DatabricksStep - how to get keys or principal within Python script

Dear All,

I am trying to submit a pipeline using a DatabricksStep on databricks cluster following [the official doc](https://github.com/Azure/MachineLearningNotebooks/blob/master//how-to-use-azureml/machine-learning-pipelines/intro-to-pipelines/aml-pipelines-use-databricks-as-compute-target.ipynb)

As far I have seen if we execute a databricks notebook we are able to get "input_blob_secretname" and input_blob_config" or input_adls_clientid input_adls_clientsecret.

What if I need to use a standard local python script ? Is there any way to pass automacilatty credential from an existeing dataset (internal or external to AzureML).

Thanks
K

Start with how-to-use-azureml/machine-learning-pipelines/intro-to-pipelines/aml-pipelines-use-databricks-as-compute-target.ipynb and compare its Databricks notebook credential examples with the standard local Python script case described here. Done means the supported way to pass credentials from an existing dataset, or the limitation, is established and documented.

Written by the indexing model from the issue text.