Get Started with mx-bluesky#

Development Environment#

We recommend using devcontainers with vscode as the IDE when developing using a DLS machine. Here’s how you can set this up:

  • Clone this repo using SSH: git clone git@github.com:DiamondLightSource/mx-bluesky.git.

  • At the same directory level as you were in for step 1, clone dodal using SSH: git clone git@github.com:DiamondLightSource/dodal.git. You should now have a directory structure looking like/some_path/mx-bluesky and /some_path/dodal.

  • In a terminal, move to the mx-bluesky directory and open vscode:

cd /some_path/mx-bluesky  
module load vscode  
code .

  • Make sure you have the devcontainers extension installed in vscode.

  • In vscode, press shift+ctrl+p, and type “Dev Containers: Open Folder in Container”, then select the mx-bluesky folder.

  • Once this has built, in vscode at the top right, click File -> Open workspace from file, then select mx-bluesky/.vscode/mx-bluesky-dev-container.code-workspace.

  • This should get dodal and mx-bluesky opened up in your workspace with all the correct settings. To prove to yourself that the environment is correct, type tox -e tests in a terminal in your devcontainer while inside the dodal and inside the mx-bluesky directory.

Notes#

  • The devcontainer initially takes quite a long time to build. It will be much faster every other time, and only needs to be rebuild if we change the container environment.

  • The first time you make a commit after building, you will need to follow some git prompts to set your git name and email.

  • The old dls_dev_env.sh script creates a virtual environment in mx-bluesky/.venv, while the devcontainer creates one in mx-bluesky/venv. This is the only reason why we need separate code-workspace files.

Supported Python versions#

As a standard for the python versions to support, we are using the numpy deprecation policy

Currently supported versions are: 3.11, 3.12.