Applies to: SQL Server 2017 (14.x) and later Azure SQL Database Azure SQL Managed Instance

Returns the character representation (JSON) of the node ID for a given object ID and graph ID.


NODE_ID_FROM_PARTS ( object_id, graph_id )



An int representing the object ID for the node table.


A bigint value for the graph ID for a node.

Return value

Returns an nvarchar(1000) character representation (JSON) of the node ID. The return value can be NULL if any of the supplied arguments are invalid.


  • The character representation (JSON) of the node ID returned by NODE_ID_FROM_PARTS is an implementation specific detail, and is subject to change.
  • NODE_ID_FROM_PARTS is the only supported way to construct a suitable character representation of the node ID.
  • NODE_ID_FROM_PARTS is useful for bulk inserting of data into a graph table, when the source data has a suitable natural or surrogate key with an integer data type.
  • The value returned from NODE_ID_FROM_PARTS can be used to populate the $node_id column in a node table. It can also be used to populate the $from_id / $to_id columns in an edge table.
  • For NODE_ID_FROM_PARTS to return valid character representation (JSON) of a node ID, the object_id parameter must correspond to an existing node table. The graph_id parameter can be any valid integer, but it need not exist in that node table. If any of these checks fail, NODE_ID_FROM_PARTS returns NULL.


The following example uses the OPENROWSET Bulk Rowset Provider to retrieve the ID and name columns from a CSV file stored on an Azure Storage account. It then uses NODE_ID_FROM_PARTS to create the appropriate character representation of $node_id for eventual (bulk) insert into the Person node table. This transformed data is then (bulk) inserted into the Person node table.

INSERT INTO Person($node_id, ID, [name])
SELECT NODE_ID_FROM_PARTS(OBJECT_ID('Person'), ID) as node_id, ID, [name]
FROM OPENROWSET (BULK 'person_0_0.csv',
    DATA_SOURCE = 'staging_data_source',
    FORMATFILE = 'format-files/person.xml',
    FORMATFILE_DATA_SOURCE = 'format_files_source',
    FIRSTROW = 2) AS staging_data;

See also