Smart Connector Node

Smart Connector Node

Overview

The Smart Connector Node is a dynamic connector that decides which connection and which function to use at runtime based on pipeline data.
Instead of hard-coding a single connection profile and function in the node configuration, you provide expressions that resolve to the right connection and function for each execution, with optional caching for high performance.

This makes Smart Connector ideal for:

• Multi-tenant routing where each tenant has its own connection profile.
• Conditional system selection (premium vs. standard processors).
• Dynamic protocol or data-source selection (PLC vs. MQTT vs. REST) using a single node.

Configuration Reference

Execution Tab

Execution Tab

Advanced Tab

Advanced Tab

Settings Tab

Settings Tab

Execution Tab

Field	Type	Default	Description
Lookup Mode	string	"name"	How identifiers are resolved: "name" (cached name→ID) or "id" (direct UUID).
Connection Identifier	string	""	Expression resolving to a connection name or ID. Required.
Function Identifier	string	""	Expression resolving to a function name or ID. Required.
Parameters	object	{}	Key/value map passed to the function. Each value may be a literal or expression.

Advanced Tab

Field	Type	Default	Description
Timeout Override	number	empty	Optional timeout (seconds) overriding the function’s default timeout.
Debug Mode	boolean	false	Enables detailed logging and additional metrics.
Cache TTL	number	300 (name mode)	Cache time-to-live in seconds for name lookups. 0 disables caching.

Expression Examples

• Connection identifier:
  • {{$input.targetConnection}}
  • {{$input.tenantId}}-database
• Function identifier:
  • {{$input.targetFunction}}
  • {{$input.amount > 10000 ? "process-premium" : "process-standard"}}
• Parameters object (JSON editor):

{
  "orderId": "{{$input.orderId}}",
  "amount": "{{$input.amount}}",
  "currency": "{{$input.currency || 'USD'}}"
}

Settings Tab

Field	Type	Default	Description
Retry on Fail	boolean	false	Automatically retry transient failures from the connection function.
On Error	string	"stopPipeline"	Error strategy: stop the pipeline or continue via error branch depending on your UI options.
Notes	string	""	Internal documentation about routing logic and usage.
Display Note in Pipeline	boolean	false	Shows notes on the canvas near the node.
Input Sync Mode	string	"first"	When multiple upstream inputs feed this node, "first" executes as soon as one packet arrives.
Wait Timeout	string	"30s"	Max time to wait when sync modes are changed (for advanced scenarios).

---

Usage Examples

Single Pipeline for Multiple Connections (For-Each + Smart Connector)

In many production scenarios you maintain a list of connections and functions (for different lines, machines, or tenants). Instead of adding many static connector nodes to the pipeline, you can:

• Use a For-Each Loop node that iterates over the list.
• Place one Smart Connector node inside the loop.
• Let Smart Connector resolve the correct connection and function for each item at runtime.

Assume the incoming payload to the For-Each Loop looks like:

{
  "connections": [
    { "connectionName": "line-1-plc", "functionName": "target-function", "payload": { "line": 1 } },
    { "connectionName": "line-2-plc", "functionName": "target-function", "payload": { "line": 2 } }
  ]
}

Configure the For-Each Loop to iterate over connections, so the Smart Connector receives each item as $input.

Field	Value
Lookup Mode	name
Connection Identifier	{{$input.connectionName}}
Function Identifier	{{$input.functionName}}
Parameters	{{$input.payload}}

With this pattern:

• You keep a single pipeline definition instead of duplicating many connector nodes.
• Adding a new connection is as simple as updating the list (for example via an upstream node or configuration).
• The rest of the pipeline can remain the same, regardless of how many connections you have.

---

Best Practices

• Prefer name mode while designing and validating pipelines; switch to ID mode later for ultra-low latency when IDs are readily available.
• Keep expressions simple and readable; complex routing logic can be extracted to upstream nodes (e.g., Data Instance or JavaScript) that compute a targetConnection or targetFunction.
• Combine Smart Connector with Condition, For-Each Loop, or Data Instance nodes to build powerful, dynamic routing and transformation flows with minimal duplication.