Re: Need Suggestions on the Requirement!!

Posted by Fabian Hueske-2 on
URL: http://deprecated-apache-flink-user-mailing-list-archive.369.s1.nabble.com/Need-Suggestions-on-the-Requirement-tp17938p17939.html

Hi,

Ingesting a table from a retraction stream is not supported yet.

You can emulate an upsert ingestion mode with a LAST_VALUE aggregation function and a GROUP BY clause (see mail thread [1]).

Best, Fabian

[1] https://lists.apache.org/thread.html/b9fc2926f6fa460c70ebee8541b05f9af5cfe60c61f35637ccbf41a8@%3Cdev.flink.apache.org%3E

2018-01-23 13:36 GMT+01:00 Puneet Kinra <[hidden email]>:

Hi

Actually we are building a CRM (offer management tool) ,Currently Working on that..

1)  Multiple Streaming Table Sources.

2)  We are hitting a query on those Tables for getting aggregated values.

3)  Want to store the output of these aggregated values into the another table & then want to run the 

selection criteria.

Ex. 

Source A----->Table Registered as Table1 (AppendMode)----->Select count(*) as TableCount1 from Table1 group by UID ----> Count Per Subscriber(retract Stream) .

 

Source B----->Table Registered as Table2 (AppendMode)----->Select count(*) as TableCount2 from Table2 group by UID----> Count Per Subscriber(retract Stream) .

Now want to put this aggregator into the third table keeping Key UID (upsert Mode) & then want to run selection criteria on the third table

example :select * from Table3 where TableCount1==?  && TableCount1== ?

--