JozoVilcek commented on code in PR #24837: URL: https://github.com/apache/beam/pull/24837#discussion_r1062662250
########## runners/spark/src/main/java/org/apache/beam/runners/spark/translation/SparkInputDataProcessor.java: ########## @@ -0,0 +1,394 @@ +/* + * 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.beam.runners.spark.translation; + +import static org.apache.beam.vendor.guava.v26_0_jre.com.google.common.base.Preconditions.checkArgument; + +import java.util.Collections; +import java.util.Iterator; +import java.util.Map; +import java.util.concurrent.ExecutorService; +import java.util.concurrent.Executors; +import java.util.concurrent.Future; +import java.util.concurrent.LinkedBlockingQueue; +import java.util.concurrent.TimeUnit; +import org.apache.beam.runners.core.DoFnRunners.OutputManager; +import org.apache.beam.runners.core.StateNamespace; +import org.apache.beam.runners.core.StateNamespaces; +import org.apache.beam.runners.core.TimerInternals; +import org.apache.beam.sdk.transforms.reflect.DoFnInvokers; +import org.apache.beam.sdk.transforms.windowing.BoundedWindow; +import org.apache.beam.sdk.util.WindowedValue; +import org.apache.beam.sdk.values.TupleTag; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.base.Function; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.collect.AbstractIterator; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.collect.Iterators; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.collect.LinkedListMultimap; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.collect.Multimap; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.util.concurrent.ThreadFactoryBuilder; +import org.checkerframework.checker.nullness.qual.Nullable; +import scala.Tuple2; + +/** Processes Spark partitions using Beam's {@link org.apache.beam.runners.core.DoFnRunner}. */ +interface SparkInputDataProcessor<FnInputT, FnOutputT, OutputT> { + + /** + * @return {@link OutputManager} to be used by {@link org.apache.beam.runners.core.DoFnRunner} for + * emitting processing results + */ + OutputManager getOutputManager(); + + /** + * Processes input partition data and return results as {@link Iterable}. + * + * @param input input partition iterator + * @param ctx current processing context + */ + <K> Iterable<OutputT> process( + Iterator<WindowedValue<FnInputT>> input, SparkProcessContext<K, FnInputT, FnOutputT> ctx); + + /** + * Creates a synchronous {@link SparkInputDataProcessor} which does processing in calling thread. + * It is doing so by processing fully input element and then iterating over the output retrieved + * from that processing. The result of processing one element fit into memory. + */ + static <FnInputT, FnOutputT> + SparkInputDataProcessor<FnInputT, FnOutputT, Tuple2<TupleTag<?>, WindowedValue<?>>> + createSync() { + return new SyncSparkInputDataProcessor<>(); + } + + /** + * Create and asynchronous {@link SparkInputDataProcessor} which does process input elements in + * separate thread and observes produced outputs asynchronously. This does not require results of + * processing one element to fit into the memory. + */ + static <FnInputT, FnOutputT> + SparkInputDataProcessor<FnInputT, FnOutputT, Tuple2<TupleTag<?>, WindowedValue<?>>> + createAsync() { + return new AsyncSparkInputDataProcessor<>(); + } +} + +class SyncSparkInputDataProcessor<FnInputT, FnOutputT> + implements SparkInputDataProcessor<FnInputT, FnOutputT, Tuple2<TupleTag<?>, WindowedValue<?>>> { + + private final UnboundedDoFnOutputManager outputManager = new UnboundedDoFnOutputManager(); + + @Override + public OutputManager getOutputManager() { + return outputManager; + } + + @Override + public <K> Iterable<Tuple2<TupleTag<?>, WindowedValue<?>>> process( + Iterator<WindowedValue<FnInputT>> input, SparkProcessContext<K, FnInputT, FnOutputT> ctx) { + if (!input.hasNext()) { + return Collections.emptyList(); + } + return () -> new SyncInOutIterator<>(input, ctx); + } + + private static class UnboundedDoFnOutputManager + implements OutputManager, Iterable<Tuple2<TupleTag<?>, WindowedValue<?>>> { + + private final Multimap<TupleTag<?>, WindowedValue<?>> outputs = LinkedListMultimap.create(); Review Comment: Addressed by next commits ########## runners/spark/src/main/java/org/apache/beam/runners/spark/translation/SparkInputDataProcessor.java: ########## @@ -0,0 +1,394 @@ +/* + * 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.beam.runners.spark.translation; + +import static org.apache.beam.vendor.guava.v26_0_jre.com.google.common.base.Preconditions.checkArgument; + +import java.util.Collections; +import java.util.Iterator; +import java.util.Map; +import java.util.concurrent.ExecutorService; +import java.util.concurrent.Executors; +import java.util.concurrent.Future; +import java.util.concurrent.LinkedBlockingQueue; +import java.util.concurrent.TimeUnit; +import org.apache.beam.runners.core.DoFnRunners.OutputManager; +import org.apache.beam.runners.core.StateNamespace; +import org.apache.beam.runners.core.StateNamespaces; +import org.apache.beam.runners.core.TimerInternals; +import org.apache.beam.sdk.transforms.reflect.DoFnInvokers; +import org.apache.beam.sdk.transforms.windowing.BoundedWindow; +import org.apache.beam.sdk.util.WindowedValue; +import org.apache.beam.sdk.values.TupleTag; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.base.Function; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.collect.AbstractIterator; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.collect.Iterators; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.collect.LinkedListMultimap; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.collect.Multimap; +import org.apache.beam.vendor.guava.v26_0_jre.com.google.common.util.concurrent.ThreadFactoryBuilder; +import org.checkerframework.checker.nullness.qual.Nullable; +import scala.Tuple2; + +/** Processes Spark partitions using Beam's {@link org.apache.beam.runners.core.DoFnRunner}. */ +interface SparkInputDataProcessor<FnInputT, FnOutputT, OutputT> { + + /** + * @return {@link OutputManager} to be used by {@link org.apache.beam.runners.core.DoFnRunner} for + * emitting processing results + */ + OutputManager getOutputManager(); + + /** + * Processes input partition data and return results as {@link Iterable}. + * + * @param input input partition iterator + * @param ctx current processing context + */ + <K> Iterable<OutputT> process( Review Comment: Addressed by next commits -- 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: github-unsubscr...@beam.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
