This repository contains the basic prerequisites to get started with Observable Framework on Github Codespaces, using micromamba as Python package manager running on Alpine Linux.

1. Sign in to your GitHub account.
2. Open https://github.com/dleeftink/observable-codespace (this repository).
3. Select Use this template > Open in a codespace.

Wait for the codespace to be built and the setup scripts to finish running (this may take some time).

• After succesfully setting up the environment, codespaces should have automatically opened index.md and penguins.csv.py to the left and a live preview pane to the right.
• Additionally, a popup might open asking you to trust the live server destination.

You can edit the index.md contents, after which the content changes are reflected (with a slight delay) in the preview pane.

• Navigate to penguin-classification/docs/data/penguins.csv.py and uncomment X = df.iloc[ ... ] to observe how changes to the Python data-loader are propagated to the live-preview.

If updates stop being reflected in the preview pane, refresh the page. If updates aren't propagated after refreshing, execute CTRL/CMD + C in the terminal after which you can restart the live server using npm run obs.

• If you accidentally close the preview pane, you can re-open this by navigating to the Ports tab on the bottom half of the screen and hovering the Forwarded Address. Click the Preview in Editor icon that shows on hover to re-open the preview inside your codespace.
• For more information on getting started, visit the Observable Framework Documentation.

You can create new dashboards ('sub-repositories') as follows:

1. Open the Terminal tab and cd to the /workspace/observable-python-codespace directory if not already there.

2. Run the npm init @observablehq command.

   • If this is your first time running this command, you will be prompted to install create. Type y.

3. Follow the CLI to create the sub-reposistory and additional template files.