com.google.cloud.dataflow.sdk.io

Interface UnboundedSource.UnboundedReader<OutputT>

All Superinterfaces:

AutoCloseable, Source.Reader<OutputT>

Enclosing class:

UnboundedSource<OutputT,CheckpointMarkT extends UnboundedSource.CheckpointMark>

@Experimental(value=SOURCE_SINK)
public static interface UnboundedSource.UnboundedReader<OutputT>
extends Source.Reader<OutputT>

A Reader that reads an unbounded amount of input.

A given UnboundedReader object will only be accessed by a single thread at once.

Method Summary

Modifier and Type	Method and Description
boolean	advance() Advances the reader to the next valid record.
UnboundedSource.CheckpointMark	getCheckpointMark() Returns a UnboundedSource.CheckpointMark representing the progress of this UnboundedReader.
byte[]	getCurrentRecordId() Returns a unique identifier for the current record.
UnboundedSource<OutputT,?>	getCurrentSource() Returns the UnboundedSource that created this reader.
org.joda.time.Instant	getWatermark() Returns a lower bound on timestamps of future elements read by this reader.
boolean	start() Initializes the reader and advances the reader to the first record.

Methods inherited from interface com.google.cloud.dataflow.sdk.io.Source.Reader

close, getCurrent, getCurrentTimestamp

Method Detail

start

boolean start()
       throws IOException

Initializes the reader and advances the reader to the first record.

This method should be called exactly once. The invocation should occur prior to calling advance() or Source.Reader.getCurrent(). This method may perform expensive operations that are needed to initialize the reader.

Returns true if a record was read, false if there is no more input currently available. Future calls to advance() may return true once more data is available. Regardless of the return value of start, start will not be called again on the same UnboundedReader object; it will only be called again when a new reader object is constructed for the same source, e.g. on recovery.

Specified by:

start in interface Source.Reader<OutputT>

Returns:

true if a record was read, false if there is no more input available.

Throws:

IOException

advance

boolean advance()
         throws IOException

Advances the reader to the next valid record.

Returns true if a record was read, false if there is no more input available. Future calls to advance() may return true once more data is available.

Specified by:

advance in interface Source.Reader<OutputT>

Returns:

true if a record was read, false if there is no more input available.

Throws:

IOException

getCurrentRecordId

byte[] getCurrentRecordId()
                   throws NoSuchElementException

Returns a unique identifier for the current record. This should be the same for each instance of the same logical record read from the underlying data source.

For example, this could be a hash of the record contents, or a logical ID present in the record. If this is generated as a hash of the record contents, it should be at least 16 bytes (128 bits) to avoid collisions.

This method has the same restrictions on when it can be called as Source.Reader.getCurrent() and Source.Reader.getCurrentTimestamp().

Note: this is not yet supported by the DataflowPipelineRunner, and it will be ignored.

Throws:

NoSuchElementException - if the reader is at the beginning of the input and start() or advance() wasn't called, or if the last start() or advance() returned false.

getWatermark

org.joda.time.Instant getWatermark()

Returns a lower bound on timestamps of future elements read by this reader.

This can be approximate. If records are read that violate this guarantee, they will be considered late, which will affect how they will be processed. See Window for more information on late data and how to handle it.

This bound should be as tight as possible. Downstream windows will not be able to close until this watermark passes the end of the window.

For example, a source may know that the records it reads will be in timestamp order. In this case, the watermark can be the timestamp of the last record read minus one. For a source that does not have natural timestamps, timestamps can be set to the time of reading, in which case the watermark is the current clock time.

See Window and Trigger for more information on timestamps and watermarks.

May be called after advance() or start() has returned false, but not before start() has been called.

getCheckpointMark

UnboundedSource.CheckpointMark getCheckpointMark()

Returns a UnboundedSource.CheckpointMark representing the progress of this UnboundedReader.

The elements read up until this is called will be processed together as a bundle. Once the result of this processing has been durably committed, UnboundedSource.CheckpointMark.finalizeCheckpoint() will be called on the UnboundedSource.CheckpointMark object.

The returned object should not be modified.

May be called after advance() or start() has returned false, but not before start() has been called.

getCurrentSource

UnboundedSource<OutputT,?> getCurrentSource()

Returns the UnboundedSource that created this reader. This will not change over the life of the reader.

Specified by:

getCurrentSource in interface Source.Reader<OutputT>