This sbt plugin attempts to provide an easy to use interface with the Databricks 2.0 rest API for managing artifacts.
masonClusterId- the Databricks cluster ID of the cluster for which you are attempting to manage libraries.masonConfigFile- if your Databricks API key and URL config is stored somewhere other than$HOME/.databrickscfgyou need to set this to that path.masonEnvironmentName- if you would like to use an API key or URL for an environment other than DEFAULT defined in your config file, set that environment name here.masonLibraryName- the name of the project itself is used by default to identify the appropriate published artifacts to use. It is recommended you not set this unless you see error messages asking you to do so.
masonPublishLibrary- this task looks for a published JAR matchingdatabricksLibraryNameand then 1) uploads that artifact to Databricks file storage, 2) adds that uploaded library to the cluster specified withmasonClusterId3) restarts that cluster if it is running to make sure the updated library takes effect.masonRemoveLibrary- this task looks for libraries installed on themasonClusterIdcluster and removes them if they matchmasonLibraryName.
All changes merged to main are automatically released to Maven central as snapshot releases. In
order to make a full release, take the following actions:
-
check out the latest changes on the
mainbranch -
tag the branch with the next appropriate release version and push the tag
git tag -a vX.Y.Z -m "vX.Y.Z" git push origin vX.Y.Z -
go to the releases page and publish the draft release notes