dlg99 commented on a change in pull request #9927: URL: https://github.com/apache/pulsar/pull/9927#discussion_r626187058
########## File path: pulsar-io/kafka-connect-adaptor/src/main/java/org/apache/pulsar/io/kafka/connect/PulsarKafkaSinkTaskContext.java ########## @@ -0,0 +1,212 @@ +/** + * 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.pulsar.io.kafka.connect; + +import com.google.common.collect.Lists; +import com.google.common.collect.Maps; +import lombok.SneakyThrows; +import lombok.extern.slf4j.Slf4j; +import org.apache.kafka.clients.consumer.OffsetAndMetadata; +import org.apache.kafka.common.TopicPartition; +import org.apache.kafka.connect.sink.SinkTaskContext; +import org.apache.kafka.connect.storage.OffsetBackingStore; + +import java.nio.ByteBuffer; +import java.util.Collection; +import java.util.List; +import java.util.Map; +import java.util.Optional; +import java.util.Set; +import java.util.concurrent.CompletableFuture; +import java.util.concurrent.ConcurrentHashMap; +import java.util.concurrent.atomic.AtomicBoolean; +import java.util.concurrent.atomic.AtomicLong; +import java.util.function.Consumer; + +import static java.nio.charset.StandardCharsets.UTF_8; +import static org.apache.pulsar.io.kafka.connect.PulsarKafkaWorkerConfig.TOPIC_NAMESPACE_CONFIG; + +@Slf4j +public class PulsarKafkaSinkTaskContext implements SinkTaskContext { + + private final Map<String, String> config; + + private final OffsetBackingStore offsetStore; + private final String topicNamespace; + private final Consumer<Collection<TopicPartition>> onPartitionChange; + private final AtomicBoolean runRepartition = new AtomicBoolean(false); + + private final ConcurrentHashMap<TopicPartition, AtomicLong> currentOffsets = new ConcurrentHashMap<>(); + + public PulsarKafkaSinkTaskContext(Map<String, String> config, + Consumer<Collection<TopicPartition>> onPartitionChange) { + this.config = config; + + offsetStore = new PulsarOffsetBackingStore(); + PulsarKafkaWorkerConfig pulsarKafkaWorkerConfig = new PulsarKafkaWorkerConfig(config); + offsetStore.configure(pulsarKafkaWorkerConfig); + offsetStore.start(); + + this.onPartitionChange = onPartitionChange; + this.topicNamespace = pulsarKafkaWorkerConfig.getString(TOPIC_NAMESPACE_CONFIG); + } + + public void close() { + offsetStore.stop(); + } + + @Override + public Map<String, String> configs() { + return config; + } + + public AtomicLong currentOffset(String topic, int partition) { + return currentOffset(new TopicPartition(topic, partition)); + } + + public AtomicLong currentOffset(TopicPartition topicPartition) { + AtomicLong offset = currentOffsets.computeIfAbsent(topicPartition, kv -> { + List<ByteBuffer> req = Lists.newLinkedList(); + ByteBuffer key = topicPartitionAsKey(topicPartition); + req.add(key); + CompletableFuture<Long> offsetFuture = new CompletableFuture<>(); + offsetStore.get(req, (Throwable ex, Map<ByteBuffer, ByteBuffer> result) -> { + if (ex == null) { + if (result != null && result.size() != 0) { + Optional<ByteBuffer> val = result.entrySet().stream() + .filter(entry -> entry.getKey().equals(key)) + .findFirst().map(entry -> entry.getValue()); + if (val.isPresent()) { + long received = val.get().getLong(); + if (log.isDebugEnabled()) { + log.debug("read initial offset for {} == {}", topicPartition, received); + } + offsetFuture.complete(received); + return; + } + } + offsetFuture.complete(-1L); + } else { + offsetFuture.completeExceptionally(ex); + } + }); + + runRepartition.set(true); + try { + return new AtomicLong(offsetFuture.get()); + } catch (Exception e) { + log.error("error getting initial state of " + topicPartition.toString(), e); + return new AtomicLong(-1L); + } + }); + if (runRepartition.compareAndSet(true, false)) { + onPartitionChange.accept(currentOffsets.keySet()); + } + return offset; + } + + public Map<TopicPartition, OffsetAndMetadata> currentOffsets() { + + Map<TopicPartition, OffsetAndMetadata> snapshot = Maps.newHashMapWithExpectedSize(currentOffsets.size()); + currentOffsets.forEach((topicPartition, offset) -> { + if (offset.get() > 0) { + snapshot.put(topicPartition, + new OffsetAndMetadata(offset.get(), Optional.empty(), null)); + } + }); + return snapshot; + } + + private ByteBuffer topicPartitionAsKey(TopicPartition topicPartition) { + return ByteBuffer.wrap((topicNamespace + "/" + topicPartition.toString()).getBytes(UTF_8)); + + } + + private void fillOffsetMap(Map<ByteBuffer, ByteBuffer> offsetMap, TopicPartition topicPartition, long l) { + ByteBuffer key = topicPartitionAsKey(topicPartition); + ByteBuffer value = ByteBuffer.allocate(Long.BYTES); + value.putLong(l); + value.flip(); + offsetMap.put(key, value); + } + + @SneakyThrows + @Override + public void offset(Map<TopicPartition, Long> map) { Review comment: Looking at this closer, to actually reset the consumer (and `pause()`/`resume()`) properly we'll need a way for the sink to communicate to the source. I'll think about it/draft the implementation but I think it can be done as a follow up. Not every kafka sink actually uses it AFAICT -- 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]
