Hi,
this is a known bug: https://issues.apache.org/jira/browse/FLINK-3869.
I'm still hoping that we can get a workaround in for Flink 1.2. See my last
comment in the Jira Issue.
Cheers,
Aljoscha
On Mon, 14 Nov 2016 at 14:49 Stephan Epping <stephan.epp...@zweitag.de>
wrote:
> Hello,
>
> I wondered if there is a particular reason for the window function to have
> explicitly the same input/output type?
>
> public <R> SingleOutputStreamOperator<R> apply(R initialValue,
> FoldFunction<T, R> foldFunction, WindowFunction<R, R, K, W> function)
>
> for example (the following does not work):
>
> DataStream<SensorAggregatedValue> aggregates = values
> .assignTimestampsAndWatermarks(new
> SensorTimestampsAndWatermarks())
> .keyBy("id")
> .timeWindow(Time.minutes(1))
> .apply(new SensorValueAccumulator(), new
> AccumulateSensorValues(), new AggregateSensorValues());
>
> because in this case my accumulator object does not have any id or
> timestamp information - just count, sum, min, max etc. And finally in the
> window function I receive the key (sensorId) and time window (start/end)
> and can build an aggregated value with all information needed. But
> currently the apply function forces me to use one cluttered class with id,
> count, sum, …, where the id,start,end time are invalid during the fold
> function.
>
> kind regards,
> Stephan
>
