Though voila-gallery is deployed as a TLJH plugin, you can develop it locally without needing to install TLJH. You just need docker and Python 3.6+ installed.
-
Create a virtual environment to work in, with your favorite Python virtual environment tool. With Python's built in
venvmodule, you can make a virtual environment with:python3 -m venv . source bin/activate
-
Install development requirements
pip install -r dev-requirements.txt
-
Do a dev install of the
tljh_voila_gallerypackage.pip install -e tljh-voila-gallery
-
Install configurable-http-proxy. This is needed by JupyterHub, and you need nodejs / npm to install it.
npm -g install configurable-http-proxy
-
Build all the images
python3 -m tljh_voila_gallery.build_images
This requires docker be available and usable from your current user without root. If you need to use
sudobefore yourdockercalls, you should usesudo $(which python3)rather than justpython3 -
Start JupyterHub with the testing config
python3 -m jupyterhub -f jupyterhub_config.py
This requires docker be available and usable from your current user without root. If you need to use
sudobefore yourdockercalls, you should usesudo $(which python3)rather than justpython3
This should start a JupyterHub running at https://localhost:8000
with the current code. If you modify the JupyterHub config or the
service code / template, you need to restart JupyterHub to see their
effects.