martongreber commented on code in PR #8: URL: https://github.com/apache/flink-connector-kudu/pull/8#discussion_r1979989809
########## flink-connector-kudu/src/main/java/org/apache/flink/connector/kudu/source/enumerator/KuduSourceEnumerator.java: ########## @@ -0,0 +1,264 @@ +/* + * 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.kudu.source.enumerator; + +import org.apache.flink.api.connector.source.Boundedness; +import org.apache.flink.api.connector.source.SourceEvent; +import org.apache.flink.api.connector.source.SplitEnumerator; +import org.apache.flink.api.connector.source.SplitEnumeratorContext; +import org.apache.flink.connector.kudu.connector.KuduTableInfo; +import org.apache.flink.connector.kudu.connector.reader.KuduReaderConfig; +import org.apache.flink.connector.kudu.source.split.KuduSourceSplit; +import org.apache.flink.connector.kudu.source.split.SplitFinishedEvent; +import org.apache.flink.connector.kudu.source.utils.KuduSplitGenerator; +import org.apache.flink.connector.kudu.source.utils.KuduSplitRetriever; + +import org.apache.kudu.util.HybridTimeUtil; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import javax.annotation.Nullable; + +import java.io.IOException; +import java.time.Duration; +import java.util.ArrayList; +import java.util.Iterator; +import java.util.List; +import java.util.Objects; +import java.util.concurrent.TimeUnit; +import java.util.function.Supplier; + +import static org.apache.flink.util.Preconditions.checkNotNull; + +/** + * The Kudu source enumerator is responsible for discovering and assigning splits. + * + * <ul> + * <li>If {@code Boundedness.BOUNDED} is set, the enumerator generates splits corresponding to the + * current snapshot time and emits records only for this bounded set. + * <li>If {@code Boundedness.CONTINUOUS_UNBOUNDED} is set, the enumerator follows a CDC-like + * approach as described below. + * </ul> + * + * <p>To provide CDC-like functionality, the enumeration works as follows: Initially, we perform a + * snapshot read of the table and mark the snapshot time as t0. From that point onward, we perform + * differential scans in the time intervals t0 - t1, t1 - t2, and so on. + * + * <p>This approach means that new splits can only be enumerated once the current time range is + * fully processed. + * + * <p>The process is controlled as follows: + * + * <ul> + * <li>Once a set of splits is enumerated for a time range, we track: + * <ul> + * <li><b>Unassigned splits</b>: Discovered but not yet assigned to readers. + * <li><b>Pending splits</b>: Assigned but not yet fully processed. + * </ul> + * <li>A new set of splits is generated only when there are no remaining unassigned or pending + * splits. + * </ul> + */ +public class KuduSourceEnumerator + implements SplitEnumerator<KuduSourceSplit, KuduSourceEnumeratorState> { + private static final Logger LOG = LoggerFactory.getLogger(KuduSourceEnumerator.class); + + private final SplitEnumeratorContext<KuduSourceSplit> context; + private final List<Integer> readersAwaitingSplit; + private final KuduSplitGenerator splitGenerator; + private final Boundedness boundedness; + private final Duration discoveryInterval; + + private long lastEndTimestamp; + private final List<KuduSourceSplit> unassigned; + private final List<KuduSourceSplit> pending; + + public KuduSourceEnumerator( + KuduTableInfo tableInfo, + KuduReaderConfig readerConfig, + Boundedness boundedness, + Duration discoveryInterval, + SplitEnumeratorContext<KuduSourceSplit> context) { + this( + tableInfo, + readerConfig, + boundedness, + discoveryInterval, + context, + KuduSourceEnumeratorState.empty()); + } + + public KuduSourceEnumerator( + KuduTableInfo tableInfo, + KuduReaderConfig readerConfig, + Boundedness boundedness, + Duration discoveryInterval, + SplitEnumeratorContext<KuduSourceSplit> context, + KuduSourceEnumeratorState enumState) { + this.boundedness = boundedness; + this.discoveryInterval = discoveryInterval; + this.context = checkNotNull(context); + this.readersAwaitingSplit = new ArrayList<>(); + this.unassigned = enumState.getUnassigned(); + this.pending = enumState.getPending(); + this.splitGenerator = new KuduSplitGenerator(readerConfig, tableInfo); + this.lastEndTimestamp = enumState.getLastEndTimestamp(); + } + + @Override + public void start() { + if (boundedness == Boundedness.CONTINUOUS_UNBOUNDED && Objects.nonNull(discoveryInterval)) { + context.callAsync( + () -> enumerateNewSplits(this::shouldEnumerateNewSplits), + this::assignSplits, + 0, + discoveryInterval.toMillis()); + } else if (boundedness.equals(Boundedness.BOUNDED)) { 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. To unsubscribe, e-mail: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
