nebari-jupyterlite-plugin

A Nebari plugin that deploys JupyterLite - a lightweight Jupyter environment that runs entirely in the browser.

Installation

pip install git+https://github.com/nebari-dev/nebari-jupyterlite-plugin.git

Or for development:

git clone https://github.com/nebari-dev/nebari-jupyterlite-plugin.git
pip install -e nebari-jupyterlite-plugin

Configuration

Add the following to your nebari-config.yaml:

jupyterlite:
  enabled: true
  auth_enabled: true  # Set to false for public access

Configuration Options

Option	Type	Default	Description
enabled	bool	true	Whether to deploy JupyterLite
auth_enabled	bool	true	Whether to require Nebari authentication
content_repo	string	""	Git repository URL for custom notebooks/files
content_branch	string	"main"	Git branch to use for content repository
overrides	dict	{}	Override Kubernetes resource settings

Custom Content

Load notebooks and files from a git repository:

jupyterlite:
  enabled: true
  content_repo: "https://github.com/your-org/notebooks"
  content_branch: "main"

The content repository is cloned at pod startup and built into JupyterLite using jupyter lite build. Files will appear in the JupyterLite file browser.

Installing Packages

JupyterLite uses Pyodide to run Python in the browser. Packages can be installed in two ways:

1. Pyodide packages (numpy, pandas, sympy, xarray, etc.) - Available immediately, just import them
2. Other PyPI packages - Install at runtime with %pip install <package>

# Pyodide packages - just import
import numpy as np
import pandas as pd

# Other packages - install first
%pip install pyfiglet
import pyfiglet

See the list of packages in Pyodide and the JupyterLite packages documentation for more details.

Overrides

You can customize the deployment using the overrides option:

Option	Type	Default	Description
image	string	quay.io/nebari/nebari-jupyterlite-plugin:latest	Docker image to use
resources.cpu_request	string	"100m"	CPU request
resources.memory_request	string	"128Mi"	Memory request
resources.cpu_limit	string	"500m"	CPU limit
resources.memory_limit	string	"256Mi"	Memory limit

Example:

jupyterlite:
  enabled: true
  overrides:
    image: "quay.io/nebari/nebari-jupyterlite-plugin:v0.1.0"
    resources:
      cpu_request: "200m"
      memory_request: "256Mi"
      cpu_limit: "1000m"
      memory_limit: "512Mi"

Usage

After deploying Nebari with this plugin:

1. Run nebari deploy
2. Access JupyterLite at https://<nebari-domain>/jupyterlite/

Features

• Browser-based: JupyterLite runs entirely in the browser with no server-side computation
• Optional Authentication: Can be protected by Nebari's authentication or made public
• Lightweight: Minimal resource requirements as it serves static files
• Quick Access: Ideal for demos, tutorials, or quick experiments

How It Works

This plugin:

1. Deploys JupyterLite as a Kubernetes Deployment using quay.io/nebari/nebari-jupyterlite-plugin
2. Creates a ClusterIP Service to expose the deployment
3. Configures Traefik IngressRoute to route /jupyterlite/ traffic
4. Optionally integrates with Nebari's forward authentication
5. If content_repo is set, an init container clones the repo and builds JupyterLite with the content

Requirements

• Nebari >= 2025.1.1
• Kubernetes cluster with Traefik ingress controller

License

Apache-2.0