Construct Overview

The Data Ops Job CDK L3 construct is used to deploy the resources required to support and perform data operations on top of a Data Lake using Glue Jobs.

Deployed Resources

• Glue Jobs - Glue Jobs will be created for each job specification in the configs

Optional Features

Asset Script Resolution

Set assetBasePath on the construct props to enable asset: prefixed paths in scriptLocation and additionalScripts. When a path starts with asset:, it is resolved relative to the configured base path (e.g., asset:dq-main.py resolves to <assetBasePath>/dq-main.py).

DataZone Data Lineage

Set the optional lineage section on a job config (enabled, domainId, optional accountId, optional manageExecutionRolePolicy) to emit OpenLineage events to an Amazon DataZone / SageMaker Unified Studio domain. When enabled, the construct injects the OpenLineage Spark listener configuration into the job's --conf argument, preserving any user-supplied --conf. Requires Glue version 5.0 or higher (enforced with a synth-time error). When enabled, the construct attaches a managed policy granting datazone:PostLineageEvent to the job execution role, scoped to the domain ARN in the deploying account; set manageExecutionRolePolicy: false to skip the attachment when the role is externally managed. The optional accountId sets only the Glue Data Catalog account (spark.glue.accountId) and must be a 12-digit AWS account ID.