Many resources are needed to download a project. Please understand that we have to compensate our server costs. Thank you in advance. Project price only 1 $
You can buy this project and download/modify it how often you want.
/*
* Copyright 2018-2024 OVO Energy Limited
*
* SPDX-License-Identifier: Apache-2.0
*/
package fs2.kafka
import scala.annotation.nowarn
import cats.effect.{Async, Outcome, Resource}
import cats.effect.syntax.all.*
import cats.syntax.all.*
import fs2.{Chunk, Stream}
import fs2.kafka.internal.*
import fs2.kafka.internal.converters.collection.*
import fs2.kafka.producer.MkProducer
import org.apache.kafka.clients.consumer.ConsumerGroupMetadata
import org.apache.kafka.clients.producer.RecordMetadata
import org.apache.kafka.common.{Metric, MetricName}
/**
* Represents a producer of Kafka records specialized for 'read-process-write' streams, with the
* ability to atomically produce `ProducerRecord`s and commit corresponding [[CommittableOffset]]s
* using [[produce]].
*
* Records are wrapped in [[TransactionalProducerRecords]], which is a chunk of
* [[CommittableProducerRecord]] which wrap zero or more records together with a
* [[CommittableOffset]].
*/
abstract class TransactionalKafkaProducer[F[_], K, V] {
/**
* Produces the `ProducerRecord`s in the specified [[TransactionalProducerRecords]] in four
* steps: first a transaction is initialized, then the records are placed in the buffer of the
* producer, then the offsets of the records are sent to the transaction, and lastly the
* transaction is committed. If errors or cancellation occurs, the transaction is aborted. The
* returned effect succeeds if the whole transaction completes successfully.
*/
def produce(
records: TransactionalProducerRecords[F, K, V]
): F[ProducerResult[K, V]]
}
object TransactionalKafkaProducer {
/**
* [[TransactionalKafkaProducer.Metrics]] extends [[TransactionalKafkaProducer]] to provide
* access to the underlying producer metrics.
*/
abstract class Metrics[F[_], K, V] extends TransactionalKafkaProducer[F, K, V] {
/**
* Returns producer metrics.
*
* @see
* org.apache.kafka.clients.producer.KafkaProducer#metrics
*/
def metrics: F[Map[MetricName, Metric]]
}
/**
* [[TransactionalKafkaProducer.WithoutOffsets]] extends [[TransactionalKafkaProducer.Metrics]]
* to allow producing of records without corresponding upstream offsets.
*/
abstract class WithoutOffsets[F[_], K, V] extends Metrics[F, K, V] {
/**
* Produces the `ProducerRecord`s in the specified [[ProducerRecords]] in three steps: first a
* transaction is initialized, then the records are placed in the buffer of the producer, and
* lastly the transaction is committed. If errors or cancellation occurs, the transaction is
* aborted. The returned effect succeeds if the whole transaction completes successfully.
*/
def produceWithoutOffsets(records: ProducerRecords[K, V]): F[ProducerResult[K, V]]
}
/**
* Creates a new [[TransactionalKafkaProducer]] in the `Resource` context, using the specified
* [[TransactionalProducerSettings]]. Note that there is another version where `F[_]` is
* specified explicitly and the key and value type can be inferred, which allows you to use the
* following syntax.
*
* {{{
* TransactionalKafkaProducer.resource[F].using(settings)
* }}}
*/
def resource[F[_], K, V](
settings: TransactionalProducerSettings[F, K, V]
)(implicit
F: Async[F],
mk: MkProducer[F]
): Resource[F, TransactionalKafkaProducer.WithoutOffsets[F, K, V]] =
(
settings.producerSettings.keySerializer,
settings.producerSettings.valueSerializer,
WithTransactionalProducer(mk, settings)
).mapN { (keySerializer, valueSerializer, withProducer) =>
new TransactionalKafkaProducer.WithoutOffsets[F, K, V] {
override def produce(
records: TransactionalProducerRecords[F, K, V]
): F[ProducerResult[K, V]] =
produceTransactionWithOffsets(records)
private[this] def produceTransactionWithOffsets(
records: Chunk[CommittableProducerRecords[F, K, V]]
): F[Chunk[(ProducerRecord[K, V], RecordMetadata)]] =
if (records.isEmpty) F.pure(Chunk.empty)
else {
val batch =
CommittableOffsetBatch.fromFoldableMap(records)(_.offset)
val consumerGroupId =
if (batch.consumerGroupIdsMissing || batch.consumerGroupIds.size != 1)
F.raiseError(ConsumerGroupException(batch.consumerGroupIds))
else F.pure(batch.consumerGroupIds.head)
consumerGroupId.flatMap { groupId =>
val sendOffsets: (KafkaByteProducer, Blocking[F]) => F[Unit] = (producer, blocking) =>
blocking {
producer.sendOffsetsToTransaction(
batch.offsets.asJava,
new ConsumerGroupMetadata(groupId)
)
}
produceTransaction(records.flatMap(_.records), Some(sendOffsets))
}
}
override def produceWithoutOffsets(
records: ProducerRecords[K, V]
): F[ProducerResult[K, V]] =
produceTransaction(records, None)
private[this] def produceTransaction(
records: Chunk[ProducerRecord[K, V]],
sendOffsets: Option[(KafkaByteProducer, Blocking[F]) => F[Unit]]
): F[Chunk[(ProducerRecord[K, V], RecordMetadata)]] =
withProducer.exclusiveAccess { (producer, blocking) =>
blocking(producer.beginTransaction()).bracketCase { _ =>
val produce = records
.traverse(
KafkaProducer.produceRecord(keySerializer, valueSerializer, producer, blocking)
)
.flatMap(_.sequence)
sendOffsets.fold(produce)(f => produce.flatTap(_ => f(producer, blocking)))
} {
case (_, Outcome.Succeeded(_)) =>
blocking(producer.commitTransaction())
case (_, Outcome.Canceled() | Outcome.Errored(_)) =>
blocking(producer.abortTransaction())
}
}
override def metrics: F[Map[MetricName, Metric]] =
withProducer.blocking(_.metrics().asScala.toMap)
override def toString: String =
"TransactionalKafkaProducer$" + System.identityHashCode(this)
}
}
/**
* Creates a new [[TransactionalKafkaProducer]] in the `Stream` context, using the specified
* [[TransactionalProducerSettings]]. Note that there is another version where `F[_]` is
* specified explicitly and the key and value type can be inferred, which allows you to use the
* following syntax.
*
* {{{
* TransactionalKafkaProducer.stream[F].using(settings)
* }}}
*/
def stream[F[_], K, V](
settings: TransactionalProducerSettings[F, K, V]
)(implicit
F: Async[F],
mk: MkProducer[F]
): Stream[F, TransactionalKafkaProducer.WithoutOffsets[F, K, V]] =
Stream.resource(resource(settings)(F, mk))
def apply[F[_]]: TransactionalProducerPartiallyApplied[F] =
new TransactionalProducerPartiallyApplied
final private[kafka] class TransactionalProducerPartiallyApplied[F[_]](val dummy: Boolean = true)
extends AnyVal {
/**
* Alternative version of `resource` where the `F[_]` is specified explicitly, and where the
* key and value type can be inferred from the [[TransactionalProducerSettings]]. This allows
* you to use the following syntax.
*
* {{{
* KafkaProducer[F].resource(settings)
* }}}
*/
def resource[K, V](settings: TransactionalProducerSettings[F, K, V])(implicit
F: Async[F],
mk: MkProducer[F]
): Resource[F, TransactionalKafkaProducer.WithoutOffsets[F, K, V]] =
TransactionalKafkaProducer.resource(settings)(F, mk)
/**
* Alternative version of `stream` where the `F[_]` is specified explicitly, and where the key
* and value type can be inferred from the [[TransactionalProducerSettings]]. This allows you
* to use the following syntax.
*
* {{{
* KafkaProducer[F].stream(settings)
* }}}
*/
def stream[K, V](settings: TransactionalProducerSettings[F, K, V])(implicit
F: Async[F],
mk: MkProducer[F]
): Stream[F, TransactionalKafkaProducer.WithoutOffsets[F, K, V]] =
TransactionalKafkaProducer.stream(settings)(F, mk)
override def toString: String =
"TransactionalProducerPartiallyApplied$" + System.identityHashCode(this)
}
/*
* Prevents the default `MkProducer` instance from being implicitly available
* to code defined in this object, ensuring factory methods require an instance
* to be provided at the call site.
*/
@nowarn("msg=never used")
implicit private def mkAmbig1[F[_]]: MkProducer[F] =
throw new AssertionError("should not be used")
@nowarn("msg=never used")
implicit private def mkAmbig2[F[_]]: MkProducer[F] =
throw new AssertionError("should not be used")
}
