Re: Categorize or GroupBy datastream data and process with CEP separately

Hi, Flink’s Pattern API allows you to specify a time interval for a pattern to occur. Use the within call to specify a valid time interval for your pattern. Pattern pattern = ...; Pattern timedPattern = pattern.within(Time.minutes(10)); Cheers, Till ​ On Thu, Jan 5, 2017 at

Re: Categorize or GroupBy datastream data and process with CEP separately

Thanks for you reply and absolutely apologies for my insanity Thats the place I confuse a lot. The keyedstream, processe by unique key. But as per the example is see, I can make a pattern like temperature > 90...temperature > 80... within 'T' minutes. What am trying to achieve ,is "Temperate

Re: Categorize or GroupBy datastream data and process with CEP separately

If you are trying to do the same CEP computation for each event type that's exactly what will happen for a KeyedStream. For example if you key by event type you can think of this like creating a separate substream for each key/eventType and then applying the CEP operations to each substream

Re: Categorize or GroupBy datastream data and process with CEP separately

This is my code. *SplitStream splitStream = inputStream.split(new OutputSelector() { @Override public Iterable select(MonitoringEvent me) { List ml = new ArrayList(); ml.add(me.getEventType()); return ml; }* I

Re: Categorize or GroupBy datastream data and process with CEP separately

Hi Thanks for the response. But sorry . Am not sure how is it inline with my requirement Right now, am using SplitterStream to split it based . Which helps me splitting the response to separate streams. The problem is I have a dynamic set of splitstream . the x, y, z is not predefined. It can

Re: Categorize or GroupBy datastream data and process with CEP separately

I think what you want here is to apply CEP processing on a KeyedStream - see the last CEP Example here: https://ci.apache.org/projects/flink/flink-docs-release-1.1/apis/streaming/libs/cep.html#examples On Tue, Jan 3, 2017 at 3:30 AM, madhairsilence wrote: > Assume I have

Categorize or GroupBy datastream data and process with CEP separately

Assume I have a datastream *x:1, y:2 , z:3 , x:7 , y:-1, z:0, z:3 , z:2, y:3 ,x: 2 ,y:6* How do I put x,y,z in their own bucket and apply my CEP rule on it. *x:1, x:7,x: 2 y:2, y:-1, y:3 , y:6 z:3, z:0 , z:3, z:2* Or to put it in other way. How do I split the stream in to these categories(one