All Downloads are FREE. Search and download functionalities are using the official Maven repository.

org.apache.spark.sql.execution.SQLExecution.scala Maven / Gradle / Ivy

There is a newer version: 2.4.8
Show newest version
/*
 * 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

import java.util.concurrent.ConcurrentHashMap
import java.util.concurrent.atomic.AtomicLong

import org.apache.spark.SparkContext
import org.apache.spark.sql.SparkSession
import org.apache.spark.sql.execution.ui.{SparkListenerSQLExecutionEnd, SparkListenerSQLExecutionStart}

object SQLExecution {

  val EXECUTION_ID_KEY = "spark.sql.execution.id"

  private val _nextExecutionId = new AtomicLong(0)

  private def nextExecutionId: Long = _nextExecutionId.getAndIncrement

  private val executionIdToQueryExecution = new ConcurrentHashMap[Long, QueryExecution]()

  def getQueryExecution(executionId: Long): QueryExecution = {
    executionIdToQueryExecution.get(executionId)
  }

  private val testing = sys.props.contains("spark.testing")

  private[sql] def checkSQLExecutionId(sparkSession: SparkSession): Unit = {
    val sc = sparkSession.sparkContext
    // only throw an exception during tests. a missing execution ID should not fail a job.
    if (testing && sc.getLocalProperty(EXECUTION_ID_KEY) == null) {
      // Attention testers: when a test fails with this exception, it means that the action that
      // started execution of a query didn't call withNewExecutionId. The execution ID should be
      // set by calling withNewExecutionId in the action that begins execution, like
      // Dataset.collect or DataFrameWriter.insertInto.
      throw new IllegalStateException("Execution ID should be set")
    }
  }

  /**
   * Wrap an action that will execute "queryExecution" to track all Spark jobs in the body so that
   * we can connect them with an execution.
   */
  def withNewExecutionId[T](
      sparkSession: SparkSession,
      queryExecution: QueryExecution)(body: => T): T = {
    val sc = sparkSession.sparkContext
    val oldExecutionId = sc.getLocalProperty(EXECUTION_ID_KEY)
    val executionId = SQLExecution.nextExecutionId
    sc.setLocalProperty(EXECUTION_ID_KEY, executionId.toString)
    executionIdToQueryExecution.put(executionId, queryExecution)
    try {
      // sparkContext.getCallSite() would first try to pick up any call site that was previously
      // set, then fall back to Utils.getCallSite(); call Utils.getCallSite() directly on
      // streaming queries would give us call site like "run at :0"
      val callSite = sparkSession.sparkContext.getCallSite()

      sparkSession.sparkContext.listenerBus.post(SparkListenerSQLExecutionStart(
        executionId, callSite.shortForm, callSite.longForm, queryExecution.toString,
        SparkPlanInfo.fromSparkPlan(queryExecution.executedPlan), System.currentTimeMillis()))
      try {
        body
      } finally {
        sparkSession.sparkContext.listenerBus.post(SparkListenerSQLExecutionEnd(
          executionId, System.currentTimeMillis()))
      }
    } finally {
      executionIdToQueryExecution.remove(executionId)
      sc.setLocalProperty(EXECUTION_ID_KEY, oldExecutionId)
    }
  }

  /**
   * Wrap an action with a known executionId. When running a different action in a different
   * thread from the original one, this method can be used to connect the Spark jobs in this action
   * with the known executionId, e.g., `BroadcastHashJoin.broadcastFuture`.
   */
  def withExecutionId[T](sc: SparkContext, executionId: String)(body: => T): T = {
    val oldExecutionId = sc.getLocalProperty(SQLExecution.EXECUTION_ID_KEY)
    try {
      sc.setLocalProperty(SQLExecution.EXECUTION_ID_KEY, executionId)
      body
    } finally {
      sc.setLocalProperty(SQLExecution.EXECUTION_ID_KEY, oldExecutionId)
    }
  }
}




© 2015 - 2024 Weber Informatics LLC | Privacy Policy