Complete dataset registration

PUT 
/datasets/:id

Completes the registration of a dataset by creating its backend infrastructure (Glue table, Redshift table, DynamoDB table, or Iceberg/Hudi/DeltaLake table) based on TargetLocation. The caller must be an owner of the dataset. The dataset's RegistrationStatus must NOT already be "completed" — calling this on a fully registered dataset returns GE-1034.

Required request body field: DatasetSchema (array of column definitions). Other fields are conditionally required based on the dataset's TargetLocation:

• Redshift: SortType, DistType, SortKeys, DistKey, PrimaryKeys, UniqueColumns, IsIdentityNeeded, IdentityAttributeValues, DataConversionParams

• DynamoDB: PartitionKey (required), SortKey (optional), StorageType, LocalSecondaryIndexes, GlobalSecondaryIndexes

• LF/S3Athena (Iceberg): handled via IcebergTableOptions on the dataset record

• LF/S3Athena (Hudi): PrimaryKey, PrecombineKey

• All targets with TableUpdate=update: RecordKeys + LatestRecordIndicator required

On success, asynchronous operations are triggered (LF access grants, DQ constraints, NL2SQL schema sync, observability etc).

Request

Responses

200

Registration completed successfully

Response Headers

Access-Control-Allow-Origin
Access-Control-Allow-Methods
Access-Control-Allow-Headers

400

Invalid input or dataset state

Response Headers

Access-Control-Allow-Origin
Access-Control-Allow-Methods
Access-Control-Allow-Headers

500

Internal failure during registration

Response Headers

Access-Control-Allow-Origin
Access-Control-Allow-Methods
Access-Control-Allow-Headers