Closed nBhabish closed 1 year ago
When authenticating on Databricks or any other environment where you're in a shell, you have to use the device_code flow.
get_business_outlook(app="xxx", auth_type="device_code")
@hongooi73 if you don't mind me asking, what exactly is a device_code? Also, where can I find it/ how do I find it?
@hongooi73 how do I get the azure login creds to be saved in Databricks? Every time the script is scheduled to run it will eventually come to a point where I'll have to go and authenticate it manually. I was wondering if there's a way for it to be a one-time thing in Databricks if that makes sense. It says the following: AzureR data directory does not exist; login credentials not saved
You can create the directory it's looking for with dir.create(AzureAuth::AzureR_dir())
. I'm not that familiar with Databricks but I assume the node you're running this on is persistent, so the dir should stick around between sessions.
@hongooi73 I think this might help me a great deal. I was able to create a custom app registration and use create_graph_login to authenticate to Azure in an unattended script to automate sending emails. However, I can't pull my outlook credentials using get_business_outlook() in that unattended script. I was wondering if you could help me figure out a way to pull my outlook credentials after the graph login.
Technically you can use auth_type="resource_owner"
to authenticate a user account in an unattended script. See the vignettes for the AzureAuth package. This is discouraged however, and your org may also have disabled it.
More info: https://learn.microsoft.com/en-au/azure/active-directory/develop/v2-oauth-ropc
I am trying to automate one of my R scripts that will be sending out weekly emails to my external stakeholders, but I am struggling with the authentication piece. I was wondering if someone could help me with this