Data Catalog

Live: https://fair-forward.github.io/datasets/

Quick start (local)

python -m venv .venv && source .venv/bin/activate    # Windows: .venv\Scripts\activate
pip install -r requirements.txt

Front-end dev server (Vite):

npm install      # one-time per clone (or after deleting node_modules)
npm run dev      # starts Vite (uses node_modules/.bin/vite)

Create a local .env (gitignored):

GOOGLE_APPLICATION_CREDENTIALS=./data_sources/google_sheets_api/service_account_JN.json
PEXELS_API_KEY=your_pexels_key   # optional, for placeholder images

Place your service account JSON at that path (do not commit it).

Build end-to-end from Google Sheets:

python scripts/build_from_google_sheets.py

Rebuild from existing docs/data_catalog.xlsx (no Sheets fetch):

python scripts/build.py

Contribute data

1. Add or update rows in the source Google Sheet: https://docs.google.com/spreadsheets/d/18sgZgPGZuZjeBTHrmbr1Ra7mx8vSToUqnx8vCjhIp0c/edit?gid=561894456#gid=561894456
2. Run the build locally (above) or trigger the GitHub Action “Manually Update Website from Google Sheets” on branch main.

Secrets & safety

• Service account keys stay local (ignored by git). Never commit JSON keys.
• If a key was ever committed, rotate it in Google Cloud and update your local .env path.
• Pexels API key is only needed for scripts/download_placeholder_images.py.

Scripts

• scripts/build_from_google_sheets.py — fetch Sheet, generate markdown/JSON, run Vite build to docs/.
• scripts/build.py — regenerate JSON + Vite build from existing docs/data_catalog.xlsx.
• scripts/download_placeholder_images.py — optional placeholders (needs PEXELS_API_KEY).

Deploy

docs/ is the static output (Vite). GitHub Pages serves from docs/. Running the build scripts updates it. The GitHub Action “Manually Update Website from Google Sheets” can do this without local setup.

Contributing (code)

PRs welcome; follow existing style and keep secrets out of git.