Value Changed
![]()
Description
The Value Changed processor detects changes in field values and outputs events only when changes occur. It supports:
- Any data type comparison
- Dimension-based state tracking
- Change timestamp tracking
- State size management
- Multi-dimensional monitoring
This processor is essential for:
- Detecting value changes
- Tracking state transitions
- Monitoring field updates
- Building change logs
- Implementing change triggers
Required input
The processor requires a data stream containing at least one field to monitor for changes.
Configuration
Keep Fields
Select the field to monitor for changes. The processor will output an event only when the selected field's value changes.
Output
The processor creates a new event containing:
- All original fields from the input event
- A new field named "change_detected" containing the timestamp when the change occurred
- Only when the monitored field changes value
Example
Input Event Stream
{
"deviceId": "sensor01",
"location": "l1",
"value": 12,
"timestamp": 1586380104915
}
{
"deviceId": "sensor01",
"location": "l1",
"value": 15,
"timestamp": 1586380105015
}
Configuration
- Keep Fields: value
Output Event
{
"deviceId": "sensor01",
"location": "l1",
"value": 15,
"timestamp": 1586380105015,
"change_detected": 1586380105015
}
Use Cases
-
Change Detection
- Monitor value changes
- Track state transitions
- Detect field updates
- Build change logs
-
State Monitoring
- Track state changes
- Monitor transitions
- Detect updates
- Log changes
Notes
- Supports any data type
- Processing is stateful
- State size limited to 5000 entries
- Events only emitted on value changes
- Original fields are preserved
- Change timestamps are added
- Dimension-based state tracking
- Multi-dimensional monitoring supported