Split Array

---

Description

The Split Array processor transforms array fields into multiple individual events, with each array element becoming a separate event. It supports:

• Array element extraction
• Context preservation
• Nested field handling
• Custom field naming

This processor is essential for:

• Converting batch data into individual events
• Processing array elements independently
• Distributing array data across streams
• Enabling element-wise analysis

---

Required input

The processor requires a data stream containing at least one array field. The array can contain elements of any supported data type:

• Numbers (integer or float)
• Strings
• Booleans
• Objects
• Nested arrays

---

Configuration

Array Field Selection

Select the array field to split into individual events. The field must be an array type.

Keep Fields

Select one or more fields from the input event that should be preserved in each output event. These fields will be copied to each output event.

Output

For each element in the input array, the processor creates a new event containing:

• The array element as a single value in a field named "array_value"
• All selected fields from the original event

Example

Input Event

{
  "deviceId": "sensor123",
  "timestamp": 1586380104915,
  "measurements": [22.5, 23.1, 22.8, 23.4],
  "status": "active"
}

Configuration

• Array Field: measurements
• Keep Fields: deviceId, timestamp, status

Output Events

// First element
{
  "deviceId": "sensor123",
  "timestamp": 1586380104915,
  "status": "active",
  "array_value": 22.5
}

// Second element
{
  "deviceId": "sensor123",
  "timestamp": 1586380104915,
  "status": "active",
  "array_value": 23.1
}

// Third element
{
  "deviceId": "sensor123",
  "timestamp": 1586380104915,
  "status": "active",
  "array_value": 22.8
}

// Fourth element
{
  "deviceId": "sensor123",
  "timestamp": 1586380104915,
  "status": "active",
  "array_value": 23.4
}

Use Cases

1. Batch Processing

   • Split batch sensor readings
   • Process multi-measurement data
   • Handle grouped observations
   • Transform batch uploads

2. Data Distribution

   • Distribute workload across processors
   • Enable parallel processing
   • Balance processing load
   • Scale data processing

Notes

• Output events maintain original event order
• Empty arrays produce no output events
• Null array elements are preserved
• Processing is stateless
• Memory usage scales with array size
• Nested fields are handled automatically