Github user JoshRosen commented on a diff in the pull request:
https://github.com/apache/spark/pull/2940#discussion_r19499585
--- Diff:
streaming/src/main/scala/org/apache/spark/streaming/receiver/ReceivedBlockHandler.scala
---
@@ -0,0 +1,144 @@
+package org.apache.spark.streaming.receiver
+
+import java.nio.ByteBuffer
+
+import scala.collection.mutable.ArrayBuffer
+import scala.concurrent.{Await, ExecutionContext, Future}
+import scala.concurrent.duration._
+import scala.language.{existentials, postfixOps}
+
+import WriteAheadLogBasedBlockHandler._
+import org.apache.hadoop.conf.Configuration
+import org.apache.hadoop.fs.Path
+import org.apache.spark.{SparkException, Logging, SparkConf}
+import org.apache.spark.storage.{BlockManager, StorageLevel, StreamBlockId}
+import org.apache.spark.streaming.util.{Clock, SystemClock,
WriteAheadLogManager}
+import org.apache.spark.util.Utils
+
+private[streaming] sealed trait ReceivedBlock
+private[streaming] case class ArrayBufferBlock(arrayBuffer:
ArrayBuffer[_]) extends ReceivedBlock
+private[streaming] case class IteratorBlock(iterator: Iterator[_]) extends
ReceivedBlock
+private[streaming] case class ByteBufferBlock(byteBuffer: ByteBuffer)
extends ReceivedBlock
+
+
+/** Trait that represents a class that handles the storage of blocks
received by receiver */
+private[streaming] trait ReceivedBlockHandler {
+
+ /** Store a received block with the given block id */
+ def storeBlock(blockId: StreamBlockId, receivedBlock: ReceivedBlock):
Option[AnyRef]
+
+ /** Cleanup old blocks older than the given threshold time */
+ def cleanupOldBlock(threshTime: Long)
+}
+
+/**
+ * Implementation of a
[[org.apache.spark.streaming.receiver.ReceivedBlockHandler]] which
+ * stores the received blocks into a block manager with the specified
storage level.
+ */
+private[streaming] class BlockManagerBasedBlockHandler(
+ blockManager: BlockManager, storageLevel: StorageLevel)
+ extends ReceivedBlockHandler with Logging {
+
+ def storeBlock(blockId: StreamBlockId, receivedBlock: ReceivedBlock):
Option[AnyRef] = {
+ val putResult = receivedBlock match {
+ case ArrayBufferBlock(arrayBuffer) =>
+ blockManager.putIterator(blockId, arrayBuffer.iterator,
storageLevel, tellMaster = true)
+ case IteratorBlock(iterator) =>
+ blockManager.putIterator(blockId, iterator, storageLevel,
tellMaster = true)
+ case ByteBufferBlock(byteBuffer) =>
+ blockManager.putBytes(blockId, byteBuffer, storageLevel,
tellMaster = true)
+ case _ =>
+ throw new SparkException(s"Could not store $blockId to block
manager, unexpected block type")
--- End diff --
For these types of "unexpected case" errors, I find it helpful to log the
name of the unexpected object's class, e.g.
```scala
case o =>
throw new SparkException(s"... unexpected block type
${o.getClass.getName}")
```
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
