Github user tdas commented on a diff in the pull request: https://github.com/apache/spark/pull/21477#discussion_r193892514 --- Diff: sql/core/src/main/scala/org/apache/spark/sql/execution/python/PythonForeachWriter.scala --- @@ -0,0 +1,161 @@ +/* + * 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.python + +import java.io.File +import java.util.concurrent.TimeUnit +import java.util.concurrent.locks.ReentrantLock + +import org.apache.spark.{SparkEnv, TaskContext} +import org.apache.spark.api.python._ +import org.apache.spark.internal.Logging +import org.apache.spark.memory.TaskMemoryManager +import org.apache.spark.sql.ForeachWriter +import org.apache.spark.sql.catalyst.expressions.UnsafeRow +import org.apache.spark.sql.types.StructType +import org.apache.spark.util.{NextIterator, Utils} + +class PythonForeachWriter(func: PythonFunction, schema: StructType) + extends ForeachWriter[UnsafeRow] { + + private lazy val context = TaskContext.get() + private lazy val buffer = new PythonForeachWriter.UnsafeRowBuffer( + context.taskMemoryManager, new File(Utils.getLocalDir(SparkEnv.get.conf)), schema.fields.length) + private lazy val inputRowIterator = buffer.iterator + + private lazy val inputByteIterator = { + EvaluatePython.registerPicklers() + val objIterator = inputRowIterator.map { row => EvaluatePython.toJava(row, schema) } + new SerDeUtil.AutoBatchedPickler(objIterator) + } + + private lazy val pythonRunner = { + val conf = SparkEnv.get.conf + val bufferSize = conf.getInt("spark.buffer.size", 65536) + val reuseWorker = conf.getBoolean("spark.python.worker.reuse", true) + PythonRunner(func, bufferSize, reuseWorker) + } + + private lazy val outputIterator = + pythonRunner.compute(inputByteIterator, context.partitionId(), context) + + override def open(partitionId: Long, version: Long): Boolean = { + outputIterator // initialize everything + TaskContext.get.addTaskCompletionListener { _ => buffer.close() } + true + } + + override def process(value: UnsafeRow): Unit = { + buffer.add(value) + } + + override def close(errorOrNull: Throwable): Unit = { + buffer.allRowsAdded() + if (outputIterator.hasNext) outputIterator.next() // to throw python exception if there was one + } +} + +object PythonForeachWriter { + + /** + * A buffer that is designed for the sole purpose of buffering UnsafeRows in PythonForeahWriter. --- End diff -- done.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org