use of org.apache.flink.streaming.api.windowing.assigners.MergingWindowAssigner in project flink by apache.
the class EvictingWindowOperator method onProcessingTime.
@Override
public void onProcessingTime(InternalTimer<K, W> timer) throws Exception {
context.key = timer.getKey();
context.window = timer.getNamespace();
evictorContext.key = timer.getKey();
evictorContext.window = timer.getNamespace();
MergingWindowSet<W> mergingWindows = null;
if (windowAssigner instanceof MergingWindowAssigner) {
mergingWindows = getMergingWindowSet();
W stateWindow = mergingWindows.getStateWindow(context.window);
if (stateWindow == null) {
// window and therefore the Trigger state, however, so nothing to do.
return;
} else {
evictingWindowState.setCurrentNamespace(stateWindow);
}
} else {
evictingWindowState.setCurrentNamespace(context.window);
}
Iterable<StreamRecord<IN>> contents = evictingWindowState.get();
if (contents != null) {
TriggerResult triggerResult = context.onProcessingTime(timer.getTimestamp());
if (triggerResult.isFire()) {
emitWindowContents(context.window, contents, evictingWindowState);
}
if (triggerResult.isPurge()) {
evictingWindowState.clear();
}
}
if (!windowAssigner.isEventTime() && isCleanupTime(context.window, timer.getTimestamp())) {
clearAllState(context.window, evictingWindowState, mergingWindows);
}
if (mergingWindows != null) {
// need to make sure to update the merging state in state
mergingWindows.persist();
}
}
use of org.apache.flink.streaming.api.windowing.assigners.MergingWindowAssigner in project flink by apache.
the class WindowOperator method onProcessingTime.
@Override
public void onProcessingTime(InternalTimer<K, W> timer) throws Exception {
context.key = timer.getKey();
context.window = timer.getNamespace();
MergingWindowSet<W> mergingWindows;
if (windowAssigner instanceof MergingWindowAssigner) {
mergingWindows = getMergingWindowSet();
W stateWindow = mergingWindows.getStateWindow(context.window);
if (stateWindow == null) {
// window and therefore the Trigger state, however, so nothing to do.
return;
} else {
windowState.setCurrentNamespace(stateWindow);
}
} else {
windowState.setCurrentNamespace(context.window);
mergingWindows = null;
}
ACC contents = null;
if (windowState != null) {
contents = windowState.get();
}
if (contents != null) {
TriggerResult triggerResult = context.onProcessingTime(timer.getTimestamp());
if (triggerResult.isFire()) {
emitWindowContents(context.window, contents);
}
if (triggerResult.isPurge()) {
windowState.clear();
}
}
if (!windowAssigner.isEventTime() && isCleanupTime(context.window, timer.getTimestamp())) {
clearAllState(context.window, windowState, mergingWindows);
}
if (mergingWindows != null) {
// need to make sure to update the merging state in state
mergingWindows.persist();
}
}
use of org.apache.flink.streaming.api.windowing.assigners.MergingWindowAssigner in project flink by apache.
the class WindowOperator method processElement.
@Override
public void processElement(StreamRecord<IN> element) throws Exception {
final Collection<W> elementWindows = windowAssigner.assignWindows(element.getValue(), element.getTimestamp(), windowAssignerContext);
//if element is handled by none of assigned elementWindows
boolean isSkippedElement = true;
final K key = this.<K>getKeyedStateBackend().getCurrentKey();
if (windowAssigner instanceof MergingWindowAssigner) {
MergingWindowSet<W> mergingWindows = getMergingWindowSet();
for (W window : elementWindows) {
// adding the new window might result in a merge, in that case the actualWindow
// is the merged window and we work with that. If we don't merge then
// actualWindow == window
W actualWindow = mergingWindows.addWindow(window, new MergingWindowSet.MergeFunction<W>() {
@Override
public void merge(W mergeResult, Collection<W> mergedWindows, W stateWindowResult, Collection<W> mergedStateWindows) throws Exception {
context.key = key;
context.window = mergeResult;
context.onMerge(mergedWindows);
for (W m : mergedWindows) {
context.window = m;
context.clear();
deleteCleanupTimer(m);
}
// merge the merged state windows into the newly resulting state window
windowMergingState.mergeNamespaces(stateWindowResult, mergedStateWindows);
}
});
// drop if the window is already late
if (isWindowLate(actualWindow)) {
mergingWindows.retireWindow(actualWindow);
continue;
}
isSkippedElement = false;
W stateWindow = mergingWindows.getStateWindow(actualWindow);
if (stateWindow == null) {
throw new IllegalStateException("Window " + window + " is not in in-flight window set.");
}
windowState.setCurrentNamespace(stateWindow);
windowState.add(element.getValue());
context.key = key;
context.window = actualWindow;
TriggerResult triggerResult = context.onElement(element);
if (triggerResult.isFire()) {
ACC contents = windowState.get();
if (contents == null) {
continue;
}
emitWindowContents(actualWindow, contents);
}
if (triggerResult.isPurge()) {
windowState.clear();
}
registerCleanupTimer(actualWindow);
}
// need to make sure to update the merging state in state
mergingWindows.persist();
} else {
for (W window : elementWindows) {
// drop if the window is already late
if (isWindowLate(window)) {
continue;
}
isSkippedElement = false;
windowState.setCurrentNamespace(window);
windowState.add(element.getValue());
context.key = key;
context.window = window;
TriggerResult triggerResult = context.onElement(element);
if (triggerResult.isFire()) {
ACC contents = windowState.get();
if (contents == null) {
continue;
}
emitWindowContents(window, contents);
}
if (triggerResult.isPurge()) {
windowState.clear();
}
registerCleanupTimer(window);
}
}
// windowAssigner is event time and current timestamp + allowed lateness no less than element timestamp
if (isSkippedElement && lateDataOutputTag != null && isElementLate(element)) {
sideOutput(element);
}
}
use of org.apache.flink.streaming.api.windowing.assigners.MergingWindowAssigner in project flink by apache.
the class AllWindowedStream method fold.
/**
* Applies the given window function to each window. The window function is called for each
* evaluation of the window for each key individually. The output of the window function is
* interpreted as a regular non-windowed stream.
*
* <p>
* Arriving data is incrementally aggregated using the given fold function.
*
* @param initialValue The initial value of the fold.
* @param foldFunction The fold function that is used for incremental aggregation.
* @param function The process window function.
* @param foldAccumulatorType Type information for the result type of the fold function
* @param resultType Type information for the result type of the window function
* @return The data stream that is the result of applying the window function to the window.
*/
@PublicEvolving
public <ACC, R> SingleOutputStreamOperator<R> fold(ACC initialValue, FoldFunction<T, ACC> foldFunction, ProcessAllWindowFunction<ACC, R, W> function, TypeInformation<ACC> foldAccumulatorType, TypeInformation<R> resultType) {
if (foldFunction instanceof RichFunction) {
throw new UnsupportedOperationException("FoldFunction of fold can not be a RichFunction.");
}
if (windowAssigner instanceof MergingWindowAssigner) {
throw new UnsupportedOperationException("Fold cannot be used with a merging WindowAssigner.");
}
//clean the closures
function = input.getExecutionEnvironment().clean(function);
foldFunction = input.getExecutionEnvironment().clean(foldFunction);
String callLocation = Utils.getCallLocationName();
String udfName = "AllWindowedStream." + callLocation;
String opName;
KeySelector<T, Byte> keySel = input.getKeySelector();
OneInputStreamOperator<T, R> operator;
if (evictor != null) {
@SuppressWarnings({ "unchecked", "rawtypes" }) TypeSerializer<StreamRecord<T>> streamRecordSerializer = (TypeSerializer<StreamRecord<T>>) new StreamElementSerializer(input.getType().createSerializer(getExecutionEnvironment().getConfig()));
ListStateDescriptor<StreamRecord<T>> stateDesc = new ListStateDescriptor<>("window-contents", streamRecordSerializer);
opName = "TriggerWindow(" + windowAssigner + ", " + stateDesc + ", " + trigger + ", " + evictor + ", " + udfName + ")";
operator = new EvictingWindowOperator<>(windowAssigner, windowAssigner.getWindowSerializer(getExecutionEnvironment().getConfig()), keySel, input.getKeyType().createSerializer(getExecutionEnvironment().getConfig()), stateDesc, new InternalIterableProcessAllWindowFunction<>(new FoldApplyProcessAllWindowFunction<>(initialValue, foldFunction, function, foldAccumulatorType)), trigger, evictor, allowedLateness, lateDataOutputTag);
} else {
FoldingStateDescriptor<T, ACC> stateDesc = new FoldingStateDescriptor<>("window-contents", initialValue, foldFunction, foldAccumulatorType.createSerializer(getExecutionEnvironment().getConfig()));
opName = "TriggerWindow(" + windowAssigner + ", " + stateDesc + ", " + trigger + ", " + udfName + ")";
operator = new WindowOperator<>(windowAssigner, windowAssigner.getWindowSerializer(getExecutionEnvironment().getConfig()), keySel, input.getKeyType().createSerializer(getExecutionEnvironment().getConfig()), stateDesc, new InternalSingleValueProcessAllWindowFunction<>(function), trigger, allowedLateness, lateDataOutputTag);
}
return input.transform(opName, resultType, operator).forceNonParallel();
}
use of org.apache.flink.streaming.api.windowing.assigners.MergingWindowAssigner in project flink by apache.
the class AllWindowedStream method apply.
/**
* Applies the given window function to each window. The window function is called for each
* evaluation of the window for each key individually. The output of the window function is
* interpreted as a regular non-windowed stream.
*
* <p>
* Arriving data is incrementally aggregated using the given fold function.
*
* @param initialValue The initial value of the fold.
* @param foldFunction The fold function that is used for incremental aggregation.
* @param function The window function.
* @param resultType Type information for the result type of the window function
* @return The data stream that is the result of applying the window function to the window.
*
* @deprecated Use {@link #fold(Object, FoldFunction, AllWindowFunction, TypeInformation, TypeInformation)} instead.
*/
@Deprecated
public <R> SingleOutputStreamOperator<R> apply(R initialValue, FoldFunction<T, R> foldFunction, AllWindowFunction<R, R, W> function, TypeInformation<R> resultType) {
if (foldFunction instanceof RichFunction) {
throw new UnsupportedOperationException("ReduceFunction of apply can not be a RichFunction.");
}
if (windowAssigner instanceof MergingWindowAssigner) {
throw new UnsupportedOperationException("Fold cannot be used with a merging WindowAssigner.");
}
//clean the closures
function = input.getExecutionEnvironment().clean(function);
foldFunction = input.getExecutionEnvironment().clean(foldFunction);
String callLocation = Utils.getCallLocationName();
String udfName = "AllWindowedStream." + callLocation;
String opName;
KeySelector<T, Byte> keySel = input.getKeySelector();
OneInputStreamOperator<T, R> operator;
if (evictor != null) {
@SuppressWarnings({ "unchecked", "rawtypes" }) TypeSerializer<StreamRecord<T>> streamRecordSerializer = (TypeSerializer<StreamRecord<T>>) new StreamElementSerializer(input.getType().createSerializer(getExecutionEnvironment().getConfig()));
ListStateDescriptor<StreamRecord<T>> stateDesc = new ListStateDescriptor<>("window-contents", streamRecordSerializer);
opName = "TriggerWindow(" + windowAssigner + ", " + stateDesc + ", " + trigger + ", " + evictor + ", " + udfName + ")";
operator = new EvictingWindowOperator<>(windowAssigner, windowAssigner.getWindowSerializer(getExecutionEnvironment().getConfig()), keySel, input.getKeyType().createSerializer(getExecutionEnvironment().getConfig()), stateDesc, new InternalIterableAllWindowFunction<>(new FoldApplyAllWindowFunction<>(initialValue, foldFunction, function, resultType)), trigger, evictor, allowedLateness, lateDataOutputTag);
} else {
FoldingStateDescriptor<T, R> stateDesc = new FoldingStateDescriptor<>("window-contents", initialValue, foldFunction, resultType.createSerializer(getExecutionEnvironment().getConfig()));
opName = "TriggerWindow(" + windowAssigner + ", " + stateDesc + ", " + trigger + ", " + udfName + ")";
operator = new WindowOperator<>(windowAssigner, windowAssigner.getWindowSerializer(getExecutionEnvironment().getConfig()), keySel, input.getKeyType().createSerializer(getExecutionEnvironment().getConfig()), stateDesc, new InternalSingleValueAllWindowFunction<>(function), trigger, allowedLateness, lateDataOutputTag);
}
return input.transform(opName, resultType, operator).forceNonParallel();
}
Aggregations