Ngulia Website

Standalone GitHub Pages repository for the core Ngulia website.

What is in this repo

• src/: Vue app code
• public/generated/: committed processed website data
• public/history-archive/: curated historical images used by the site
• public/partner-logos/: partner assets used by the site
• scripts/: sync and helper scripts for the website repo

What is not in this repo

Raw research inputs and the website data pipeline stay outside the repo. In particular, the large ringing spreadsheets, recovery workbook, BirdLife geodatabase, source photo archive, and the R export script are not part of this website repository.

Environment split

Public website config goes in .env.local:

VITE_MAPBOX_TOKEN=...

Private pipeline config goes in .env.pipeline.local:

NGULIA_PROJECT_DATA_DIR=/absolute/path/to/Ngulia/data
GOOGLE_MAPS_API_KEY=...

Only the VITE_ token is used by the frontend build. The Google key is only used by scripts/geocode_recovery_sites.py. NGULIA_PROJECT_DATA_DIR should point to the research workspace data/ directory so scripts/preprocess.py can copy the prepared website export from derived/website/.

Local development

npm install
npm run dev

Build

Build the website from the committed processed data:

npm run build

Regenerate processed data first, then build:

Rscript /absolute/path/to/Ngulia/scripts/07_build_curated_recoveries.R
Rscript /absolute/path/to/Ngulia/scripts/08_build_website_data.R
npm run preprocess
npm run build

Or in one step:

npm run build:full

Preprocessing notes

The website no longer builds its data payload from raw research inputs.

The research workspace now prepares website data in two steps:

Rscript /absolute/path/to/Ngulia/scripts/07_build_curated_recoveries.R
Rscript /absolute/path/to/Ngulia/scripts/08_build_website_data.R

The website export step reads only:

• data/curated/daily_counts.csv
• data/curated/recoveries.csv
• taxonomy files in data/staging/website/config/, data/staging/ring_events/config/, and data/reference/taxonomy/

It writes the website-ready files to:

• NGULIA_PROJECT_DATA_DIR/derived/website/

Then this repo only copies that directory into:

• public/generated/

The sync script expects these paths to exist in derived/website/:

• dashboard.json
• recoveries.json
• migration-probabilities.json
• species-ranges-index.json

Static publications data and species photos now live in this repo and are not part of the generated data sync.