Option A: Configure an input that only emits a single event, e.g. JDBC with a query that returns nothing but the latest ID, so you have a starting point. Then use a Jdbc_streaming filter for the actual query that returns multiple rows. Then there would be only one result with all the rows in one field. In the last step one could restructure the fields with Ruby as one pleases. Option B: Execute your usual query that returns multiple events. Use an aggregate filter and in the code for the aggregation you would append the current event's ID to an array of IDs in the aggregation map and cancel the event. Then you would either need a timeout or a condition to detect when to push the map data as an event.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.