T - Type of the elements in this Stream@PublicEvolving public class IterativeStream<T> extends SingleOutputStreamOperator<T>
DataStream.| Modifier and Type | Class and Description |
|---|---|
static class |
IterativeStream.ConnectedIterativeStreams<I,F>
The
IterativeStream.ConnectedIterativeStreams represent a start of an
iterative part of a streaming program, where the original input of the
iteration and the feedback of the iteration are connected as in a
ConnectedStreams. |
environment, transformation| Modifier | Constructor and Description |
|---|---|
protected |
IterativeStream(DataStream<T> dataStream,
long maxWaitTime) |
| Modifier and Type | Method and Description |
|---|---|
DataStream<T> |
closeWith(DataStream<T> feedbackStream)
Closes the iteration.
|
<F> IterativeStream.ConnectedIterativeStreams<T,F> |
withFeedbackType(Class<F> feedbackTypeClass)
Changes the feedback type of the iteration and allows the user to apply
co-transformations on the input and feedback stream, as in a
ConnectedStreams. |
<F> IterativeStream.ConnectedIterativeStreams<T,F> |
withFeedbackType(String feedbackTypeString)
Changes the feedback type of the iteration and allows the user to apply
co-transformations on the input and feedback stream, as in a
ConnectedStreams. |
<F> IterativeStream.ConnectedIterativeStreams<T,F> |
withFeedbackType(org.apache.flink.api.common.typeinfo.TypeInformation<F> feedbackType)
Changes the feedback type of the iteration and allows the user to apply
co-transformations on the input and feedback stream, as in a
ConnectedStreams. |
disableChaining, getName, name, returns, returns, returns, returns, setBufferTimeout, setConnectionType, setParallelism, slotSharingGroup, startNewChain, uidaddSink, assignTimestamps, assignTimestampsAndWatermarks, assignTimestampsAndWatermarks, broadcast, clean, coGroup, connect, countWindowAll, countWindowAll, filter, flatMap, forward, getExecutionConfig, getExecutionEnvironment, getId, getParallelism, getTransformation, getType, global, iterate, iterate, join, keyBy, keyBy, keyBy, map, partitionCustom, partitionCustom, partitionCustom, print, printToErr, project, rebalance, rescale, shuffle, split, timeWindowAll, timeWindowAll, transform, union, windowAll, writeAsCsv, writeAsCsv, writeAsCsv, writeAsText, writeAsText, writeToSocket, writeUsingOutputFormatprotected IterativeStream(DataStream<T> dataStream, long maxWaitTime)
public DataStream<T> closeWith(DataStream<T> feedbackStream)
A common usage pattern for streaming iterations is to use output
splitting to send a part of the closing data stream to the head. Refer to
DataStream.split(org.apache.flink.streaming.api.collector.selector.OutputSelector)
for more information.
feedbackStream - DataStream that will be used as input to the iteration
head.public <F> IterativeStream.ConnectedIterativeStreams<T,F> withFeedbackType(String feedbackTypeString)
ConnectedStreams.
For type safety the user needs to define the feedback type
feedbackTypeString - String describing the type information of the feedback stream.IterativeStream.ConnectedIterativeStreams.public <F> IterativeStream.ConnectedIterativeStreams<T,F> withFeedbackType(Class<F> feedbackTypeClass)
ConnectedStreams.
For type safety the user needs to define the feedback type
feedbackTypeClass - Class of the elements in the feedback stream.IterativeStream.ConnectedIterativeStreams.public <F> IterativeStream.ConnectedIterativeStreams<T,F> withFeedbackType(org.apache.flink.api.common.typeinfo.TypeInformation<F> feedbackType)
ConnectedStreams.
For type safety the user needs to define the feedback type
feedbackType - The type information of the feedback stream.IterativeStream.ConnectedIterativeStreams.Copyright © 2014–2016 The Apache Software Foundation. All rights reserved.