gargvishesh commented on code in PR #16291: URL: https://github.com/apache/druid/pull/16291#discussion_r1620343918
########## extensions-core/multi-stage-query/src/main/java/org/apache/druid/msq/indexing/MSQCompactionRunner.java: ########## @@ -0,0 +1,508 @@ +/* + * 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.druid.msq.indexing; + +import com.fasterxml.jackson.annotation.JacksonInject; +import com.fasterxml.jackson.annotation.JsonIgnore; +import com.fasterxml.jackson.core.JsonProcessingException; +import com.fasterxml.jackson.databind.ObjectMapper; +import com.google.common.collect.ImmutableList; +import com.google.inject.Injector; +import org.apache.druid.data.input.impl.DimensionSchema; +import org.apache.druid.error.InvalidInput; +import org.apache.druid.indexer.TaskStatus; +import org.apache.druid.indexer.partitions.DimensionRangePartitionsSpec; +import org.apache.druid.indexer.partitions.DynamicPartitionsSpec; +import org.apache.druid.indexer.partitions.PartitionsSpec; +import org.apache.druid.indexer.partitions.SecondaryPartitionType; +import org.apache.druid.indexing.common.TaskToolbox; +import org.apache.druid.indexing.common.task.CompactionRunner; +import org.apache.druid.indexing.common.task.CompactionTask; +import org.apache.druid.indexing.common.task.CurrentSubTaskHolder; +import org.apache.druid.java.util.common.NonnullPair; +import org.apache.druid.java.util.common.StringUtils; +import org.apache.druid.java.util.common.granularity.AllGranularity; +import org.apache.druid.java.util.common.granularity.Granularities; +import org.apache.druid.java.util.common.granularity.Granularity; +import org.apache.druid.java.util.common.granularity.PeriodGranularity; +import org.apache.druid.java.util.common.logger.Logger; +import org.apache.druid.math.expr.ExprMacroTable; +import org.apache.druid.msq.indexing.destination.DataSourceMSQDestination; +import org.apache.druid.msq.kernel.WorkerAssignmentStrategy; +import org.apache.druid.msq.util.MultiStageQueryContext; +import org.apache.druid.query.Druids; +import org.apache.druid.query.Query; +import org.apache.druid.query.QueryContext; +import org.apache.druid.query.TableDataSource; +import org.apache.druid.query.dimension.DefaultDimensionSpec; +import org.apache.druid.query.dimension.DimensionSpec; +import org.apache.druid.query.expression.TimestampFloorExprMacro; +import org.apache.druid.query.filter.DimFilter; +import org.apache.druid.query.groupby.GroupByQuery; +import org.apache.druid.query.groupby.orderby.OrderByColumnSpec; +import org.apache.druid.query.spec.MultipleIntervalSegmentSpec; +import org.apache.druid.segment.VirtualColumn; +import org.apache.druid.segment.VirtualColumns; +import org.apache.druid.segment.column.ColumnHolder; +import org.apache.druid.segment.column.ColumnType; +import org.apache.druid.segment.column.RowSignature; +import org.apache.druid.segment.indexing.DataSchema; +import org.apache.druid.segment.virtual.ExpressionVirtualColumn; +import org.apache.druid.sql.calcite.parser.DruidSqlInsert; +import org.apache.druid.sql.calcite.planner.ColumnMapping; +import org.apache.druid.sql.calcite.planner.ColumnMappings; +import org.joda.time.Interval; + +import java.util.ArrayList; +import java.util.Arrays; +import java.util.Collections; +import java.util.HashMap; +import java.util.List; +import java.util.Map; +import java.util.Objects; +import java.util.stream.Collectors; + +public class MSQCompactionRunner implements CompactionRunner +{ + private static final Logger log = new Logger(MSQCompactionRunner.class); + private static final Granularity DEFAULT_SEGMENT_GRANULARITY = Granularities.ALL; + + private final ObjectMapper jsonMapper; + private final Injector injector; + + public static final String TIME_VIRTUAL_COLUMN = "__vTime"; + public static final String TIME_COLUMN = ColumnHolder.TIME_COLUMN_NAME; + + @JsonIgnore + private final CurrentSubTaskHolder currentSubTaskHolder = new CurrentSubTaskHolder( + (taskObject, config) -> { + final MSQControllerTask msqControllerTask = (MSQControllerTask) taskObject; + msqControllerTask.stopGracefully(config); + }); + + + public MSQCompactionRunner(@JacksonInject ObjectMapper jsonMapper, @JacksonInject Injector injector) + { + this.jsonMapper = jsonMapper; + this.injector = injector; + } + + @Override + public NonnullPair<Boolean, String> supportsCompactionSpec( + CompactionTask compactionTask + ) + { + if (compactionTask.getTuningConfig() != null) { + PartitionsSpec partitionsSpec = compactionTask.getTuningConfig().getPartitionsSpec(); + if (!(partitionsSpec instanceof DynamicPartitionsSpec + || partitionsSpec instanceof DimensionRangePartitionsSpec)) { + return new NonnullPair<>(false, StringUtils.format( + "Invalid partition spec type[%s] for MSQ compaction engine." + + " Type must be either DynamicPartitionsSpec or DynamicRangePartitionsSpec.", + partitionsSpec.getClass() + )); + } + if (partitionsSpec instanceof DynamicPartitionsSpec + && ((DynamicPartitionsSpec) partitionsSpec).getMaxTotalRows() != null) { + return new NonnullPair<>(false, StringUtils.format( + "maxTotalRows[%d] in DynamicPartitionsSpec not supported for MSQ compaction engine.", + ((DynamicPartitionsSpec) partitionsSpec).getMaxTotalRows() + )); + } + } + if (compactionTask.getMetricsSpec() != null + && compactionTask.getGranularitySpec() != null + && !compactionTask.getGranularitySpec() + .isRollup()) { + return new NonnullPair<>( + false, + "rollup in granularitySpec must be set to True if metricsSpec is specifed " + + "for MSQ compaction engine." + ); Review Comment: When running with native engine, `rollup=false` doesn't do any aggregation, so the number of rows are as is. It is only when `rollup=true` that the rows are grouped, but in a way equivalent to `finalizeAggregations=false`, leading to metrics being represented with their intermediate types. I believe there is no way to specify `finalizeAggregations=true` equivalent there which is to group but with the final field type for metrics. With MSQ, we don't support `rollup=false` if metricsSpec is specified since the metricsSpec is then meaningless. With `rollup=true`, we create a group-by query which can have `finalizeAggregations` as either `true` or `false`. Right now only `true` is supported. The support for `false` will come in a follow-up PR. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
