[Question] A question about the future of this Databricks CLI
Closed this issue · 2 comments
I am using the Python Databricks CLI since two years and It simplifies the way I am interacting with REST APIs.
However, in the last months you have created a Python SDK and a new (and more complete) CLI for the REST APIs.
So, my question is: what is the planned future of the Python Databricks CLI?
The Python-based CLI (this repo) will not receive new feature work but will continue to receive critical bug fixes and maintenance. We might cut a 0.18 version to drop Python 2 support but otherwise, nothing new is planned here.
The new CLI is built from an API specification and because of this any updates for the API are automatically propagated into the CLI, as opposed to having to manually mirror all changes.
For users who have been using this repository as a sort-of Python SDK (by importing databricks_cli.sdk), we have built a proper Python SDK with API coverage generated similar to the new CLI.
Both the new CLI and SDKs support all authentication mechanisms supported by the platform instead of only personal access token (PAT) based auth in this repository. You can read more about this in the SDK README or in our documentation (differs per cloud with authentication options are available):
Thanks for raising this issue. We're actively working on the new CLI, new SDKs, and more, and hadn't taken the time yet to add a notice to this repository. I have updated the README and added a pinned issue to address this.
If you have feedback or experience issues with either the new CLI or SDKs, please let us know!