Neo4j Output

Description

The Neo4j Output transform allows you to do high performance updates in one node, two nodes or two nodes and a relationship.

The transform generates the required Cypher statements with accompanying parameters.

Supported Engines

Hop Engine

Supported

Spark

Supported

Flink

Supported

Dataflow

Supported

Usage

You can update one node or two nodes and a relationship. Nodes and relationships can have properties and the appropriate MERGE statements will be generated based on the information you provided. Make use of the "Get fields" buttons on the right hand side of the dialog to prevent you from having to type too much.

Labels

You can get one node labels from a list of fields or you can use static values (with optionally variables). If you specify both the static value will be used if the field value is null. A label will not be used for the node if the label is null or empty.

Array Properties

The Neo4j Output transform supports Array type properties for nodes and relationships. When a property is configured with type "Array", the transform will parse the input string value and convert it to a Neo4j array.

Array properties require two configuration options:

Array separator: Character used to separate array elements (default: comma ,) Examples: , (comma), ; (semicolon), | (pipe)
Array enclosure: Character used to enclose each array element (default: empty) Examples: " (double quotes), ' (single quotes), empty for no enclosure

Array values should be provided as strings in the input data, e.g., "1.0,2.0,3.0" or "tag1,tag2,tag3". The transform will automatically split and convert these to Neo4j array types based on the separator and enclosure settings.

Notable options

Option Description

Option	Description
Batch size	for better performance the transform will group records together in a single transaction
Use CREATE instead of MERGE	this bypasses any lookup and is faster.
Create indexes	creates unique field constraints for all Primary properties in the nodes to make sure performance stays good when dealing with merges and lookup of nodes for relationships. NOTE: Please consider only using this when doing prototyping and testing. It’s better to put the index creation statements explicitly in a Cypher Script action to make sure you never forget it even when moving to a new database.
Array separator	comma (`,`)
Character used to separate array elements when property type is Array	Array enclosure
empty	Character used to enclose each array element when property type is Array. Leave empty for no enclosure.

Batch size

for better performance the transform will group records together in a single transaction

Use CREATE instead of MERGE

this bypasses any lookup and is faster.

Create indexes

creates unique field constraints for all Primary properties in the nodes to make sure performance stays good when dealing with merges and lookup of nodes for relationships. NOTE: Please consider only using this when doing prototyping and testing. It’s better to put the index creation statements explicitly in a Cypher Script action to make sure you never forget it even when moving to a new database.

Array separator

comma (,)

Character used to separate array elements when property type is Array

Array enclosure

empty

Character used to enclose each array element when property type is Array. Leave empty for no enclosure.

Show Cypher Preview

The transform dialog includes a "Show Cypher" button that displays a preview of the generated Cypher statements without executing them. This preview shows:

The UNWIND-based batch processing structure
Node creation/merge statements with labels and properties
Relationship creation/merge statements
Property mappings and types

The preview is read-only and helps you verify the Cypher that will be generated before running the pipeline.