Yanikovic commented on a change in pull request #15140: URL: https://github.com/apache/flink/pull/15140#discussion_r594623672
########## File path: flink-connectors/flink-connector-rabbitmq2/src/main/java/org/apache/flink/connector/rabbitmq2/sink/RabbitMQSink.java ########## @@ -0,0 +1,229 @@ +/* + * 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.flink.connector.rabbitmq2.sink; + +import org.apache.flink.api.common.serialization.DeserializationSchema; +import org.apache.flink.api.common.serialization.SerializationSchema; +import org.apache.flink.api.connector.sink.Committer; +import org.apache.flink.api.connector.sink.GlobalCommitter; +import org.apache.flink.api.connector.sink.Sink; +import org.apache.flink.api.connector.sink.SinkWriter; +import org.apache.flink.connector.rabbitmq2.ConsistencyMode; +import org.apache.flink.connector.rabbitmq2.RabbitMQConnectionConfig; +import org.apache.flink.connector.rabbitmq2.sink.state.RabbitMQSinkWriterState; +import org.apache.flink.connector.rabbitmq2.sink.state.RabbitMQSinkWriterStateSerializer; +import org.apache.flink.connector.rabbitmq2.sink.writer.RabbitMQSinkWriterBase; +import org.apache.flink.connector.rabbitmq2.sink.writer.specalized.RabbitMQSinkWriterAtLeastOnce; +import org.apache.flink.connector.rabbitmq2.sink.writer.specalized.RabbitMQSinkWriterAtMostOnce; +import org.apache.flink.connector.rabbitmq2.sink.writer.specalized.RabbitMQSinkWriterExactlyOnce; +import org.apache.flink.core.io.SimpleVersionedSerializer; +import org.apache.flink.util.Preconditions; + +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import javax.annotation.Nullable; + +import java.util.List; +import java.util.Optional; + +/** + * RabbitMQ sink (publisher) that publishes messages from upstream flink tasks to a RabbitMQ queue. + * It provides at-most-once, at-least-once and exactly-once processing semantics. For at-least-once + * and exactly-once, checkpointing needs to be enabled. The sink operates as a StreamingSink and + * thus works in a streaming fashion. + * + * <pre>{@code + * RabbitMQSink + * .builder() + * .setConnectionConfig(connectionConfig) + * .setQueueName("queue") + * .setSerializationSchema(new SimpleStringSchema()) + * .setConsistencyMode(ConsistencyMode.AT_LEAST_ONCE) + * .setMinimalResendInterval(10L) + * .build(); + * }</pre> + * + * <p>When creating the sink a {@code connectionConfig} must be specified via {@link + * RabbitMQConnectionConfig}. It contains required information for the RabbitMQ java client to + * connect to the RabbitMQ server. A minimum configuration contains a (virtual) host, a username, a + * password and a port. Besides that, the {@code queueName} to publish to and a {@link + * SerializationSchema} for the sink input type is required. {@code publishOptions} can be added to + * route messages in RabbitMQ. + * + * <p>If at-least-once is required, an optional number of {@code maxRetry} attempts can be specified + * until a failure is triggered. Generally, messages are buffered until an acknowledgement arrives + * because delivery needs to be guaranteed. On each checkpoint, all unacknowledged messages will be + * resent to RabbitMQ. If the checkpointing interval is set low or a high frequency of resending is + * not desired, the {@code minimalResendIntervalMilliseconds} can be specified to prevent the sink + * from resending data that has just arrived. In case of a failure, all unacknowledged messages can + * be restored and resend. + * + * <p>In the case of exactly-once a transactional RabbitMQ channel is used to achieve that all + * messages within a checkpoint are delivered once and only once. All messages that arrive in a + * checkpoint interval are buffered and sent to RabbitMQ in a single transaction when the checkpoint + * is triggered. If the transaction fails, all messages that were a part of the transaction are put + * back into the buffer and a resend is issued in the next checkpoint. + * + * <p>Keep in mind that the transactional channels are heavyweight and performance will drop. Under Review comment: Done ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
