com.google.cloud.dataflow.sdk.util.common.worker

Class BatchingShuffleEntryReader

java.lang.Object

com.google.cloud.dataflow.sdk.util.common.worker.BatchingShuffleEntryReader

All Implemented Interfaces:

ShuffleEntryReader

@NotThreadSafe
public final class BatchingShuffleEntryReader
extends Object
implements ShuffleEntryReader

BatchingShuffleEntryReader provides a mechanism for reading entries from a shuffle dataset.

Constructor Summary

Constructors

Constructor and Description
BatchingShuffleEntryReader(ShuffleBatchReader batchReader) Constructs a BatchingShuffleEntryReader.

Method Summary

Modifier and Type	Method and Description
Reiterator<ShuffleEntry>	read(ShufflePosition startPosition, ShufflePosition endPosition) Returns an iterator that reads a range of entries from a shuffle dataset.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

BatchingShuffleEntryReader

public BatchingShuffleEntryReader(ShuffleBatchReader batchReader)

Constructs a BatchingShuffleEntryReader.

Parameters:

batchReader - supplies the underlying ShuffleBatchReader to read batches of entries from

Method Detail

read

public Reiterator<ShuffleEntry> read(@Nullable
                                     ShufflePosition startPosition,
                                     @Nullable
                                     ShufflePosition endPosition)

Description copied from interface: ShuffleEntryReader

Returns an iterator that reads a range of entries from a shuffle dataset.

Specified by:

read in interface ShuffleEntryReader

Parameters:

startPosition - encodes the initial key from where to read. This parameter may be null, indicating that the read should start with the first key in the dataset.

endPosition - encodes the key "just past" the end of the range to be read; keys up to endPosition will be returned, but keys equal to or greater than endPosition will not. This parameter may be null, indicating that the read should end just past the last key in the dataset (that is, the last key in the dataset will be included in the read, as long as that key is greater than or equal to startPosition).

Returns:

a Reiterator over the requested range of entries.