Summary
The Databricks Java SDK adds Kinesis stream configuration support for ML stream sources and an update mode field for pipelines. Breaking changes convert bundle deployment state fields to strings and remove direct cloud credential fields from Unity Catalog model provider configurations.
Summary generated by brickster.ai. For the full changelog and any code/binary attachments, follow the GitHub link above.
More from databricks/databricks-sdk-java
This release removes the includeBrowse request field and browseOnly response field across several model and MCP service APIs in Unity Catalog. It also adds inputColumn and outputColumn fields to pipeline transformers and introduces a NetSuite catalog connection type.
The Java SDK now stops retrying consumed streaming requests like file uploads to prevent writing empty files, surfacing the original error to the caller instead. Additionally, the Billing and Unity Catalog services received new API fields and enum values for managing budget alerts, scope types, and policy grants.
* Bump Jackson and Apache Commons Configuration dependencies to address security vulnerabilities.
The Databricks SDK for Java introduces breaking changes to IAM v2 models by replacing the internalId field with dedicated groupId, servicePrincipalId, and userId fields. This update also adds a pauseStatus field to job trigger states and a directoryInfo field to workspace object metadata.