Skip to content

Latest commit

 

History

History
44 lines (28 loc) · 1.97 KB

CONTRIBUTING.md

File metadata and controls

44 lines (28 loc) · 1.97 KB
Raw

Setting up the dev environment

To isolate dependencies from your global python installation, it is important to use a tool like virtualenv. With virtualenv you can install the dev environment by doing the following.

Another option would be to use docker directly, i.e.

docker run -it -v `echo $PWD`:/root python:3.6.8 bash
docker run -it -v `echo $PWD`:/root python:2.7.12 bash
  • pip install -e .
  • pip install -r dev-requirements-py3.txt

To verify that the installation of databricks-cli is the one checked out from VCS, you can check by doing python -c "import databricks_cli; print databricks_cli.__file__".

Developing using VSCode dev containers

This repo comes pre-configured with a devolpment container for the VSCode Remote Containers extension. When opening this project in VSCode you will be asked if you want to open it in a dev container. Click yes and VSCode will build a docker container which everything needed to develop the Databricks CLI and attach VSCode to the container.

Requirements:

  1. VSCode with the remote containers extension installed
  2. A working docker installation

Developing using Github CodeSpaces

The same development container setup used for local VSCode also works with GitHub CodeSpaces. If you have CodeSpaces enabled in your Github account then can just create a CodeSpace from the repoand start coding.

In order to test the CLI against a Databricks cluster you can define the these secrets for your CodeSpace so you don't have to run databricks init eacht time you open it:

  • DATABRICKS_HOST: Workspace URL
  • DATABRICKS_TOKEN: Personal access token

https://docs.github.com/en/codespaces/managing-your-codespaces/managing-encrypted-secrets-for-your-codespaces

Running Tests

  • tox