Barrinha Experience 🌅

A client-side-only web app that scores each upcoming day in Faro, Portugal from 1–5 ★ for how good the environmental conditions are for the target activity.

No backend, no API keys — safe to host on GitHub Pages.

What it scores

Each day gets a weighted score from five factors (equal weight, tunable):

Factor	Best condition	Source
Tide	as low as possible	Open-Meteo Marine (sea_level_height_msl)
Air temp	as high as possible	IPMA city daily forecast
Water temp	as high as possible	IPMA oceanography daily forecast
Wind	as close to 0 as possible	Open-Meteo forecast (wind_speed_10m_max)
Sunset ↔ moonrise gap	as small as possible	SunCalc (computed in-browser)

Scored horizon ≈ 3 days (bounded by IPMA's oceanography files).

Run locally

# from this directory (repo root)
python3 -m http.server 8080
# then open http://localhost:8080

(A plain static server is required — ES modules don't load over file://.)

Deploy (GitHub Pages)

Push to main, then in repo Settings → Pages set source to Deploy from a branch, branch main, folder / (root). The site serves index.html directly.

Structure

index.html        # shell; loads SunCalc then the ES-module app
styles.css
lib/suncalc.js    # vendored (MIT) — astronomy, no network
js/
  config.js       # location ids, endpoints, normalization bands, weights
  fetch.js        # all data fetches + localStorage cache (6 h TTL)
  astronomy.js    # sunset & closest moonrise per date
  normalize.js    # merge sources by date → per-day records
  scoring.js      # pure: (dayRecord, weights) → { subScores, stars }
  render.js       # one card per day: stars + factor breakdown
  main.js         # orchestration
docs/spec.md      # full specification & locked decisions

Tuning

All normalization bands and factor weights live in js/config.js. Scoring is a pure function in js/scoring.js, easy to test and adjust.