Hartsfield

Hartsfield humbly supports UC Berkeley's DataHub.

To run this locally:

Install a python 3.11 venv in the project directory, activate it, and install requirements:

python3.11 -m venv ./venv
source ./venv/bin/activate
pip install -r requirements.txt

Install npm (the Node.js package manager), adjust the version, install the project dependencies, and do the "audit fix" if it complains that you should do so:

https://docs.npmjs.com/downloading-and-installing-node-js-and-npm

npm install -g npm@9.8.1
npm install
npm audit fix

Securely install local configurations with secrets

Put your configurations in a separately encrypted area outside of the project folder, which you will later export to environment variables. Ensure that your uid is in the AUTHORIZED_USERS list within that file.

mkdir /Volumes/XYZ/hartsfield_config

Run one terminal session for the python back end...

source venv/bin/activate
source .env.development
export HARTSFIELD_LOCAL_CONFIGS=/Volumes/XYZ/hartsfield_config
export HARTSFIELD_ENV=development
venv/bin/python application.py

...and another terminal session for the Node.js front end:

source .env.development
export HARTSFIELD_LOCAL_CONFIGS=/Volumes/XYZ/hartsfield_config
export HARTSFIELD_ENV=development
npm run serve-vue

Using the application

Browse to http://localhost:8080/ -- but note that the first access will take up to several minutes as all of the Node.js stuff does its thing! Subsequent access are fast.

A diagram of the intended function of the application: