spark ParquetScan 源码

spark ParquetScan 代码

文件路径：/sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/parquet/ParquetScan.scala

/*
 * 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.datasources.v2.parquet

import scala.collection.JavaConverters._

import org.apache.hadoop.conf.Configuration
import org.apache.hadoop.fs.Path
import org.apache.parquet.hadoop.ParquetInputFormat

import org.apache.spark.sql.SparkSession
import org.apache.spark.sql.catalyst.expressions.Expression
import org.apache.spark.sql.connector.expressions.aggregate.Aggregation
import org.apache.spark.sql.connector.read.PartitionReaderFactory
import org.apache.spark.sql.execution.datasources.{AggregatePushDownUtils, PartitioningAwareFileIndex, RowIndexUtil}
import org.apache.spark.sql.execution.datasources.parquet.{ParquetOptions, ParquetReadSupport, ParquetWriteSupport}
import org.apache.spark.sql.execution.datasources.v2.FileScan
import org.apache.spark.sql.internal.SQLConf
import org.apache.spark.sql.sources.Filter
import org.apache.spark.sql.types.StructType
import org.apache.spark.sql.util.CaseInsensitiveStringMap
import org.apache.spark.util.SerializableConfiguration

case class ParquetScan(
    sparkSession: SparkSession,
    hadoopConf: Configuration,
    fileIndex: PartitioningAwareFileIndex,
    dataSchema: StructType,
    readDataSchema: StructType,
    readPartitionSchema: StructType,
    pushedFilters: Array[Filter],
    options: CaseInsensitiveStringMap,
    pushedAggregate: Option[Aggregation] = None,
    partitionFilters: Seq[Expression] = Seq.empty,
    dataFilters: Seq[Expression] = Seq.empty) extends FileScan {
  override def isSplitable(path: Path): Boolean = {
    // If aggregate is pushed down, only the file footer will be read once,
    // so file should not be split across multiple tasks.
    pushedAggregate.isEmpty &&
      // SPARK-39634: Allow file splitting in combination with row index generation once
      // the fix for PARQUET-2161 is available.
      !RowIndexUtil.isNeededForSchema(readSchema)
  }

  override def readSchema(): StructType = {
    // If aggregate is pushed down, schema has already been pruned in `ParquetScanBuilder`
    // and no need to call super.readSchema()
    if (pushedAggregate.nonEmpty) readDataSchema else super.readSchema()
  }

  override def createReaderFactory(): PartitionReaderFactory = {
    val readDataSchemaAsJson = readDataSchema.json
    hadoopConf.set(ParquetInputFormat.READ_SUPPORT_CLASS, classOf[ParquetReadSupport].getName)
    hadoopConf.set(
      ParquetReadSupport.SPARK_ROW_REQUESTED_SCHEMA,
      readDataSchemaAsJson)
    hadoopConf.set(
      ParquetWriteSupport.SPARK_ROW_SCHEMA,
      readDataSchemaAsJson)
    hadoopConf.set(
      SQLConf.SESSION_LOCAL_TIMEZONE.key,
      sparkSession.sessionState.conf.sessionLocalTimeZone)
    hadoopConf.setBoolean(
      SQLConf.NESTED_SCHEMA_PRUNING_ENABLED.key,
      sparkSession.sessionState.conf.nestedSchemaPruningEnabled)
    hadoopConf.setBoolean(
      SQLConf.CASE_SENSITIVE.key,
      sparkSession.sessionState.conf.caseSensitiveAnalysis)

    // Sets flags for `ParquetToSparkSchemaConverter`
    hadoopConf.setBoolean(
      SQLConf.PARQUET_BINARY_AS_STRING.key,
      sparkSession.sessionState.conf.isParquetBinaryAsString)
    hadoopConf.setBoolean(
      SQLConf.PARQUET_INT96_AS_TIMESTAMP.key,
      sparkSession.sessionState.conf.isParquetINT96AsTimestamp)
    hadoopConf.setBoolean(
      SQLConf.PARQUET_TIMESTAMP_NTZ_ENABLED.key,
      sparkSession.sessionState.conf.parquetTimestampNTZEnabled)

    val broadcastedConf = sparkSession.sparkContext.broadcast(
      new SerializableConfiguration(hadoopConf))
    val sqlConf = sparkSession.sessionState.conf
    ParquetPartitionReaderFactory(
      sqlConf,
      broadcastedConf,
      dataSchema,
      readDataSchema,
      readPartitionSchema,
      pushedFilters,
      pushedAggregate,
      new ParquetOptions(options.asCaseSensitiveMap.asScala.toMap, sqlConf))
  }

  override def equals(obj: Any): Boolean = obj match {
    case p: ParquetScan =>
      val pushedDownAggEqual = if (pushedAggregate.nonEmpty && p.pushedAggregate.nonEmpty) {
        AggregatePushDownUtils.equivalentAggregations(pushedAggregate.get, p.pushedAggregate.get)
      } else {
        pushedAggregate.isEmpty && p.pushedAggregate.isEmpty
      }
      super.equals(p) && dataSchema == p.dataSchema && options == p.options &&
        equivalentFilters(pushedFilters, p.pushedFilters) && pushedDownAggEqual
    case _ => false
  }

  override def hashCode(): Int = getClass.hashCode()

  lazy private val (pushedAggregationsStr, pushedGroupByStr) = if (pushedAggregate.nonEmpty) {
    (seqToString(pushedAggregate.get.aggregateExpressions),
      seqToString(pushedAggregate.get.groupByExpressions))
  } else {
    ("[]", "[]")
  }

  override def description(): String = {
    super.description() + ", PushedFilters: " + seqToString(pushedFilters) +
      ", PushedAggregation: " + pushedAggregationsStr +
      ", PushedGroupBy: " + pushedGroupByStr
  }

  override def getMetaData(): Map[String, String] = {
    super.getMetaData() ++ Map("PushedFilters" -> seqToString(pushedFilters)) ++
      Map("PushedAggregation" -> pushedAggregationsStr) ++
      Map("PushedGroupBy" -> pushedGroupByStr)
  }
}

相关信息

spark 源码目录

相关文章

spark ParquetDataSourceV2 源码

spark ParquetPartitionReaderFactory 源码

spark ParquetScanBuilder 源码

spark ParquetTable 源码

spark ParquetWrite 源码