Github user koeninger commented on a diff in the pull request:
https://github.com/apache/spark/pull/4805#discussion_r25990749
--- Diff:
external/kafka/src/main/scala/org/apache/spark/streaming/kafka/DirectKafkaInputDStream.scala
---
@@ -158,4 +166,37 @@ class DirectKafkaInputDStream[
}
}
+ private[streaming]
+ class DirectKafkaStreamingListener extends StreamingListener {
+
+ val offsetCommitEnabled =
context.conf.getBoolean("spark.streaming.kafka.offsetCommit", false)
+
+ val zkClientOpt = if (offsetCommitEnabled) {
+ kafkaParams.get("zookeeper.connect").map { zkConnect =>
+ val zkSessionTimeoutMs =
kafkaParams.getOrElse("zookeeper.session.timeout.ms", "6000").toInt
+ val zkConnectionTimeoutMs =
+ kafkaParams.getOrElse("zookeeper.session.timeout.ms",
"6000").toInt
+ new ZkClient(zkConnect, zkSessionTimeoutMs, zkConnectionTimeoutMs,
ZKStringSerializer)
+ }
+ } else {
+ None
+ }
+
+ val groupId = "directKafkaConsumer"
+
+ override def onBatchCompleted(batchCompleted:
StreamingListenerBatchCompleted) = synchronized {
+ if (offsetCommitEnabled) {
+ for (offsets <- offsetMap.get(batchCompleted.batchInfo.batchTime);
client <- zkClientOpt) {
--- End diff --
if offsetCommitEnabled, zkClientOpt should always be defined, yes? Why two
separate checks?
From a style point of view, people were complaining about monadic syntax
earlier, I don't know that semicolon separated on a single line is an
improvement.
---
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
