[SPARK-1332] Improve Spark Streaming's Network Receiver and InputDStream API [WIP]

The current Network Receiver API makes it slightly complicated to right a new receiver as one needs to create an instance of BlockGenerator as shown in SocketReceiver
https://github.com/apache/spark/blob/master/streaming/src/main/scala/org/apache/spark/streaming/dstream/SocketInputDStream.scala#L51

Exposing the BlockGenerator interface has made it harder to improve the receiving process. The API of NetworkReceiver (which was not a very stable API anyways) needs to be change if we are to ensure future stability.

Additionally, the functions like streamingContext.socketStream that create input streams, return DStream objects. That makes it hard to expose functionality (say, rate limits) unique to input dstreams. They should return InputDStream or NetworkInputDStream. This is still not yet implemented.

This PR is blocked on the graceful shutdown PR #247

Author: Tathagata Das <tathagata.das1565@gmail.com>

Closes #300 from tdas/network-receiver-api and squashes the following commits:

ea27b38 [Tathagata Das] Merge remote-tracking branch 'apache-github/master' into network-receiver-api
3a4777c [Tathagata Das] Renamed NetworkInputDStream to ReceiverInputDStream, and ActorReceiver related stuff.
838dd39 [Tathagata Das] Added more events to the StreamingListener to report errors and stopped receivers.
a75c7a6 [Tathagata Das] Address some PR comments and fixed other issues.
91bfa72 [Tathagata Das] Fixed bugs.
8533094 [Tathagata Das] Scala style fixes.
028bde6 [Tathagata Das] Further refactored receiver to allow restarting of a receiver.
43f5290 [Tathagata Das] Made functions that create input streams return InputDStream and NetworkInputDStream, for both Scala and Java.
2c94579 [Tathagata Das] Fixed graceful shutdown by removing interrupts on receiving thread.
9e37a0b [Tathagata Das] Merge remote-tracking branch 'apache-github/master' into network-receiver-api
3223e95 [Tathagata Das] Refactored the code that runs the NetworkReceiver into further classes and traits to make them more testable.
a36cc48 [Tathagata Das] Refactored the NetworkReceiver API for future stability.

4 files changed, 26 insertions, 24 deletions

diff --git a/external/flume/src/main/scala/org/apache/spark/streaming/flume/FlumeInputDStream.scala b/external/flume/src/main/scala/org/apache/spark/streaming/flume/FlumeInputDStream.scala
index 34012b846e..df7605fe57 100644
--- a/external/flume/src/main/scala/org/apache/spark/streaming/flume/FlumeInputDStream.scala
+++ b/external/flume/src/main/scala/org/apache/spark/streaming/flume/FlumeInputDStream.scala
@@ -34,6 +34,8 @@ import org.apache.spark.util.Utils
 import org.apache.spark.storage.StorageLevel
 import org.apache.spark.streaming.StreamingContext
 import org.apache.spark.streaming.dstream._
+import org.apache.spark.Logging
+import org.apache.spark.streaming.receiver.Receiver
 
 private[streaming]
 class FlumeInputDStream[T: ClassTag](
@@ -41,9 +43,9 @@ class FlumeInputDStream[T: ClassTag](
   host: String,
   port: Int,
   storageLevel: StorageLevel
-) extends NetworkInputDStream[SparkFlumeEvent](ssc_) {
+) extends ReceiverInputDStream[SparkFlumeEvent](ssc_) {
 
-  override def getReceiver(): NetworkReceiver[SparkFlumeEvent] = {
+  override def getReceiver(): Receiver[SparkFlumeEvent] = {
     new FlumeReceiver(host, port, storageLevel)
   }
 }
@@ -115,13 +117,13 @@ private[streaming] object SparkFlumeEvent {
 private[streaming]
 class FlumeEventServer(receiver : FlumeReceiver) extends AvroSourceProtocol {
   override def append(event : AvroFlumeEvent) : Status = {
-    receiver.blockGenerator += SparkFlumeEvent.fromAvroFlumeEvent(event)
+    receiver.store(SparkFlumeEvent.fromAvroFlumeEvent(event))
     Status.OK
   }
 
   override def appendBatch(events : java.util.List[AvroFlumeEvent]) : Status = {
     events.foreach (event =>
-      receiver.blockGenerator += SparkFlumeEvent.fromAvroFlumeEvent(event))
+      receiver.store(SparkFlumeEvent.fromAvroFlumeEvent(event)))
     Status.OK
   }
 }
@@ -133,23 +135,21 @@ class FlumeReceiver(
     host: String,
     port: Int,
     storageLevel: StorageLevel
-  ) extends NetworkReceiver[SparkFlumeEvent] {
+  ) extends Receiver[SparkFlumeEvent](storageLevel) with Logging {
 
-  lazy val blockGenerator = new BlockGenerator(storageLevel)
+  lazy val responder = new SpecificResponder(
+    classOf[AvroSourceProtocol], new FlumeEventServer(this))
+  lazy val server = new NettyServer(responder, new InetSocketAddress(host, port))
 
-  protected override def onStart() {
-    val responder = new SpecificResponder(
-      classOf[AvroSourceProtocol], new FlumeEventServer(this))
-    val server = new NettyServer(responder, new InetSocketAddress(host, port))
-    blockGenerator.start()
+  def onStart() {
     server.start()
     logInfo("Flume receiver started")
   }
 
-  protected override def onStop() {
-    blockGenerator.stop()
+  def onStop() {
+    server.close()
     logInfo("Flume receiver stopped")
   }
 
-  override def getLocationPreference = Some(host)
+  override def preferredLocation = Some(host)
 }
diff --git a/external/flume/src/main/scala/org/apache/spark/streaming/flume/FlumeUtils.scala b/external/flume/src/main/scala/org/apache/spark/streaming/flume/FlumeUtils.scala
index 654ba451e7..499f3560ef 100644
--- a/external/flume/src/main/scala/org/apache/spark/streaming/flume/FlumeUtils.scala
+++ b/external/flume/src/main/scala/org/apache/spark/streaming/flume/FlumeUtils.scala
@@ -19,8 +19,8 @@ package org.apache.spark.streaming.flume
 
 import org.apache.spark.storage.StorageLevel
 import org.apache.spark.streaming.StreamingContext
-import org.apache.spark.streaming.api.java.{JavaStreamingContext, JavaDStream}
-import org.apache.spark.streaming.dstream.DStream
+import org.apache.spark.streaming.api.java.{JavaReceiverInputDStream, JavaInputDStream, JavaStreamingContext, JavaDStream}
+import org.apache.spark.streaming.dstream.{ReceiverInputDStream, DStream}
 
 object FlumeUtils {
   /**
@@ -35,7 +35,7 @@ object FlumeUtils {
       hostname: String,
       port: Int,
       storageLevel: StorageLevel = StorageLevel.MEMORY_AND_DISK_SER_2
-    ): DStream[SparkFlumeEvent] = {
+    ): ReceiverInputDStream[SparkFlumeEvent] = {
     val inputStream = new FlumeInputDStream[SparkFlumeEvent](ssc, hostname, port, storageLevel)
     inputStream
   }
@@ -50,7 +50,7 @@ object FlumeUtils {
       jssc: JavaStreamingContext,
       hostname: String,
       port: Int
-    ): JavaDStream[SparkFlumeEvent] = {
+    ): JavaReceiverInputDStream[SparkFlumeEvent] = {
     createStream(jssc.ssc, hostname, port)
   }
 
@@ -65,7 +65,7 @@ object FlumeUtils {
       hostname: String,
       port: Int,
       storageLevel: StorageLevel
-    ): JavaDStream[SparkFlumeEvent] = {
+    ): JavaReceiverInputDStream[SparkFlumeEvent] = {
     createStream(jssc.ssc, hostname, port, storageLevel)
   }
 }
diff --git a/external/flume/src/test/java/org/apache/spark/streaming/flume/JavaFlumeStreamSuite.java b/external/flume/src/test/java/org/apache/spark/streaming/flume/JavaFlumeStreamSuite.java
index 733389b98d..e0ad4f1015 100644
--- a/external/flume/src/test/java/org/apache/spark/streaming/flume/JavaFlumeStreamSuite.java
+++ b/external/flume/src/test/java/org/apache/spark/streaming/flume/JavaFlumeStreamSuite.java
@@ -19,16 +19,16 @@ package org.apache.spark.streaming.flume;
 
 import org.apache.spark.storage.StorageLevel;
 import org.apache.spark.streaming.LocalJavaStreamingContext;
-import org.apache.spark.streaming.api.java.JavaDStream;
 
+import org.apache.spark.streaming.api.java.JavaReceiverInputDStream;
 import org.junit.Test;
 
 public class JavaFlumeStreamSuite extends LocalJavaStreamingContext {
   @Test
   public void testFlumeStream() {
     // tests the API, does not actually test data receiving
-    JavaDStream<SparkFlumeEvent> test1 = FlumeUtils.createStream(ssc, "localhost", 12345);
-    JavaDStream<SparkFlumeEvent> test2 = FlumeUtils.createStream(ssc, "localhost", 12345,
+    JavaReceiverInputDStream<SparkFlumeEvent> test1 = FlumeUtils.createStream(ssc, "localhost", 12345);
+    JavaReceiverInputDStream<SparkFlumeEvent> test2 = FlumeUtils.createStream(ssc, "localhost", 12345,
       StorageLevel.MEMORY_AND_DISK_SER_2());
   }
 }
diff --git a/external/flume/src/test/scala/org/apache/spark/streaming/flume/FlumeStreamSuite.scala b/external/flume/src/test/scala/org/apache/spark/streaming/flume/FlumeStreamSuite.scala
index 8bc43972ab..78603200d2 100644
--- a/external/flume/src/test/scala/org/apache/spark/streaming/flume/FlumeStreamSuite.scala
+++ b/external/flume/src/test/scala/org/apache/spark/streaming/flume/FlumeStreamSuite.scala
@@ -31,6 +31,7 @@ import org.apache.flume.source.avro.{AvroFlumeEvent, AvroSourceProtocol}
 import org.apache.spark.storage.StorageLevel
 import org.apache.spark.streaming.{TestOutputStream, StreamingContext, TestSuiteBase}
 import org.apache.spark.streaming.util.ManualClock
+import org.apache.spark.streaming.api.java.JavaReceiverInputDStream
 
 class FlumeStreamSuite extends TestSuiteBase {
 
@@ -39,10 +40,11 @@ class FlumeStreamSuite extends TestSuiteBase {
   test("flume input stream") {
     // Set up the streaming context and input streams
     val ssc = new StreamingContext(conf, batchDuration)
-    val flumeStream = FlumeUtils.createStream(ssc, "localhost", testPort, StorageLevel.MEMORY_AND_DISK)
+    val flumeStream: JavaReceiverInputDStream[SparkFlumeEvent] =
+      FlumeUtils.createStream(ssc, "localhost", testPort, StorageLevel.MEMORY_AND_DISK)
     val outputBuffer = new ArrayBuffer[Seq[SparkFlumeEvent]]
       with SynchronizedBuffer[Seq[SparkFlumeEvent]]
-    val outputStream = new TestOutputStream(flumeStream, outputBuffer)
+    val outputStream = new TestOutputStream(flumeStream.receiverInputDStream, outputBuffer)
     outputStream.register()
     ssc.start()