trigger-event

This agent writes a new record (an "event") to a different topic, based on a condition. It can also optionally drop the record from the main flow of the pipeline.

This agent is different from the dispatch agent as the record that is sent to the new topic is a new record, not the same record that was received.

Example

This is an example using the trigger-event agent to write a new record to a different topic when a condition is met.

  - name: "Split some text"  
    type: "text-splitter"
    input: input-topic-splitter
    configuration:
       ....

  - name: "Trigger event on last chunk"
    type: "trigger-event"
    output: output-topic-chunks
    configuration:
      destination: drop-stale-chunks-topic
      continue-processing: true
      when: fn:toInt(properties.text_num_chunks) == (fn:toInt(properties.chunk_id) + 1)
      fields:
          - name: "value.filename"
            expression: "key.filename"

In this example the text-splitter agent splits a text into a set of chunks. When the last chunk is processed, the trigger-event agent writes a new record to the drop-stale-chunks-topic topic. The new record will have a filename field in the value part ("value.filename").

Defining the contents of the output record

The trigger-event agent configures a set of fields that will be written to the output record. As usual you can write to the key part, the value part, and the properties of the record. Use the expression language to define the fields and write the expression.

Aborting the processing downstream

The trigger-event agent can also abort the processing of the record downstream by setting the continue-processing property to false. This is useful in cases where you have some system events that you want to write to a different topic, but you don't want to continue processing the record downstream.

Configuration

Check out the full configuration properties in the API Reference page.