use of org.apache.flink.annotation.PublicEvolving in project flink by apache.
the class BinaryInputFormat method reopen.
@PublicEvolving
@Override
public void reopen(FileInputSplit split, Tuple2<Long, Long> state) throws IOException {
Preconditions.checkNotNull(split, "reopen() cannot be called on a null split.");
Preconditions.checkNotNull(state, "reopen() cannot be called with a null initial state.");
try {
this.open(split);
} finally {
this.blockInfo = this.createAndReadBlockInfo();
long blockPos = state.f0;
this.readRecords = state.f1;
this.stream.seek(this.splitStart + blockPos);
this.blockBasedInput = new BlockBasedInput(this.stream, (int) blockPos, this.splitLength);
this.dataInputStream = new DataInputViewStreamWrapper(blockBasedInput);
}
}
use of org.apache.flink.annotation.PublicEvolving in project flink by apache.
the class TypeExtractor method getAllDeclaredFields.
/**
* Recursively determine all declared fields
* This is required because class.getFields() is not returning fields defined
* in parent classes.
*
* @param clazz class to be analyzed
* @param ignoreDuplicates if true, in case of duplicate field names only the lowest one
* in a hierarchy will be returned; throws an exception otherwise
* @return list of fields
*/
@PublicEvolving
public static List<Field> getAllDeclaredFields(Class<?> clazz, boolean ignoreDuplicates) {
List<Field> result = new ArrayList<Field>();
while (clazz != null) {
Field[] fields = clazz.getDeclaredFields();
for (Field field : fields) {
if (Modifier.isTransient(field.getModifiers()) || Modifier.isStatic(field.getModifiers())) {
// we have no use for transient or static fields
continue;
}
if (hasFieldWithSameName(field.getName(), result)) {
if (ignoreDuplicates) {
continue;
} else {
throw new InvalidTypesException("The field " + field + " is already contained in the hierarchy of the " + clazz + "." + "Please use unique field names through your classes hierarchy");
}
}
result.add(field);
}
clazz = clazz.getSuperclass();
}
return result;
}
use of org.apache.flink.annotation.PublicEvolving in project flink by apache.
the class TypeExtractor method getBinaryOperatorReturnType.
/**
* Returns the binary operator's return type.
*
* @param function Function to extract the return type from
* @param baseClass Base class of the function
* @param inputTypeArgumentIndex Index of the type argument of function's first parameter
* specifying the input type if it is wrapped (Iterable, Map,
* etc.). Otherwise -1.
* @param outputTypeArgumentIndex Index of the type argument of functions second parameter
* specifying the output type if it is wrapped in a Collector.
* Otherwise -1.
* @param in1Type Type of the left side input elements (In case of an iterable, it is the element type)
* @param in2Type Type of the right side input elements (In case of an iterable, it is the element type)
* @param functionName Function name
* @param allowMissing Can the type information be missing
* @param <IN1> Left side input type
* @param <IN2> Right side input type
* @param <OUT> Output type
* @return TypeInformation of the return type of the function
*/
@SuppressWarnings("unchecked")
@PublicEvolving
public static <IN1, IN2, OUT> TypeInformation<OUT> getBinaryOperatorReturnType(Function function, Class<?> baseClass, int inputTypeArgumentIndex, int outputTypeArgumentIndex, TypeInformation<IN1> in1Type, TypeInformation<IN2> in2Type, String functionName, boolean allowMissing) {
try {
final LambdaExecutable exec;
try {
exec = checkAndExtractLambda(function);
} catch (TypeExtractionException e) {
throw new InvalidTypesException("Internal error occurred.", e);
}
if (exec != null) {
// check for lambda type erasure
validateLambdaGenericParameters(exec);
// parameters must be accessed from behind, since JVM can add additional parameters e.g. when using local variables inside lambda function
final int paramLen = exec.getParameterTypes().length - 1;
final Type input1 = (outputTypeArgumentIndex >= 0) ? exec.getParameterTypes()[paramLen - 2] : exec.getParameterTypes()[paramLen - 1];
final Type input2 = (outputTypeArgumentIndex >= 0) ? exec.getParameterTypes()[paramLen - 1] : exec.getParameterTypes()[paramLen];
validateInputType((inputTypeArgumentIndex >= 0) ? extractTypeArgument(input1, inputTypeArgumentIndex) : input1, in1Type);
validateInputType((inputTypeArgumentIndex >= 0) ? extractTypeArgument(input2, inputTypeArgumentIndex) : input2, in2Type);
if (function instanceof ResultTypeQueryable) {
return ((ResultTypeQueryable<OUT>) function).getProducedType();
}
return new TypeExtractor().privateCreateTypeInfo((outputTypeArgumentIndex >= 0) ? extractTypeArgument(exec.getParameterTypes()[paramLen], outputTypeArgumentIndex) : exec.getReturnType(), in1Type, in2Type);
} else {
validateInputType(baseClass, function.getClass(), 0, in1Type);
validateInputType(baseClass, function.getClass(), 1, in2Type);
if (function instanceof ResultTypeQueryable) {
return ((ResultTypeQueryable<OUT>) function).getProducedType();
}
return new TypeExtractor().privateCreateTypeInfo(baseClass, function.getClass(), 2, in1Type, in2Type);
}
} catch (InvalidTypesException e) {
if (allowMissing) {
return (TypeInformation<OUT>) new MissingTypeInfo(functionName != null ? functionName : function.toString(), e);
} else {
throw e;
}
}
}
use of org.apache.flink.annotation.PublicEvolving in project flink by apache.
the class AllWindowedStream method aggregate.
/**
* 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 aggregate function. This means
* that the window function typically has only a single value to process when called.
*
* @param aggregateFunction The aggregation function that is used for incremental aggregation.
* @param windowFunction The process window function.
* @param accumulatorType Type information for the internal accumulator type of the aggregation 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.
*
* @param <ACC> The type of the AggregateFunction's accumulator
* @param <V> The type of AggregateFunction's result, and the WindowFunction's input
* @param <R> The type of the elements in the resulting stream, equal to the
* WindowFunction's result type
*/
@PublicEvolving
public <ACC, V, R> SingleOutputStreamOperator<R> aggregate(AggregateFunction<T, ACC, V> aggregateFunction, ProcessAllWindowFunction<V, R, W> windowFunction, TypeInformation<ACC> accumulatorType, TypeInformation<V> aggregateResultType, TypeInformation<R> resultType) {
checkNotNull(aggregateFunction, "aggregateFunction");
checkNotNull(windowFunction, "windowFunction");
checkNotNull(accumulatorType, "accumulatorType");
checkNotNull(aggregateResultType, "aggregateResultType");
checkNotNull(resultType, "resultType");
if (aggregateFunction instanceof RichFunction) {
throw new UnsupportedOperationException("This aggregate function cannot be a RichFunction.");
}
//clean the closures
windowFunction = input.getExecutionEnvironment().clean(windowFunction);
aggregateFunction = input.getExecutionEnvironment().clean(aggregateFunction);
final String callLocation = Utils.getCallLocationName();
final String udfName = "AllWindowedStream." + callLocation;
final String opName;
final 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 InternalAggregateProcessAllWindowFunction<>(aggregateFunction, windowFunction), trigger, evictor, allowedLateness, lateDataOutputTag);
} else {
AggregatingStateDescriptor<T, ACC, V> stateDesc = new AggregatingStateDescriptor<>("window-contents", aggregateFunction, accumulatorType.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<>(windowFunction), trigger, allowedLateness, lateDataOutputTag);
}
return input.transform(opName, resultType, operator).forceNonParallel();
}
use of org.apache.flink.annotation.PublicEvolving 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();
}
Aggregations