jupyter-earth

Jupyter meets the Earth: combining research use cases in geosciences with technical developments within the Jupyter and Pangeo ecosystems.

This repository consists of two parts:

1. The Dockerfile that builds into the experimental, cloud-based JupyterHub environment.
2. The source files and notebooks for the Jupyter meets the Earth project website.

Dockerfile

The hub.jupytearth.org-image/ folder hosts the Dockerfile that builds into the base environment used in https://hub.jupytearth.org. This is an experimental environment our team uses to perform various geoscience workflows and co-develop itself based on user feedback / issue reports.

The cloud infrastructure declaration and Helm chart configuration for https://hub.jupytearth.org lives in https://github.com/2i2c-org/pilot-hubs/.

How to work with this Dockerfile?

Whenever a pull request is made modifying the image, a GitHub Workflow will try verify if it can successfully rebuild.

Whenever a pull request is merged modifying the image, a GitHub Workflow will build and push the image and emit a message on how to update https://hub.jupytearth.org to start making use of the new image. In short, it is to visit https://hub.jupytearth.org/services/configurator/ and enter the image name:tag which is described in the executed GitHub Workflow.

The corresponding GitHub Workflow: .github/workflows/build-image.yaml

Project website

The docs/ folder contains source files for the Jupyter meets the Earth project website: https://pangeo-data.github.io/jupyter-earth/. This is a Binder-ready site, and you can execute the example notebooks presented in the documentation interactively using the following Binder link:

How to work with this documentation?

The documentation of this repository is built using the beta version of Jupyter Book. It will be automatically updated any time changes are made to the docs/ folder and pushed to master.

The corresponding GitHub Workflow: .github/workflows/build-doc.yaml

To build the documentation locally...

Run these steps:

• Install the Jupyter Book 2.0 CLI using conda:

  conda env create -f environment.yml
  conda activate jupyter-earth
  

• Build the docs/ folder:

  jupyter-book build docs/
  

This will create a folder in docs/_build/html where you can preview your site. For example, with chrome docs/_build/html/index.html.