Github user aljoscha commented on a diff in the pull request: https://github.com/apache/flink/pull/1929#discussion_r61228546 --- Diff: flink-streaming-java/src/main/java/org/apache/flink/streaming/api/functions/source/FileSplitMonitoringFunction.java --- @@ -0,0 +1,234 @@ +/* + * 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.streaming.api.functions.source; + +import org.apache.flink.api.common.io.FileInputFormat; +import org.apache.flink.configuration.Configuration; +import org.apache.flink.core.fs.FileInputSplit; +import org.apache.flink.core.fs.FileStatus; +import org.apache.flink.core.fs.FileSystem; +import org.apache.flink.core.fs.Path; +import org.apache.flink.runtime.JobException; +import org.apache.flink.util.Preconditions; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import java.io.IOException; +import java.net.URI; +import java.util.ArrayList; +import java.util.HashSet; +import java.util.List; +import java.util.Set; + +/** + * This is the single (non-parallel) task which monitors a user-provided path and assigns splits + * to downstream tasks for further reading and processing. Which splits will be further processed + * depends on the user-provided {@link FileSplitMonitoringFunction.WatchType}. + */ +public class FileSplitMonitoringFunction<OUT> + extends RichSourceFunction<FileInputSplit> { + + private static final long serialVersionUID = 1L; + + private static final Logger LOG = LoggerFactory.getLogger(FileSplitMonitoringFunction.class); + + /** + * Specifies when computation will be triggered. + */ + public enum WatchType { + REPROCESS_WITH_APPENDED // Reprocesses the whole file when new data is appended. + } + + /** The path to monitor. */ + private final String path; + + /** The default parallelism for the job, as this is going to be the parallelism of the downstream readers. */ + private final int readerParallelism; + + /** The {@link FileInputFormat} to be read. */ + private FileInputFormat<OUT> format; + + /** How often to monitor the state of the directory for new data. */ + private final long interval; + + /** Which new data to process (see {@link WatchType}. */ + private final WatchType watchType; + + private long globalModificationTime; + + private FilePathFilter pathFilter; + + private volatile boolean isRunning = false; + + private Configuration configuration; + + public FileSplitMonitoringFunction( + FileInputFormat<OUT> format, String path, Configuration configuration, + WatchType watchType, int readerParallelism, long interval) { + + this(format, path, configuration, FilePathFilter.DefaultFilter.getInstance(), watchType, readerParallelism, interval); + } + + public FileSplitMonitoringFunction( + FileInputFormat<OUT> format, String path, Configuration configuration, + FilePathFilter filter, WatchType watchType, int readerParallelism, long interval) { + + this.format = Preconditions.checkNotNull(format); + this.path = Preconditions.checkNotNull(path); + this.configuration = Preconditions.checkNotNull(configuration); + + Preconditions.checkArgument(interval >= 100, + "The specified monitoring interval is smaller than the minimum allowed one (100 ms)."); + this.interval = interval; + + this.watchType = watchType; + + this.pathFilter = Preconditions.checkNotNull(filter); + + this.readerParallelism = Math.max(readerParallelism, 1); + this.globalModificationTime = Long.MIN_VALUE; + } + + @Override + @SuppressWarnings("unchecked") + public void open(Configuration parameters) throws Exception { + super.open(parameters); + format.configure(this.configuration); + } + + /** + * Creates the input splits for the path to be assigned to the downstream tasks. + * Those are going to read their contents for further processing. + */ + private FileInputSplit[] createInputSplits() throws JobException { + FileInputSplit[] inputSplits; + try { + inputSplits = format.createInputSplits(readerParallelism); + } catch (Throwable t) { + throw new JobException("Creating the input splits caused an error: " + t.getMessage(), t); + } + return inputSplits; + } + + @Override + public void run(SourceFunction.SourceContext<FileInputSplit> context) throws Exception { + if(!isRunning) { --- End diff -- We should not change this here. I think it can be problematic with race conditions. Just initializing `isRunning` to `true` when declaring the field should do the trick. (This is also how the Kafka Consumer does it and I think as this one as the reference consumer since it's most widely used.)
--- 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. ---