[GitHub] spark pull request: [SPARK-2377] Python API for Streaming

[tdas <gi...@git.apache.org>]

Github user tdas commented on a diff in the pull request:

    https://github.com/apache/spark/pull/2538#discussion_r18251679
  
    --- Diff: streaming/src/main/scala/org/apache/spark/streaming/api/python/PythonDStream.scala ---
    @@ -0,0 +1,301 @@
    +/*
    + * 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.streaming.api.python
    +
    +import java.io.{ObjectInputStream, ObjectOutputStream}
    +import java.lang.reflect.Proxy
    +import java.util.{ArrayList => JArrayList, List => JList}
    +import scala.collection.JavaConversions._
    +import scala.collection.JavaConverters._
    +
    +import org.apache.spark.api.java._
    +import org.apache.spark.api.python._
    +import org.apache.spark.rdd.RDD
    +import org.apache.spark.storage.StorageLevel
    +import org.apache.spark.streaming.{Interval, Duration, Time}
    +import org.apache.spark.streaming.dstream._
    +import org.apache.spark.streaming.api.java._
    +
    +
    +/**
    + * Interface for Python callback function with three arguments
    + */
    +private[python] trait PythonRDDFunction {
    +  def call(time: Long, rdds: JList[_]): JavaRDD[Array[Byte]]
    +}
    +
    +/**
    + * Wrapper for PythonRDDFunction
    + * TODO: support checkpoint
    + */
    +private[python] class RDDFunction(@transient var pfunc: PythonRDDFunction)
    +  extends function.Function2[JList[JavaRDD[_]], Time, JavaRDD[Array[Byte]]] with Serializable {
    +
    +  def apply(rdd: Option[RDD[_]], time: Time): Option[RDD[Array[Byte]]] = {
    +    Option(pfunc.call(time.milliseconds, List(rdd.map(JavaRDD.fromRDD(_)).orNull).asJava)).map(_.rdd)
    +  }
    +
    +  def apply(rdd: Option[RDD[_]], rdd2: Option[RDD[_]], time: Time): Option[RDD[Array[Byte]]] = {
    +    val rdds = List(rdd.map(JavaRDD.fromRDD(_)).orNull, rdd2.map(JavaRDD.fromRDD(_)).orNull).asJava
    +    Option(pfunc.call(time.milliseconds, rdds)).map(_.rdd)
    +  }
    +
    +  // for function.Function2
    +  def call(rdds: JList[JavaRDD[_]], time: Time): JavaRDD[Array[Byte]] = {
    +    pfunc.call(time.milliseconds, rdds)
    +  }
    +
    +  private def writeObject(out: ObjectOutputStream): Unit = {
    +    assert(PythonDStream.serializer != null, "Serializer has not been registered!")
    +    val bytes = PythonDStream.serializer.serialize(pfunc)
    +    out.writeInt(bytes.length)
    +    out.write(bytes)
    +  }
    +
    +  private def readObject(in: ObjectInputStream): Unit = {
    +    assert(PythonDStream.serializer != null, "Serializer has not been registered!")
    +    val length = in.readInt()
    +    val bytes = new Array[Byte](length)
    +    in.readFully(bytes)
    +    pfunc = PythonDStream.serializer.deserialize(bytes)
    +  }
    +}
    +
    +/**
    + * Inferface for Python Serializer to serialize PythonRDDFunction
    + */
    +private[python] trait PythonRDDFunctionSerializer {
    +  def dumps(id: String): Array[Byte]  //
    +  def loads(bytes: Array[Byte]): PythonRDDFunction
    +}
    +
    +/**
    + * Wrapper for PythonRDDFunctionSerializer
    + */
    +private[python] class RDDFunctionSerializer(pser: PythonRDDFunctionSerializer) {
    +  def serialize(func: PythonRDDFunction): Array[Byte] = {
    +    // get the id of PythonRDDFunction in py4j
    +    val h = Proxy.getInvocationHandler(func.asInstanceOf[Proxy])
    +    val f = h.getClass().getDeclaredField("id");
    +    f.setAccessible(true);
    +    val id = f.get(h).asInstanceOf[String];
    +    pser.dumps(id)
    +  }
    +
    +  def deserialize(bytes: Array[Byte]): PythonRDDFunction = {
    +    pser.loads(bytes)
    +  }
    +}
    +
    +/**
    + * Helper functions
    + */
    +private[python] object PythonDStream {
    +
    +  // A serializer in Python, used to serialize PythonRDDFunction
    +  var serializer: RDDFunctionSerializer = _
    +
    +  // Register a serializer from Python, should be called during initialization
    +  def registerSerializer(ser: PythonRDDFunctionSerializer) = {
    +    serializer = new RDDFunctionSerializer(ser)
    +  }
    +
    +  // helper function for DStream.foreachRDD(),
    +  // cannot be `foreachRDD`, it will confusing py4j
    +  def callForeachRDD(jdstream: JavaDStream[Array[Byte]], pfunc: PythonRDDFunction) {
    +    val func = new RDDFunction((pfunc))
    +    jdstream.dstream.foreachRDD((rdd, time) => func(Some(rdd), time))
    +  }
    +
    +  // convert list of RDD into queue of RDDs, for ssc.queueStream()
    +  def toRDDQueue(rdds: JArrayList[JavaRDD[Array[Byte]]]): java.util.Queue[JavaRDD[Array[Byte]]] = {
    +    val queue = new java.util.LinkedList[JavaRDD[Array[Byte]]]
    +    rdds.forall(queue.add(_))
    +    queue
    +  }
    +}
    +
    +/**
    + * Base class for PythonDStream with some common methods
    + */
    +private[python]
    +abstract class PythonDStream(parent: DStream[_], @transient pfunc: PythonRDDFunction)
    +  extends DStream[Array[Byte]] (parent.ssc) {
    +
    +  val func = new RDDFunction(pfunc)
    +
    +  override def dependencies = List(parent)
    +
    +  override def slideDuration: Duration = parent.slideDuration
    +
    +  val asJavaDStream  = JavaDStream.fromDStream(this)
    +}
    +
    +/**
    + * Transformed DStream in Python.
    + *
    + * If `reuse` is true and the result of the `func` is an PythonRDD, then it will cache it
    + * as an template for future use, this can reduce the Python callbacks.
    + */
    +private[python]
    +class PythonTransformedDStream (parent: DStream[_], @transient pfunc: PythonRDDFunction,
    +                                var reuse: Boolean = false)
    +  extends PythonDStream(parent, pfunc) {
    +
    +  // rdd returned by func
    +  var lastResult: PythonRDD = _
    +
    +  override def compute(validTime: Time): Option[RDD[Array[Byte]]] = {
    +    val rdd = parent.getOrCompute(validTime)
    +    if (rdd.isEmpty) {
    +      return None
    +    }
    +    if (reuse && lastResult != null) {
    +      // use the previous result as the template to generate new RDD
    +      Some(lastResult.copyTo(rdd.get))
    +    } else {
    +      val r = func(rdd, validTime)
    +      if (reuse && r.isDefined && lastResult == null) {
    +        // try to use the result as a template
    +        r.get match {
    +          case pyrdd: PythonRDD =>
    +            if (pyrdd.parent(0) == rdd) {
    +              // only one PythonRDD
    +              lastResult = pyrdd
    +            } else {
    +              // maybe have multiple stages, don't check it anymore
    +              reuse = false
    +            }
    +        }
    +      }
    +      r
    +    }
    +  }
    +}
    +
    +/**
    + * Transformed from two DStreams in Python.
    + */
    +private[python]
    +class PythonTransformed2DStream(parent: DStream[_], parent2: DStream[_],
    +                                @transient pfunc: PythonRDDFunction)
    +  extends DStream[Array[Byte]] (parent.ssc) {
    +
    +  val func = new RDDFunction(pfunc)
    +
    +  override def slideDuration: Duration = parent.slideDuration
    +
    +  override def dependencies = List(parent, parent2)
    +
    +  override def compute(validTime: Time): Option[RDD[Array[Byte]]] = {
    +    func(parent.getOrCompute(validTime), parent2.getOrCompute(validTime), validTime)
    +  }
    +
    +  val asJavaDStream = JavaDStream.fromDStream(this)
    +}
    +
    +/**
    + * similar to StateDStream
    + */
    +private[python]
    +class PythonStateDStream(parent: DStream[Array[Byte]], @transient reduceFunc: PythonRDDFunction)
    +  extends PythonDStream(parent, reduceFunc) {
    +
    +  super.persist(StorageLevel.MEMORY_ONLY)
    +  override val mustCheckpoint = true
    +
    +  override def compute(validTime: Time): Option[RDD[Array[Byte]]] = {
    +    val lastState = getOrCompute(validTime - slideDuration)
    +    val rdd = parent.getOrCompute(validTime)
    +    if (rdd.isDefined) {
    +      func(lastState, rdd, validTime)
    +    } else {
    +      lastState
    +    }
    +  }
    +}
    +
    +/**
    + * similar to ReducedWindowedDStream
    + */
    +private[python]
    +class PythonReducedWindowedDStream(parent: DStream[Array[Byte]],
    --- End diff --
    
    Incorrect scala style. See http://docs.scala-lang.org/style/declarations.html


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org