Merge Two Streams

---

Description

The Merge By Enrich processor combines two event streams by enriching one stream with properties from the other. It supports:

• Real-time event stream enrichment
• Last-event state tracking
• Custom output field selection
• Dynamic event composition
• Stateful event processing

This processor is essential for:

• Enriching events with additional context
• Combining related data streams
• Creating unified event views
• Building composite event structures

---

Required Input

The processor requires two input streams:

• First stream: Any event stream with at least one property
• Second stream: Any event stream with at least one property

---

Configuration

Stream Selection

During pipeline modeling, you can:

• Select which stream should be enriched (Stream 1 or Stream 2)
• Choose which fields from each stream should be included in the output
• The processor will maintain the event frequency of the selected stream

Output

The processor creates a new event containing the selected fields from both input streams. The output is generated whenever:

• A new event arrives from the selected stream
• The last event from the other stream is available

Example

Input Event Stream 1

{
  "deviceId": "sensor01",
  "temperature": 25.5,
  "timestamp": 1586380104915
}

Input Event Stream 2

{
  "location": "room1",
  "humidity": 45.2,
  "timestamp": 1586380104915
}

Configuration

• Selected Stream: Stream 1
• Output Fields: deviceId, temperature, location, humidity

Output Event

{
  "deviceId": "sensor01",
  "temperature": 25.5,
  "location": "room1",
  "humidity": 45.2
}

Use Cases

1. Event Enrichment

   • Add context to sensor readings
   • Combine related metrics
   • Create unified views
   • Join event streams

2. Data Integration

   • Merge sensor data
   • Combine related events
   • Create composite events
   • Build rich event structures

Notes

• Events are enriched in real-time
• Last event state is maintained
• Output fields are configurable
• Original event structure is preserved
• Events are forwarded at the frequency of the selected stream
• State is cleared on pipeline stop