Package org.apache.beam.runners.spark.structuredstreaming.translation.batch.functions
Internal implementation of the Beam runner for Apache Spark.
-
Interface Summary Interface Description SideInputValues<T> SideInputValuesserves as a Kryo serializable container that contains a materialized view of side inputs.SideInputValues.Loader<T> Factory function for loadSideInputValuesfrom aDataset. -
Class Summary Class Description CachedSideInputReader SideInputReader that caches results for costlyMaterializations.GroupAlsoByWindowViaOutputBufferFn<K,InputT,W extends org.apache.beam.sdk.transforms.windowing.BoundedWindow> A FlatMap function that groups by windows in batch mode usingReduceFnRunner.NoOpStepContext AStepContextfor Spark Batch Runner execution.SideInputValues.BaseSideInputValues<BinaryT,ValuesT extends @NonNull java.lang.Object,T> SideInputValues.ByWindow<T> GeneralSideInputValuesforBoundedWindowsin two possible states.SideInputValues.Global<T> SpecializedSideInputValuesfor use with theGlobalWindowin two possible states.SparkSideInputReader SideInputReader using broadcastedSideInputValues.