[HUDI-1491] Support partition pruning for MOR snapshot query #2378
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -18,24 +18,29 @@ | |
|
|
||
| package org.apache.hudi | ||
|
|
||
| import com.google.common.annotations.VisibleForTesting | ||
| import org.apache.hudi.common.model.HoodieBaseFile | ||
| import org.apache.hudi.common.table.{HoodieTableMetaClient, TableSchemaResolver} | ||
| import org.apache.hudi.common.table.view.HoodieTableFileSystemView | ||
| import org.apache.hudi.exception.HoodieException | ||
| import org.apache.hudi.hadoop.utils.HoodieRealtimeInputFormatUtils | ||
| import org.apache.hudi.hadoop.utils.HoodieRealtimeRecordReaderUtils.getMaxCompactionMemoryInBytes | ||
| import org.apache.hadoop.fs.{FileSystem, Path} | ||
| import org.apache.hadoop.mapred.JobConf | ||
| import org.apache.spark.internal.Logging | ||
| import org.apache.spark.rdd.RDD | ||
| import org.apache.spark.sql.catalyst.InternalRow | ||
| import org.apache.spark.sql.catalyst.expressions.{Attribute, Expression} | ||
| import org.apache.spark.sql.execution.datasources.PartitionedFile | ||
| import org.apache.spark.sql.execution.datasources.parquet.ParquetFileFormat | ||
| import org.apache.spark.sql.execution.hudi.utils.PushDownUtils | ||
| import org.apache.spark.sql.{Row, SQLContext, SparkSession} | ||
| import org.apache.spark.sql.sources.{BaseRelation, CatalystScan, Filter, PrunedFilteredScan} | ||
| import org.apache.spark.sql.types.{StructField, StructType} | ||
|
|
||
| import java.util.Locale | ||
| import scala.collection.JavaConverters._ | ||
| import scala.collection.mutable | ||
|
|
||
| case class HoodieMergeOnReadFileSplit(dataFile: Option[PartitionedFile], | ||
| logPaths: Option[List[String]], | ||
|
|
@@ -55,8 +60,8 @@ class MergeOnReadSnapshotRelation(val sqlContext: SQLContext, | |
| val optParams: Map[String, String], | ||
| val userSchema: StructType, | ||
| val globPaths: Seq[Path], | ||
| val metaClient: HoodieTableMetaClient)(val sparkSession: SparkSession) | ||
| extends BaseRelation with CatalystScan with Logging { | ||
|
Contributor
There was a problem hiding this comment. @garyli1019 : what's the counter part of this file for COW? I understand this PR is one for MOR. but just for my understanding.
Member
There was a problem hiding this comment. we don't have an independent class counterpart for COW since we just need to pass a filter to a normal parquet reader. |
||
|
|
||
| private val conf = sqlContext.sparkContext.hadoopConfiguration | ||
| private val jobConf = new JobConf(conf) | ||
|
|
@@ -68,7 +73,8 @@ class MergeOnReadSnapshotRelation(val sqlContext: SQLContext, | |
| DataSourceReadOptions.REALTIME_MERGE_OPT_KEY, | ||
| DataSourceReadOptions.DEFAULT_REALTIME_MERGE_OPT_VAL) | ||
| private val maxCompactionMemoryInBytes = getMaxCompactionMemoryInBytes(jobConf) | ||
| private val inMemoryFileIndex = HoodieSparkUtils.createInMemoryFileIndex(sqlContext.sparkSession, Some(tableStructSchema), optParams, globPaths) | ||
| private var fileIndex: List[HoodieMergeOnReadFileSplit] = _ | ||
| private val preCombineField = { | ||
| val preCombineFieldFromTableConfig = metaClient.getTableConfig.getPreCombineField | ||
| if (preCombineFieldFromTableConfig != null) { | ||
|
|
@@ -79,16 +85,33 @@ class MergeOnReadSnapshotRelation(val sqlContext: SQLContext, | |
| optParams.get(DataSourceReadOptions.READ_PRE_COMBINE_FIELD) | ||
| } | ||
| } | ||
|
|
||
| val partitionStructSchema = inMemoryFileIndex.partitionSpec().partitionColumns | ||
| val overlappedPartCols = mutable.Map.empty[String, StructField] | ||
| partitionStructSchema.foreach { partitionField => | ||
| if (tableStructSchema.exists(getColName(_) == getColName(partitionField))) { | ||
| overlappedPartCols += getColName(partitionField) -> partitionField | ||
| } | ||
| } | ||
|
|
||
| // When data and partition schemas have overlapping columns, the output | ||
| // schema respects the order of the data schema for the overlapping columns, and it | ||
| // respects the data types of the partition schema. | ||
| override def schema: StructType = { | ||
| StructType(tableStructSchema.map(f => overlappedPartCols.getOrElse(getColName(f), f)) ++ | ||
| partitionStructSchema.filterNot(f => overlappedPartCols.contains(getColName(f)))) | ||
| } | ||
|
|
||
| override def needConversion: Boolean = false | ||
|
|
||
| override def buildScan(requiredColumns: Seq[Attribute], filters: Seq[Expression]): RDD[Row] = { | ||
| fileIndex = buildFileIndex(filters) | ||
| val pushedFilters = PushDownUtils.transformFilter(filters) | ||
| log.debug(s" buildScan requiredColumns = ${requiredColumns.mkString(",")}") | ||
| log.debug(s" buildScan filters = ${pushedFilters.mkString(",")}") | ||
| var requiredStructSchema = StructType(Seq()) | ||
| requiredColumns.foreach(col => { | ||
| val field = tableStructSchema.find(_.name == col.name) | ||
| if (field.isDefined) { | ||
| requiredStructSchema = requiredStructSchema.add(field.get) | ||
| } | ||
|
|
@@ -106,18 +129,18 @@ class MergeOnReadSnapshotRelation(val sqlContext: SQLContext, | |
| val fullSchemaParquetReader = new ParquetFileFormat().buildReaderWithPartitionValues( | ||
| sparkSession = sqlContext.sparkSession, | ||
| dataSchema = tableStructSchema, | ||
| partitionSchema = partitionStructSchema, | ||
| requiredSchema = tableStructSchema, | ||
| filters = pushedFilters, | ||
| options = optParams, | ||
| hadoopConf = sqlContext.sparkSession.sessionState.newHadoopConf() | ||
| ) | ||
| val requiredSchemaParquetReader = new ParquetFileFormat().buildReaderWithPartitionValues( | ||
| sparkSession = sqlContext.sparkSession, | ||
| dataSchema = tableStructSchema, | ||
| partitionSchema = partitionStructSchema, | ||
| requiredSchema = requiredStructSchema, | ||
| filters = pushedFilters, | ||
| options = optParams, | ||
| hadoopConf = sqlContext.sparkSession.sessionState.newHadoopConf() | ||
| ) | ||
|
|
@@ -132,9 +155,14 @@ class MergeOnReadSnapshotRelation(val sqlContext: SQLContext, | |
| rdd.asInstanceOf[RDD[Row]] | ||
| } | ||
|
|
||
| def buildFileIndex(filters: Seq[Expression]): List[HoodieMergeOnReadFileSplit] = { | ||
| val selectedPartitions = inMemoryFileIndex.listFiles(filters, filters) | ||
| val selectedPartitionsPathMap = selectedPartitions.flatMap(x=>{ | ||
| val files = x.files | ||
| val fileMap = files.map(file=>{(file.getPath.getName,x.values)}) | ||
| fileMap | ||
| }).toMap | ||
| val fileStatuses = selectedPartitions.flatMap(_.files) | ||
| if (fileStatuses.isEmpty) { | ||
| throw new HoodieException("No files found for reading in user provided path.") | ||
| } | ||
|
|
@@ -148,10 +176,22 @@ class MergeOnReadSnapshotRelation(val sqlContext: SQLContext, | |
| val fileSplits = fileGroup.map(kv => { | ||
| val baseFile = kv._1 | ||
| val logPaths = if (kv._2.isEmpty) Option.empty else Option(kv._2.asScala.toList) | ||
| val partitionValues = selectedPartitionsPathMap.get(baseFile.getFileName).get | ||
| val partitionedFile = PartitionedFile(partitionValues, baseFile.getPath, 0, baseFile.getFileLen) | ||
| HoodieMergeOnReadFileSplit(Option(partitionedFile), logPaths, latestCommit, | ||
| metaClient.getBasePath, maxCompactionMemoryInBytes, mergeType) | ||
| }).toList | ||
| fileSplits | ||
| } | ||
|
|
||
| private def getColName(f: StructField): String = { | ||
| if (sparkSession.sessionState.conf.caseSensitiveAnalysis) { | ||
| f.name | ||
| } else { | ||
| f.name.toLowerCase(Locale.ROOT) | ||
| } | ||
| } | ||
|
|
||
| @VisibleForTesting | ||
| def getFileIndexPaths = fileIndex.map(x => x.dataFile.get.filePath) | ||
|
Contributor
There was a problem hiding this comment. Is there any annotation like VisibleOnlyForTests as we have in java. Not a big fan of introducing public methods in source code just for test purposes. |
||
| } | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -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.spark.sql.execution.hudi.utils | ||
|
|
||
| import org.apache.spark.sql.catalyst.expressions.Expression | ||
| import org.apache.spark.sql.execution.datasources.DataSourceStrategy | ||
| import org.apache.spark.sql.execution.datasources.DataSourceStrategy.translateFilter | ||
| import org.apache.spark.sql.sources.{BaseRelation, Filter} | ||
|
|
||
| /** | ||
| * This util object is use DataSourceStrategy protected translateFilter method | ||
| * [https://github.com/apache/spark/blob/v2.4.4/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/DataSourceStrategy.scala#L439] | ||
| */ | ||
| object PushDownUtils { | ||
|
|
||
|
|
||
| /** | ||
| * Tries to translate a Catalyst Seq[Expression] into data source Array[Filter]. | ||
| */ | ||
| def transformFilter(filterPredicates: Seq[Expression]): Array[Filter] = { | ||
| val translatedMap: Map[Expression, Filter] = filterPredicates.flatMap { p => | ||
| translateFilter(p).map(f => p -> f) | ||
| }.toMap | ||
| translatedMap.values.toArray | ||
| } | ||
| } | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
minor: isHoodieMeta"data"Path