Summary
Compute cluster management APIs now support a DependencyMode field, alongside new configuration fields for AI runtime tasks, ML stream schemas, and Postgres synced tables. Additionally, a breaking change removes the
Summary generated by brickster.ai. For the full changelog and any code/binary attachments, follow the GitHub link above.
Topics
More from databricks/databricks-sdk-go
The bundle deployments service introduces breaking changes by reordering arguments in CreateDeployment and removing the DeploymentId field from CreateDeploymentRequest. The catalog service adds new owner enum values to SpecialDestination for Unity Catalog schemas, tables, volumes, functions, and registered models.
This release adds the EffectiveEntitlements field to workspace assignment details in the IAM v2 SDK service. It also introduces the ServerlessComputeId field to job cluster definitions within the Jobs service.
API endpoint paths changed for IAM v2 account and workspace assignment methods, ML window duration fields are now optional, and the LongRolling field was removed from ML time windows. New API fields were added across jobs
Version 0.160.0 of the Databricks Go SDK adds support for JAR analysis assets, asset reviews, and task run listing within Databricks Clean Rooms. Other updates include support for JDBC catalog connection types, Azure instance pool capacity reservation groups, and bundle deployment parent paths.
The Go SDK update adds the EffectiveWorkspaceId field to disaster recovery stable URLs and the SourceMetadataColumn field to pipeline table configurations. It also introduces a breaking change by removing the CodeSourcePath field from AI runtime job tasks.
The Databricks Go SDK updated Clean Rooms API models to support shared output configurations, environment versions, and notebook asset descriptions. It also added custom report options for Google Ads, Meta Marketing, and TikTok Ads pipeline ingestion sources.