Is there a way to programatically retrieve a workspace name ?

[AmarK]

Is there a spark command in databricks that will tell me what databricks workspace I am using? I’d like to parameterise my code so that I can update delta lake file paths automatically depending on the workspace (i.e. it picks up the dev workspace name when in dev and will pick up the prod workspace name in prod). Is this possible?

[Navya_R]

You may run the below in a notebook within the workspace:

dbutils.entry_point.getDbutils().notebook().getContext().toJson()

The output contains: 

browserHostName - is the workspace name

orgId - is the workspaceID

[AmarK]

Thanks Navya! But this doesn't work for me on a High Concurrency cluster. It seems that toJson() isn't whitelisted.

[mehdi_lamrani]

from databricks.sdk import WorkspaceClient

workspace_client = WorkspaceClient()

workspace_hostname = workspace_client.config.host

[SebastianRowan]

How does it even make sense? Did you even see how old the post is? It's confusing! What made you reply to this post, wondering!

[WiliamRosa]

To programmatically retrieve the Databricks workspace name from within a notebook, you can use Spark configuration or the notebook context. One method is to read the workspace URL using spark.conf.get("spark.databricks.workspaceUrl") and then extract the instance name—for example, by splitting at the first period. Alternatively, you can access the notebook context with dbutils.notebook.entry_point.getDbutils().notebook().getContext().browserHostName().toString() (or via the equivalent JSON tags), which provides the hostname corresponding to the workspace. Both approaches give you a way to dynamically capture and use the workspace identifier in your code.