Merge

Type key: merge_node
Group: Flow Control
Category: flow
Tier: CORE
Version: 2.0.0

The Merge node joins data coming from multiple upstream branches back into a single stream. It is used after If/Switch nodes when you want to re-unify the execution paths, or after parallel branches that collect results from different sources.

Handles

Handle	Direction	Description
input	Input (multiple)	Accepts connections from multiple upstream nodes
output	Output	Emits the merged result

The Merge node is the only core node that accepts multiple input connections.

Configuration

Field	Type	Default	Description
label	string	Merge	Display label on the canvas
mergeStrategy	enum	concat	How to combine the incoming data streams
mergeKey	string	id	Key field used by key-based strategies

Merge Strategies

concat

Concatenates all input arrays into a single flat array.

Input from branch A: [{id:1}, {id:2}]
Input from branch B: [{id:3}, {id:4}]
Output: [{id:1}, {id:2}, {id:3}, {id:4}]

Use this when branches produce arrays of the same type and you want to process them together downstream.

union

Concatenates all arrays then removes duplicates based on a mergeKey field (default: id). Each unique key appears only once in the output.

Input A: [{id:1}, {id:2}]
Input B: [{id:2}, {id:3}]
Output (mergeKey: id): [{id:1}, {id:2}, {id:3}]

intersect

Returns only items that are present in all connected branches, matched by mergeKey.

Input A: [{id:1}, {id:2}, {id:3}]
Input B: [{id:2}, {id:3}, {id:4}]
Output: [{id:2}, {id:3}]

Use this to find common records across datasets.

except

Returns items from the first branch that do NOT appear in any subsequent branch, matched by mergeKey.

Input A (base): [{id:1}, {id:2}, {id:3}]
Input B (exclude): [{id:2}]
Output: [{id:1}, {id:3}]

Use this to subtract one dataset from another.

join

Merges records with the same key from all branches into a single enriched record (inner join semantics — only records present in all branches are kept).

Input A: [{id:1, name:"Alice"}]
Input B: [{id:1, score:99}]
Output (mergeKey: id): [{id:1, name:"Alice", score:99}]

Merge Key

For union, intersect, except, and join strategies, configure the key field used to match records across branches. Defaults to id.

Note: even if upstream field inference is incomplete in the editor (e.g. you only see msg in a dropdown), you can still set mergeKey to a nested record key such as id — the merge strategies operate on the items inside each branch’s data array.

Behavior

When multiple upstream branches connect to a Merge node:

1. The engine collects outputs from all connected branches
2. Each branch's output data array is extracted
3. The configured merge strategy is applied
4. The merged result is wrapped in { "data": [...], "count": N } and passed downstream

Input shape

Merge expects to receive multiple upstream payloads. If an upstream payload has a data array (Data Model–style), Merge uses that array. Otherwise, the full payload is treated as a single record.

Output shape

The node returns a Data Model–style payload:

• data: merged array of records
• count: number of merged records
• merge_strategy: applied strategy
• input_count: number of connected inputs

Example: Re-Unify After If Branch

💾 Try this example

Download the workflow JSON — open an empty workflow and import the file.

Notes

• The Merge node requires at least two input connections to be meaningful; a single input just passes data through
• The concat strategy works with any data shape; the others require a consistent key field across branches
• Variables and workflow-level context are preserved from all branches (later branches override earlier for duplicate keys)
• Use intersect/except for set operations on database result arrays