = HoodieTableMetaClient.builder().setBasePath(basePath).setConf(storageConf).build() + validateDataAndRecordIndices(hudiOpts, deletedDf3) + deletedDf2.unpersist() + deletedDf3.unpersist() + } + + private def getNewInstantTime(): String = { + InProcessTimeGenerator.createNewInstantTime(); + } + + @ParameterizedTest + @EnumSource(classOf[HoodieTableType]) + def testRLIUpsertTableTypes(tableType: HoodieTableType): Unit = { + testRLIUpsert(tableType, "GLOBAL_RECORD_LEVEL_INDEX") + } + + @ParameterizedTest + @CsvSource(value = Array( + "RECORD_INDEX", + "GLOBAL_RECORD_LEVEL_INDEX") + ) + // tests b/w compatibility of configs for global RLI Review Comment: tests added for b/w compatibility -- 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]
