Github user squito commented on a diff in the pull request:
https://github.com/apache/spark/pull/15505#discussion_r95423240
--- Diff: core/src/main/scala/org/apache/spark/util/Utils.scala ---
@@ -149,7 +149,12 @@ private[spark] object Utils extends Logging {
/** Deserialize an object using Java serialization and the given
ClassLoader */
def deserialize[T](bytes: Array[Byte], loader: ClassLoader): T = {
- val bis = new ByteArrayInputStream(bytes)
+ deserialize(ByteBuffer.wrap(bytes), loader)
+ }
+
+ /** Deserialize an object using Java serialization and the given
ClassLoader */
+ def deserialize[T](bytes: ByteBuffer, loader: ClassLoader): T = {
+ val bis = new ByteBufferInputStream(bytes)
--- End diff --
I think you misunderstood my comment. I agree, this change is fine. Its
correctness is clear, and definitely covered by tests.
My initial concern was that this might be some super-common code path, and
so even a slight modification may effect performance (should be optimized away
by the jit anyway but thought it was worth thinking about).
Anyway, first of all it was a really minor concern, and second of all it
turned out to be totally irrelevant since its *not* a commonly used method. I
was just trying to help out any other reviewers from also trying to track this
down. Sorry for the confusion :)
---
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]
