Github user koeninger commented on a diff in the pull request:
https://github.com/apache/spark/pull/15102#discussion_r80607882
--- Diff:
external/kafka-0-10-sql/src/main/scala/org/apache/spark/sql/kafka010/CachedKafkaConsumer.scala
---
@@ -0,0 +1,155 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one or more
+ * contributor license agreements. See the NOTICE file distributed with
+ * this work for additional information regarding copyright ownership.
+ * The ASF licenses this file to You under the Apache License, Version 2.0
+ * (the "License"); you may not use this file except in compliance with
+ * the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.spark.sql.kafka010
+
+import java.{util => ju}
+
+import org.apache.kafka.clients.consumer.{ConsumerConfig, ConsumerRecord,
KafkaConsumer}
+import org.apache.kafka.common.TopicPartition
+
+import org.apache.spark.{SparkEnv, SparkException, TaskContext}
+import org.apache.spark.internal.Logging
+
+
+/**
+ * Consumer of single topicpartition, intended for cached reuse.
+ * Underlying consumer is not threadsafe, so neither is this,
+ * but processing the same topicpartition and group id in multiple threads
is usually bad anyway.
+ */
+private[kafka010] case class CachedKafkaConsumer private(
+ topicPartition: TopicPartition,
+ kafkaParams: ju.Map[String, Object]) extends Logging {
+
+ private val consumer = {
+ val c = new KafkaConsumer[Array[Byte], Array[Byte]](kafkaParams)
+ val tps = new ju.ArrayList[TopicPartition]()
+ tps.add(topicPartition)
+ c.assign(tps)
+ c
+ }
+
+ /**
+ * Timeout for polls to the consumer. Since the data should be already
available, the poll
+ * should get the data immediately, and the timeout value should not
matter as long as it is
+ * generous and does not cause timeout when there are not issues.
+ */
+ private val pollTimeoutMs = 60 * 1000
+
+ /** Iterator to the already fetch data */
+ private var fetchedData =
ju.Collections.emptyIterator[ConsumerRecord[Array[Byte], Array[Byte]]]
+ private var nextOffsetInFetchedData = -2L
+
+ /**
+ * Get the record for the given offset, waiting up to timeout ms if IO
is necessary.
+ * Sequential forward access will use buffers, but random access will be
horribly inefficient.
+ */
+ def get(offset: Long): ConsumerRecord[Array[Byte], Array[Byte]] = {
+ logDebug(s"Get $topicPartition nextOffset $nextOffsetInFetchedData
requested $offset")
+ if (offset != nextOffsetInFetchedData) {
+ logInfo(s"Initial fetch for $topicPartition $offset")
+ seek(offset)
+ poll()
+ }
+
+ if (!fetchedData.hasNext()) { poll() }
+ assert(fetchedData.hasNext(),
+ s"Failed to get records for $topicPartition $offset after polling
for $pollTimeoutMs")
+ var record = fetchedData.next()
+
+ if (record.offset != offset) {
+ logInfo(s"Buffer miss for $topicPartition $offset")
+ seek(offset)
+ poll()
--- End diff --
Nor is the timeout used here
---
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
