Github user mridulm commented on a diff in the pull request:
https://github.com/apache/spark/pull/17295#discussion_r106778005
--- Diff: core/src/main/scala/org/apache/spark/storage/BlockManager.scala
---
@@ -56,6 +57,43 @@ private[spark] class BlockResult(
val bytes: Long)
/**
+ * Abstracts away how blocks are stored and provides different ways to
read the underlying block
+ * data. The data for a BlockData instance can only be read once, since it
may be backed by open
+ * file descriptors that change state as data is read.
+ */
+private[spark] trait BlockData {
+
+ def toInputStream(): InputStream
+
+ def toManagedBuffer(): ManagedBuffer
+
+ def toByteBuffer(allocator: Int => ByteBuffer): ChunkedByteBuffer
+
+ def size: Long
+
+ def dispose(): Unit
+
+}
+
+private[spark] class ByteBufferBlockData(
+ val buffer: ChunkedByteBuffer,
+ autoDispose: Boolean = true) extends BlockData {
+
+ override def toInputStream(): InputStream = buffer.toInputStream(dispose
= autoDispose)
+
+ override def toManagedBuffer(): ManagedBuffer = new
NettyManagedBuffer(buffer.toNetty)
+
+ override def toByteBuffer(allocator: Int => ByteBuffer):
ChunkedByteBuffer = {
+ buffer.copy(allocator)
+ }
--- End diff --
autoDispose is not honored for toManagedBuffer and toByteBuffer ?
On first pass, it looks like it is not ...
Also, is the expectation that invoker must manually invoke dispose when not
using toInputStream ?
Would be good to add a comment about this to BlockData trait detailing the
expectation.
---
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
