Github user shivaram commented on a diff in the pull request:
https://github.com/apache/spark/pull/5096#discussion_r27847867
--- Diff: core/src/main/scala/org/apache/spark/api/r/RRDD.scala ---
@@ -0,0 +1,450 @@
+/*
+ * 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.api.r
+
+import java.io._
+import java.net.ServerSocket
+import java.util.{Map => JMap}
+
+import scala.collection.JavaConversions._
+import scala.io.Source
+import scala.reflect.ClassTag
+import scala.util.Try
+
+import org.apache.spark._
+import org.apache.spark.api.java.{JavaPairRDD, JavaRDD, JavaSparkContext}
+import org.apache.spark.broadcast.Broadcast
+import org.apache.spark.rdd.RDD
+import org.apache.spark.util.Utils
+
+private abstract class BaseRRDD[T: ClassTag, U: ClassTag](
+ parent: RDD[T],
+ numPartitions: Int,
+ func: Array[Byte],
+ deserializer: String,
+ serializer: String,
+ packageNames: Array[Byte],
+ rLibDir: String,
+ broadcastVars: Array[Broadcast[Object]])
+ extends RDD[U](parent) with Logging {
+ override def getPartitions = parent.partitions
+
+ override def compute(partition: Partition, context: TaskContext):
Iterator[U] = {
+
+ // The parent may be also an RRDD, so we should launch it first.
+ val parentIterator = firstParent[T].iterator(partition, context)
+
+ // we expect two connections
+ val serverSocket = new ServerSocket(0, 2)
+ val listenPort = serverSocket.getLocalPort()
+
+ // The stdout/stderr is shared by multiple tasks, because we use one
daemon
+ // to launch child process as worker.
+ val errThread = RRDD.createRWorker(rLibDir, listenPort)
+
+ // We use two sockets to separate input and output, then it's easy to
manage
+ // the lifecycle of them to avoid deadlock.
+ // TODO: optimize it to use one socket
+
+ // the socket used to send out the input of task
+ serverSocket.setSoTimeout(10000)
+ val inSocket = serverSocket.accept()
+ startStdinThread(inSocket.getOutputStream(), parentIterator,
partition.index)
+
+ // the socket used to receive the output of task
+ val outSocket = serverSocket.accept()
+ val inputStream = new BufferedInputStream(outSocket.getInputStream)
+ val dataStream = openDataStream(inputStream)
+ serverSocket.close()
+
+ try {
+
+ return new Iterator[U] {
+ def next(): U = {
+ val obj = _nextObj
+ if (hasNext) {
+ _nextObj = read()
+ }
+ obj
+ }
+
+ var _nextObj = read()
+
+ def hasNext(): Boolean = {
+ val hasMore = (_nextObj != null)
+ if (!hasMore) {
+ dataStream.close()
+ }
+ hasMore
+ }
+ }
+ } catch {
+ case e: Exception =>
+ throw new SparkException("R computation failed with\n " +
errThread.getLines())
+ }
+ }
+
+ /**
+ * Start a thread to write RDD data to the R process.
+ */
+ private def startStdinThread[T](
+ output: OutputStream,
+ iter: Iterator[T],
+ partition: Int) = {
--- End diff --
Done
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
