Package org.apache.kafka.streams.processor.internals

Class GlobalProcessorContextImpl

java.lang.Object

org.apache.kafka.streams.processor.internals.AbstractProcessorContext

org.apache.kafka.streams.processor.internals.GlobalProcessorContextImpl

All Implemented Interfaces:

InternalProcessorContext, ProcessorContext

public class GlobalProcessorContextImpl
extends AbstractProcessorContext

Field Summary

Fields inherited from class org.apache.kafka.streams.processor.internals.AbstractProcessorContext

cache, currentNode, NONEXIST_TOPIC, recordContext

Fields inherited from interface org.apache.kafka.streams.processor.internals.InternalProcessorContext

BYTEARRAY_VALUE_SERIALIZER, BYTES_KEY_SERIALIZER

Constructor Summary

Constructors

Constructor	Description
GlobalProcessorContextImpl(StreamsConfig config, org.apache.kafka.streams.processor.internals.StateManager stateMgr, StreamsMetricsImpl metrics, ThreadCache cache)

Method Summary

Modifier and Type	Method	Description
void	commit()	Requests a commit.
<K,V> void	forward(K key, V value)	Forwards a key/value pair to all downstream processors.
<K,V> void	forward(K key, V value, int childIndex)	Deprecated.
<K,V> void	forward(K key, V value, java.lang.String childName)	Deprecated.
<K,V> void	forward(K key, V value, To to)	No-op.
StateStore	getStateStore(java.lang.String name)	Get the state store given the store name.
void	logChange(java.lang.String storeName, org.apache.kafka.common.utils.Bytes key, byte[] value, long timestamp)
void	registerCacheFlushListener(java.lang.String namespace, ThreadCache.DirtyEntryFlushListener listener)	Register a dirty entry flush listener for a particular namespace
Cancellable	schedule(long interval, PunctuationType type, Punctuator callback)	Deprecated.
Cancellable	schedule(java.time.Duration interval, PunctuationType type, Punctuator callback)	Schedules a periodic operation for processors.
void	transitionToActive(StreamTask streamTask, RecordCollector recordCollector, ThreadCache newCache)	Transition to active task and register a new task and cache to this processor context
void	transitionToStandby(ThreadCache newCache)	Transition to standby task and register a dummy cache to this processor context

Methods inherited from class org.apache.kafka.streams.processor.internals.AbstractProcessorContext

appConfigs, appConfigsWithPrefix, applicationId, applicationInternalStream, cache, currentNode, currentSystemTimeMs, headers, initialize, keySerde, metrics, offset, partition, recordContext, register, setCurrentNode, setRecordContext, setSystemTimeMs, stateDir, taskId, taskType, timestamp, topic, uninitialize, valueSerde

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.kafka.streams.processor.internals.InternalProcessorContext

getStateStore

Constructor Detail

GlobalProcessorContextImpl

public GlobalProcessorContextImpl(StreamsConfig config,
                                  org.apache.kafka.streams.processor.internals.StateManager stateMgr,
                                  StreamsMetricsImpl metrics,
                                  ThreadCache cache)

Method Detail

getStateStore

public StateStore getStateStore(java.lang.String name)

Description copied from interface: ProcessorContext

Get the state store given the store name.

Parameters:

name - The store name

Returns:

The state store instance

forward

public <K,V> void forward(K key,
                                V value)

Description copied from interface: ProcessorContext

Forwards a key/value pair to all downstream processors. Used the input record's timestamp as timestamp for the output record.

Parameters:

key - key

value - value

forward

public <K,V> void forward(K key,
                                V value,
                                To to)

No-op. This should only be called on GlobalStateStore#flush and there should be no child nodes

Parameters:

key - key

value - value

to - the options to use when forwarding

forward

@Deprecated
public <K,V> void forward(K key,
                                V value,
                                int childIndex)

Deprecated.

Description copied from interface: ProcessorContext

Forwards a key/value pair to one of the downstream processors designated by childIndex.

Parameters:

key - key

value - value

childIndex - index in list of children of this node

Throws:

java.lang.UnsupportedOperationException - on every invocation

forward

@Deprecated
public <K,V> void forward(K key,
                                V value,
                                java.lang.String childName)

Deprecated.

Description copied from interface: ProcessorContext

Forwards a key/value pair to one of the downstream processors designated by the downstream processor name.

Parameters:

key - key

value - value

childName - name of downstream processor

Throws:

java.lang.UnsupportedOperationException - on every invocation

commit

public void commit()

Description copied from interface: ProcessorContext

Requests a commit.

schedule

@Deprecated
public Cancellable schedule(long interval,
                            PunctuationType type,
                            Punctuator callback)

Deprecated.

Description copied from interface: ProcessorContext

Schedules a periodic operation for processors. A processor may call this method during initialization or processing to schedule a periodic callback — called a punctuation — to Punctuator.punctuate(long). The type parameter controls what notion of time is used for punctuation:

• PunctuationType.STREAM_TIME — uses "stream time", which is advanced by the processing of messages in accordance with the timestamp as extracted by the TimestampExtractor in use. The first punctuation will be triggered by the first record that is processed. NOTE: Only advanced if messages arrive
• PunctuationType.WALL_CLOCK_TIME — uses system time (the wall-clock time), which is advanced independent of whether new messages arrive. The first punctuation will be triggered after interval has elapsed. NOTE: This is best effort only as its granularity is limited by how long an iteration of the processing loop takes to complete

Skipping punctuations: Punctuations will not be triggered more than once at any given timestamp. This means that "missed" punctuation will be skipped. It's possible to "miss" a punctuation if:

• with PunctuationType.STREAM_TIME, when stream time advances more than interval
• with PunctuationType.WALL_CLOCK_TIME, on GC pause, too short interval, ...

Parameters:

interval - the time interval between punctuations in milliseconds

type - one of: PunctuationType.STREAM_TIME, PunctuationType.WALL_CLOCK_TIME

callback - a function consuming timestamps representing the current stream or system time

Returns:

a handle allowing cancellation of the punctuation schedule established by this method

Throws:

java.lang.UnsupportedOperationException - on every invocation

schedule

public Cancellable schedule(java.time.Duration interval,
                            PunctuationType type,
                            Punctuator callback)

Description copied from interface: ProcessorContext

Schedules a periodic operation for processors. A processor may call this method during initialization or processing to schedule a periodic callback — called a punctuation — to Punctuator.punctuate(long). The type parameter controls what notion of time is used for punctuation:

• PunctuationType.STREAM_TIME — uses "stream time", which is advanced by the processing of messages in accordance with the timestamp as extracted by the TimestampExtractor in use. The first punctuation will be triggered by the first record that is processed. NOTE: Only advanced if messages arrive
• PunctuationType.WALL_CLOCK_TIME — uses system time (the wall-clock time), which is advanced independent of whether new messages arrive. The first punctuation will be triggered after interval has elapsed. NOTE: This is best effort only as its granularity is limited by how long an iteration of the processing loop takes to complete

Skipping punctuations: Punctuations will not be triggered more than once at any given timestamp. This means that "missed" punctuation will be skipped. It's possible to "miss" a punctuation if:

• with PunctuationType.STREAM_TIME, when stream time advances more than interval
• with PunctuationType.WALL_CLOCK_TIME, on GC pause, too short interval, ...

Parameters:

interval - the time interval between punctuations (supported minimum is 1 millisecond)

type - one of: PunctuationType.STREAM_TIME, PunctuationType.WALL_CLOCK_TIME

callback - a function consuming timestamps representing the current stream or system time

Returns:

a handle allowing cancellation of the punctuation schedule established by this method

Throws:

java.lang.UnsupportedOperationException - on every invocation

logChange

public void logChange(java.lang.String storeName,
                      org.apache.kafka.common.utils.Bytes key,
                      byte[] value,
                      long timestamp)

transitionToActive

public void transitionToActive(StreamTask streamTask,
                               RecordCollector recordCollector,
                               ThreadCache newCache)

Description copied from interface: InternalProcessorContext

Transition to active task and register a new task and cache to this processor context

transitionToStandby

public void transitionToStandby(ThreadCache newCache)

Description copied from interface: InternalProcessorContext

Transition to standby task and register a dummy cache to this processor context

registerCacheFlushListener

public void registerCacheFlushListener(java.lang.String namespace,
                                       ThreadCache.DirtyEntryFlushListener listener)

Description copied from interface: InternalProcessorContext

Register a dirty entry flush listener for a particular namespace