Class GroupedAggregator
- java.lang.Object
-
- org.apache.storm.trident.operation.impl.GroupedAggregator
-
- All Implemented Interfaces:
Serializable
,Aggregator<Object[]>
,Operation
public class GroupedAggregator extends Object implements Aggregator<Object[]>
- See Also:
- Serialized Form
-
-
Constructor Summary
Constructors Constructor Description GroupedAggregator(Aggregator agg, Fields group, Fields input, int outSize)
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description void
aggregate(Object[] arr, TridentTuple tuple, TridentCollector collector)
void
cleanup()
When running in local mode, called when the local cluster is being shut down.void
complete(Object[] arr, TridentCollector collector)
Object[]
init(Object batchId, TridentCollector collector)
void
prepare(Map<String,Object> conf, TridentOperationContext context)
Called when the `Operation` is first initialized.
-
-
-
Constructor Detail
-
GroupedAggregator
public GroupedAggregator(Aggregator agg, Fields group, Fields input, int outSize)
-
-
Method Detail
-
prepare
public void prepare(Map<String,Object> conf, TridentOperationContext context)
Description copied from interface:Operation
Called when the `Operation` is first initialized.- Specified by:
prepare
in interfaceOperation
- Parameters:
conf
- the Storm configuration mapcontext
- the operation context which provides information such as the number of partitions in the stream, and the current partition index. It also provides methods for registering operation-specific metrics.- See Also:
TridentOperationContext
-
init
public Object[] init(Object batchId, TridentCollector collector)
- Specified by:
init
in interfaceAggregator<Object[]>
-
aggregate
public void aggregate(Object[] arr, TridentTuple tuple, TridentCollector collector)
- Specified by:
aggregate
in interfaceAggregator<Object[]>
-
complete
public void complete(Object[] arr, TridentCollector collector)
- Specified by:
complete
in interfaceAggregator<Object[]>
-
-