// Generates deterministic demo data that follows CONTRACT.md. // Run: npm run mock // The output is checked in so the app renders before the Go API exists. import { mkdir, writeFile } from 'node:fs/promises' import { dirname, resolve } from 'node:path' import { fileURLToPath } from 'node:url' const root = resolve(dirname(fileURLToPath(import.meta.url)), '..') const out = (path) => resolve(root, 'mock', path) const BERLIN = 'Europe/Berlin' const DEMO_NOW = '2026-11-14T08:12:00Z' // 09:12 in Heilbronn const PLAN_START = '2026-11-15T23:00:00Z' // Mon 16 Nov 00:00 local const PLAN_END = '2026-11-22T23:00:00Z' // Mon 23 Nov 00:00 local const DATA_VERSION = '2026-11-14T03:00:00Z' const MODEL_VERSION = 'histgb-2026-11-14' const CONTEXT_VERSION = '2026-11-14T03:00:00Z' const sensors = [ { id: 'hn-stadtgalerie-01', name: 'Stadtgalerie', lat: 49.1427, lon: 9.2109, factor: 1.15 }, { id: 'hn-kaiserstrasse-02', name: 'Kaiserstraße', lat: 49.1434, lon: 9.2142, factor: 1.05 }, { id: 'hn-marktplatz-03', name: 'Marktplatz', lat: 49.1414, lon: 9.2163, factor: 1.0 }, { id: 'hn-kilianskirche-04', name: 'Kilianskirche', lat: 49.1418, lon: 9.2188, factor: 0.88 }, { id: 'hn-rathaus-05', name: 'Rathaus', lat: 49.1407, lon: 9.2196, factor: 0.82 }, { id: 'hn-neckartor-06', name: 'Neckartor', lat: 49.1389, lon: 9.2172, factor: 0.9 }, { id: 'hn-suelmerstrasse-07', name: 'Sülmerstraße', lat: 49.1451, lon: 9.2158, factor: 1.08 }, { id: 'hn-harmonie-08', name: 'Harmonie', lat: 49.1397, lon: 9.213, factor: 0.76 }, ] // Local weather for 14 Nov (now) through 22 Nov. Deterministic demo context. const weatherByDate = { '2026-11-14': { temp: 10.2, precip: 0.2, cloud: 55, wind: 14, humidity: 74 }, '2026-11-15': { temp: 8.6, precip: 0.4, cloud: 70, wind: 12, humidity: 78 }, '2026-11-16': { temp: 7.0, precip: 0.8, cloud: 85, wind: 17, humidity: 82 }, '2026-11-17': { temp: 5.4, precip: 0.2, cloud: 75, wind: 11, humidity: 80 }, '2026-11-18': { temp: 6.2, precip: 0.0, cloud: 60, wind: 9, humidity: 72 }, '2026-11-19': { temp: 8.8, precip: 2.4, cloud: 90, wind: 20, humidity: 86 }, '2026-11-20': { temp: 9.6, precip: 0.6, cloud: 80, wind: 16, humidity: 84 }, '2026-11-21': { temp: 12.4, precip: 0.0, cloud: 25, wind: 8, humidity: 65 }, '2026-11-22': { temp: 8.1, precip: 0.0, cloud: 50, wind: 10, humidity: 71 }, } const hourShape = [ 0.12, 0.08, 0.06, 0.05, 0.05, 0.07, 0.12, 0.18, 0.25, 0.32, 0.55, 0.78, 0.95, 1.0, 0.88, 0.82, 0.86, 1.02, 1.12, 0.72, 0.48, 0.35, 0.24, 0.16, ] // Monday = 0 ... Sunday = 6 const dayFactor = [0.78, 0.84, 0.9, 0.96, 1.08, 1.38, 0.7] function mulberry32(seed) { let a = seed return function () { a |= 0 a = (a + 0x6d2b79f5) | 0 let t = Math.imul(a ^ (a >>> 15), 1 | a) t = (t + Math.imul(t ^ (t >>> 7), 61 | t)) ^ t return ((t ^ (t >>> 14)) >>> 0) / 4294967296 } } const partsFormatter = new Intl.DateTimeFormat('en-CA', { timeZone: BERLIN, year: 'numeric', month: '2-digit', day: '2-digit', hour: '2-digit', hourCycle: 'h23', }) function localParts(iso) { const parts = Object.fromEntries( partsFormatter.formatToParts(new Date(iso)).map((p) => [p.type, p.value]), ) return { date: `${parts.year}-${parts.month}-${parts.day}`, hour: Number(parts.hour), weekday: ((new Date(`${parts.year}-${parts.month}-${parts.day}T12:00:00Z`).getUTCDay() + 6) % 7), } } function weatherFactor(date, hour, weekday) { const w = weatherByDate[date] ?? weatherByDate['2026-11-16'] let factor = 1 if (w.precip >= 1) factor *= 0.86 else if (w.precip > 0) factor *= 0.95 if (w.cloud < 35 && (weekday === 5 || weekday === 6)) factor *= 1.08 if (w.temp < 6) factor *= 0.95 if (w.temp > 11) factor *= 1.03 return factor } function footfallAt(sensorIndex, iso) { const { date, hour, weekday } = localParts(iso) const rand = mulberry32(1000 + sensorIndex * 7919 + Math.floor(new Date(iso).getTime() / 3_600_000)) const noise = 0.92 + rand() * 0.16 const v = 110 * dayFactor[weekday] * hourShape[hour] * sensors[sensorIndex].factor * weatherFactor(date, hour, weekday) * noise return Math.max(4, Math.round(v)) } function pointsFor(sensorIndex, fromIso, hours) { const start = new Date(fromIso).getTime() return Array.from({ length: hours }, (_, i) => { const t = new Date(start + i * 3_600_000).toISOString().replace('.000Z', 'Z') const v = footfallAt(sensorIndex, t) return { t, v, lo: Math.max(0, Math.round(v * 0.82)), hi: Math.round(v * 1.18), } }) } function meanOf(points) { return Math.round((points.reduce((sum, p) => sum + p.v, 0) / points.length) * 10) / 10 } function p95Of(points) { const sorted = points.map((p) => p.v).sort((a, b) => a - b) return sorted[Math.floor(sorted.length * 0.95)] ?? 0 } function ring(points) { return points.map(({ t, v }) => ({ t, v })) } function weatherAt(iso) { const { date, hour } = localParts(iso) const w = weatherByDate[date] ?? weatherByDate['2026-11-16'] const temp = Math.round((w.temp + 1.8 * Math.sin(((hour - 9) / 24) * Math.PI * 2)) * 10) / 10 const rainHours = w.precip > 0 ? (hour >= 6 && hour <= 9) || (hour >= 15 && hour <= 17) : false const precip = rainHours ? Math.round((w.precip / 4) * 10) / 10 : 0 return { t: iso, temp_c: temp, precip_mm: precip, cloud_pct: w.cloud, wind_kmh: w.wind, humidity_pct: w.humidity, } } function isoHour(fromIso, offset) { return new Date(new Date(fromIso).getTime() + offset * 3_600_000).toISOString().replace('.000Z', 'Z') } await mkdir(out('forecast'), { recursive: true }) await mkdir(out('series'), { recursive: true }) const seriesBySensor = {} for (const [index, sensor] of sensors.entries()) { const series = pointsFor(index, '2026-11-07T00:00:00Z', 177) seriesBySensor[sensor.id] = series await writeFile( out(`series/${sensor.id}.json`), JSON.stringify({ sensor: sensor.id, step: '1h', points: ring(series) }, null, 2) + '\n', ) const forecast = pointsFor(index, PLAN_START, 168) await writeFile( out(`forecast/${sensor.id}.json`), JSON.stringify({ sensor: sensor.id, horizon: '7d', scenario: 'base', points: forecast, cached: false, data_version: DATA_VERSION, model_version: MODEL_VERSION, context_version: CONTEXT_VERSION, }, null, 2) + '\n', ) sensor.mean = meanOf(series) sensor.p95 = p95Of(series) sensor.battery = 61 + ((index * 13) % 38) delete sensor.factor } await writeFile( out('sensors.json'), JSON.stringify(sensors.map(({ id, name, lat, lon, mean, p95, battery }) => ({ id, name, lat, lon, mean, p95, battery, })), null, 2) + '\n', ) const weatherForecast = Array.from({ length: 168 }, (_, i) => weatherAt(isoHour('2026-11-14T09:00:00Z', i))) await writeFile( out('context.json'), JSON.stringify({ weather: { now: weatherAt('2026-11-14T08:00:00Z'), forecast: weatherForecast, }, holidays: [ { date: '2026-01-01', name: 'Neujahr' }, { date: '2026-01-06', name: 'Heilige Drei Könige' }, { date: '2026-04-03', name: 'Karfreitag' }, { date: '2026-04-06', name: 'Ostermontag' }, { date: '2026-05-01', name: 'Tag der Arbeit' }, { date: '2026-05-14', name: 'Christi Himmelfahrt' }, { date: '2026-05-25', name: 'Pfingstmontag' }, { date: '2026-06-04', name: 'Fronleichnam' }, { date: '2026-10-03', name: 'Tag der Deutschen Einheit' }, { date: '2026-11-01', name: 'Allerheiligen' }, { date: '2026-12-25', name: '1. Weihnachtstag' }, { date: '2026-12-26', name: '2. Weihnachtstag' }, ], updated_at: '2026-11-14T03:00:00Z', cached: false, }, null, 2) + '\n', ) await writeFile( out('meta.json'), JSON.stringify({ data_version: DATA_VERSION, model_version: MODEL_VERSION, context_version: CONTEXT_VERSION, coverage: { sensors: 8, from: '2025-10-16T21:00:00Z', to: '2026-11-07T23:51:00Z', cadence_min: 15, }, metrics: { model: 'weather', mae: 3.36, wape: 27.8, r2: 0.779, baseline_mae: 5.4, baseline_wape: 44.7, }, limits: [ 'estimated devices, not people', '8 sensors in a 580 m cluster', 'observed weather in training; forecast weather in scenarios', ], }, null, 2) + '\n', ) await writeFile( out('scenario.json'), JSON.stringify({ version: 3, scenario: { scenario: 'base' }, at: DEMO_NOW, }, null, 2) + '\n', ) await writeFile( out('scenarios.json'), JSON.stringify([ { id: 'base', label: 'Base week', short: 'Base', description: 'Normal week, latest forecast weather.', weather: { temp_delta_c: 0, precip_mm: 0 }, }, { id: 'sunny-saturday-market', label: 'Sunny Saturday + market', short: 'Sun + market', description: 'Dry, mild Saturday with a market event; assumes +20% Saturday footfall.', weather: { temp_delta_c: 4, precip_mm: 0 }, event: { type: 'market', uplift: 0.2 }, }, { id: 'rainy-saturday', label: 'Rainy Saturday', short: 'Rainy Sat', description: 'Steady rain on Saturday; assumes −18% Saturday footfall.', weather: { temp_delta_c: -2, precip_mm: 3.2 }, }, { id: 'weekday-event', label: 'Wednesday evening event', short: 'Wed event', description: 'Evening event nearby; assumes +25% footfall Wed 16:00–20:00.', weather: { temp_delta_c: 0, precip_mm: 0 }, event: { type: 'concert', uplift: 0.25 }, }, ], null, 2) + '\n', ) await writeFile( out('demo.json'), JSON.stringify({ now: DEMO_NOW, plan_start: PLAN_START, plan_end: PLAN_END, timezone: BERLIN }, null, 2) + '\n', ) console.log('mock data written to mock/')