[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_187":188,"_189":190},"routes/_app.x.$handle.$name",{"_11":12,"_13":14,"_15":16,"_18":19,"_17":-7,"_140":-5,"_20":141,"_22":142,"_143":144,"_145":146,"_25":147,"_24":-5,"_148":-5,"_149":-5,"_150":55,"_151":152,"_153":46,"_154":-5,"_155":-5,"_156":157},"routes/_app.x.$handle.$name.branch.$branch.code.$",{"_11":12,"_13":14,"_15":16,"_17":-7,"_18":19,"_20":21,"_22":23,"_24":-7,"_25":26,"_27":-5,"_28":-5,"_29":-5,"_30":-5,"_31":-5},"handle","jvt012","name","get-high-temp","branchName","main","version","path","","project",{"_35":86,"_13":14,"_47":87,"_88":-5,"_89":55,"_90":91,"_92":91,"_93":-5,"_94":-5,"_95":96,"_97":-5,"_98":55,"_99":-5,"_100":-5,"_101":102,"_103":104,"_105":-5,"_106":-5,"_107":-5,"_108":-5,"_109":110,"_111":110,"_112":113,"_114":113,"_115":113,"_116":110,"_117":110,"_118":119,"_120":121,"_122":123,"_124":55},"tree",[73,74,75],"file","contents",[32,33,34],"upstream","pull","ahead","behind","envVars",{"_35":66,"_13":67,"_38":68,"_40":-5,"_41":42,"_43":69,"_45":46,"_47":70,"_17":49,"_50":63,"_52":71,"_54":55,"_56":-5,"_18":72},{"_35":58,"_13":59,"_38":60,"_40":-5,"_41":42,"_43":61,"_45":46,"_47":62,"_17":49,"_50":63,"_52":64,"_54":55,"_56":-5,"_18":65},{"_35":36,"_13":37,"_38":39,"_40":-5,"_41":42,"_43":44,"_45":46,"_47":48,"_17":49,"_50":51,"_52":53,"_54":55,"_56":-5,"_18":57},"id","e5cc788c-e200-11f0-a6fc-42dde27851f2","scheduled.ts","fileId","71a80db4-e1ff-11f0-9c40-42dde27851f2","parentId","branchId","0990179a-e1f4-11f0-8dec-42dde27851f2","branchVersion",37,"latest",true,"createdAt",["D",1766715393167],7,"fileType","interval","content","import { DateTime } from \"https://esm.sh/luxon@3.5.0\";\nimport { sqlite } from \"https://esm.town/v/std/sqlite\";\n\nconst USER_AGENT = \"rh-temp-forecast (jwa012@ciceroisback.com)\";\n\n// CLI locations to settle (start with SFO)\nconst SITES = [\"SFO\", \"LAX\", \"HOU\", \"SEA\", \"MIA\", \"MDW\", \"AUS\", \"DEN\", \"DFW\"];\n\ntype SqlResult = {\n columns: string[];\n rows: any[][];\n rowsAffected: number;\n lastInsertRowid: any;\n};\n\nfunction rowsToObjects(res: SqlResult): T[] {\n return (res.rows ?? []).map((row) =>\n Object.fromEntries(row.map((v, i) => [res.columns[i], v])) as T\n );\n}\n\nasync function initDb() {\n await sqlite.execute(`\n CREATE TABLE IF NOT EXISTS predictions (\n site TEXT NOT NULL,\n dateLocal TEXT NOT NULL,\n tz TEXT NOT NULL,\n issuedAtUTC TEXT NOT NULL,\n meanHigh REAL NOT NULL,\n sigma REAL NOT NULL,\n componentsJson TEXT,\n PRIMARY KEY (site, dateLocal)\n )\n `);\n\n await sqlite.execute(`\n CREATE TABLE IF NOT EXISTS settlements (\n site TEXT NOT NULL,\n dateLocal TEXT NOT NULL,\n settledHigh INTEGER NOT NULL,\n forecastMean REAL NOT NULL,\n error REAL NOT NULL,\n productId TEXT,\n issuanceDateLocal TEXT,\n settledAtUTC TEXT NOT NULL,\n PRIMARY KEY (site, dateLocal)\n )\n `);\n}\n\nasync function dbGetPrediction(site: string, dateLocal: string) {\n const res = await sqlite.execute({\n sql: `SELECT * FROM predictions WHERE site = ? AND dateLocal = ?`,\n args: [site, dateLocal],\n });\n return rowsToObjects(res as any)[0] ?? null;\n}\n\nasync function dbIsSettled(site: string, dateLocal: string) {\n const res = await sqlite.execute({\n sql:\n `SELECT 1 AS one FROM settlements WHERE site = ? AND dateLocal = ? LIMIT 1`,\n args: [site, dateLocal],\n });\n return (res.rows?.length ?? 0) > 0;\n}\n\nasync function dbSaveSettlement(row: {\n site: string;\n dateLocal: string;\n settledHigh: number;\n forecastMean: number;\n error: number;\n productId?: string;\n issuanceDateLocal?: string;\n settledAtUTC: string;\n}) {\n await sqlite.execute({\n sql: `\n INSERT INTO settlements(site, dateLocal, settledHigh, forecastMean, error, productId, issuanceDateLocal, settledAtUTC)\n VALUES (?, ?, ?, ?, ?, ?, ?, ?)\n ON CONFLICT(site, dateLocal) DO UPDATE SET\n settledHigh=excluded.settledHigh,\n forecastMean=excluded.forecastMean,\n error=excluded.error,\n productId=excluded.productId,\n issuanceDateLocal=excluded.issuanceDateLocal,\n settledAtUTC=excluded.settledAtUTC\n `,\n args: [\n row.site,\n row.dateLocal,\n row.settledHigh,\n row.forecastMean,\n row.error,\n row.productId ?? null,\n row.issuanceDateLocal ?? null,\n row.settledAtUTC,\n ],\n });\n}\n\nasync function fetchJson(url: string) {\n const r = await fetch(url, {\n headers: { \"User-Agent\": USER_AGENT, \"Accept\": \"application/geo+json\" },\n });\n if (!r.ok) throw new Error(`Fetch failed ${r.status} for ${url}`);\n return r.json();\n}\n\n// Parse month name like DECEMBER -> 12\nconst MONTHS: Record = {\n JANUARY: 1,\n FEBRUARY: 2,\n MARCH: 3,\n APRIL: 4,\n MAY: 5,\n JUNE: 6,\n JULY: 7,\n AUGUST: 8,\n SEPTEMBER: 9,\n OCTOBER: 10,\n NOVEMBER: 11,\n DECEMBER: 12,\n};\n\n// Works for the exact CLI you pasted:\n// \"...THE SAN FRANCISCO AIRPORT CLIMATE SUMMARY FOR DECEMBER 19 2025...\"\n// and the table row:\n// \"MAXIMUM 64 12:28 PM ...\"\nfunction parseCliDateAndMax(\n cliText: string,\n): { summaryDateISO?: string; maxF?: number } {\n const t = cliText.replace(/\\r/g, \"\");\n\n // 1) Date\n const m = t.match(/CLIMATE SUMMARY FOR\\s+([A-Z]+)\\s+(\\d{1,2})\\s+(\\d{4})/);\n if (!m) return {};\n const monthName = m[1].toUpperCase();\n const day = Number(m[2]);\n const year = Number(m[3]);\n const month = MONTHS[monthName];\n if (!month || !Number.isFinite(day) || !Number.isFinite(year)) return {};\n\n const summaryDateISO = DateTime.fromObject(\n { year, month, day },\n { zone: \"America/Los_Angeles\" },\n ).toISODate()!;\n\n // 2) Max temperature (TODAY / MAXIMUM)\n // Narrow to the TEMPERATURE section for robustness.\n const tempSectionMatch = t.match(/TEMPERATURE\\s+\\(F\\)([\\s\\S]*?)\\n\\n/s);\n const tempSection = tempSectionMatch?.[0] ?? t;\n\n const maxMatch = tempSection.match(/^\\s*MAXIMUM\\s+(-?\\d{1,3}|MM)\\b/m);\n if (!maxMatch) return { summaryDateISO };\n\n if (maxMatch[1] === \"MM\") return { summaryDateISO }; // missing\n const maxF = Number(maxMatch[1]);\n if (!Number.isFinite(maxF)) return { summaryDateISO };\n\n return { summaryDateISO, maxF };\n}\n\n// Find a CLI product whose \"CLIMATE SUMMARY FOR \" matches targetLocalISO\nasync function getSettlementHighForDate(\n cliLocation: string,\n targetLocalISO: string,\n) {\n const listUrl =\n `https://api.weather.gov/products/types/CLI/locations/${cliLocation}?limit=30`;\n const listing = await fetchJson(listUrl);\n const items: Array<{ id: string }> = listing[\"@graph\"] ?? [];\n\n for (const it of items) {\n const prod = await fetchJson(`https://api.weather.gov/products/${it.id}`);\n const text: string = prod?.productText;\n if (!text) continue;\n\n const parsed = parseCliDateAndMax(text);\n if (!parsed.summaryDateISO || parsed.maxF === undefined) continue;\n\n if (parsed.summaryDateISO === targetLocalISO) {\n const issuanceTime = prod?.issuanceTime as string | undefined;\n const issuanceDateLocal = issuanceTime\n ? DateTime.fromISO(issuanceTime).setZone(\"America/Los_Angeles\")\n .toISODate()\n : null;\n\n return {\n maxF: parsed.maxF,\n productId: it.id,\n issuanceDateLocal: issuanceDateLocal ?? undefined,\n };\n }\n }\n\n return null;\n}\n\nexport default async function settleCliAndCalibrate() {\n await initDb();\n\n // Settle \"yesterday\" in PT for SFO CLI\n const targetLocalISO = DateTime.now()\n .setZone(\"America/Los_Angeles\")\n .minus({ days: 1 })\n .toISODate()!;\n\n const results: any[] = [];\n\n for (const site of SITES) {\n if (await dbIsSettled(site, targetLocalISO)) {\n results.push({\n site,\n dateLocal: targetLocalISO,\n status: \"already_settled\",\n });\n continue;\n }\n\n const pred = await dbGetPrediction(site, targetLocalISO);\n if (!pred) {\n results.push({\n site,\n dateLocal: targetLocalISO,\n status: \"no_prediction_saved\",\n });\n continue;\n }\n\n const settlement = await getSettlementHighForDate(site, targetLocalISO);\n if (!settlement) {\n results.push({\n site,\n dateLocal: targetLocalISO,\n status: \"cli_not_found_yet\",\n });\n continue;\n }\n\n const settledHigh = Number(settlement.maxF);\n const forecastMean = Number(pred.meanHigh);\n const error = settledHigh - forecastMean;\n\n await dbSaveSettlement({\n site,\n dateLocal: targetLocalISO,\n settledHigh,\n forecastMean,\n error,\n productId: settlement.productId,\n issuanceDateLocal: settlement.issuanceDateLocal,\n settledAtUTC: DateTime.utc().toISO()!,\n });\n\n results.push({\n site,\n dateLocal: targetLocalISO,\n status: \"settled\",\n settledHighF: settledHigh,\n forecastMeanF: Math.round(forecastMean * 10) / 10,\n errorF: Math.round(error * 10) / 10,\n productId: settlement.productId,\n });\n }\n\n return { ok: true, targetLocalISO, results };\n}","deleted",false,"label",[37],"3d5679b8-e201-11f0-9883-42dde27851f2","main.ts","511dd6be-e1ff-11f0-a5d0-42dde27851f2",40,["D",1766715540032],"http","import { DateTime } from \"https://esm.sh/luxon@3.5.0\";\nimport { sqlite } from \"https://esm.town/v/std/sqlite\";\n\nconst USER_AGENT = \"rh-temp-forecast (jwa012@ciceroisback.com)\"; // put real contact\n\n// Robinhood settlement site: CLI location \"SFO\"\nconst SITES: Record<\n string,\n { cliLocation: string; stationId: string; lat: number; lon: number }\n> = {\n LAX: { stationId: \"KLAX\", cliLocation: \"LAX\", lat: 33.9425, lon: -118.4081 },\n HOU: { stationId: \"KHOU\", cliLocation: \"HOU\", lat: 29.6454, lon: -95.2789 },\n SEA: {\n stationId: \"KSEA\",\n cliLocation: \"SEA\",\n lat: 47.4472,\n lon: -122.31361,\n },\n MIA: { stationId: \"KMIA\", cliLocation: \"MIA\", lat: 25.79, lon: -80.32 },\n MDW: { stationId: \"KMDW\", cliLocation: \"MDW\", lat: 41.78, lon: -87.65 },\n AUS: { stationId: \"KAUS\", cliLocation: \"AUS\", lat: 30.18, lon: -97.68 },\n DEN: { stationId: \"KDEN\", cliLocation: \"DEN\", lat: 39.85, lon: -104.66 },\n SFO: {\n stationId: \"KSFO\",\n cliLocation: \"SFO\",\n lat: 37.61961,\n lon: -122.36558,\n },\n DFW: { stationId: \"KDFW\", cliLocation: \"DFW\", lat: 32.9, lon: -97.02 },\n};\n\n// Try a couple independent models. Open-Meteo availability varies; failures are handled.\nconst OPEN_METEO_MODELS = [\"hrrr\", \"gfs_global\"];\n\nfunction clamp(x: number, a: number, b: number) {\n return Math.max(a, Math.min(b, x));\n}\nfunction mean(xs: number[]) {\n return xs.reduce((a, b) => a + b, 0) / Math.max(1, xs.length);\n}\nfunction stddev(xs: number[]) {\n if (xs.length < 2) return NaN;\n const m = mean(xs);\n const v = xs.reduce((acc, x) => acc + (x - m) ** 2, 0) / (xs.length - 1);\n return Math.sqrt(v);\n}\n\n// Normal CDF approximation (sufficient for pricing)\nfunction normCdf(z: number) {\n const t = 1 / (1 + 0.2316419 * Math.abs(z));\n const d = 0.3989423 * Math.exp(-z * z / 2);\n let p = d * t *\n (0.3193815 +\n t * (-0.3565638 +\n t * (1.781478 +\n t * (-1.821256 +\n t * 1.330274))));\n if (z > 0) p = 1 - p;\n return p;\n}\n\ntype SqlResult = {\n columns: string[];\n rows: any[][];\n rowsAffected: number;\n lastInsertRowid: any;\n};\n\nfunction rowsToObjects(res: SqlResult): T[] {\n return (res.rows ?? []).map((row) =>\n Object.fromEntries(row.map((v, i) => [res.columns[i], v])) as T\n );\n}\n\nasync function initDb() {\n await sqlite.execute(`\n CREATE TABLE IF NOT EXISTS predictions (\n site TEXT NOT NULL,\n dateLocal TEXT NOT NULL,\n tz TEXT NOT NULL,\n issuedAtUTC TEXT NOT NULL,\n meanHigh REAL NOT NULL,\n sigma REAL NOT NULL,\n componentsJson TEXT,\n PRIMARY KEY (site, dateLocal)\n )\n `);\n\n await sqlite.execute(`\n CREATE TABLE IF NOT EXISTS settlements (\n site TEXT NOT NULL,\n dateLocal TEXT NOT NULL,\n settledHigh INTEGER NOT NULL,\n forecastMean REAL NOT NULL,\n error REAL NOT NULL,\n productId TEXT,\n issuanceDateLocal TEXT,\n settledAtUTC TEXT NOT NULL,\n PRIMARY KEY (site, dateLocal)\n )\n `);\n\n await sqlite.execute(`\n CREATE INDEX IF NOT EXISTS idx_settlements_site_date\n ON settlements(site, dateLocal)\n `);\n}\n\nasync function dbSavePrediction(row: {\n site: string;\n dateLocal: string;\n tz: string;\n issuedAtUTC: string;\n meanHigh: number;\n sigma: number;\n components: any;\n}) {\n await sqlite.execute({\n sql: `\n INSERT INTO predictions(site, dateLocal, tz, issuedAtUTC, meanHigh, sigma, componentsJson)\n VALUES (?, ?, ?, ?, ?, ?, ?)\n ON CONFLICT(site, dateLocal) DO UPDATE SET\n tz=excluded.tz,\n issuedAtUTC=excluded.issuedAtUTC,\n meanHigh=excluded.meanHigh,\n sigma=excluded.sigma,\n componentsJson=excluded.componentsJson\n `,\n args: [\n row.site,\n row.dateLocal,\n row.tz,\n row.issuedAtUTC,\n row.meanHigh,\n row.sigma,\n JSON.stringify(row.components ?? null),\n ],\n });\n}\n\nasync function dbGetRecentErrors(site: string, limit = 60): Promise {\n const res = await sqlite.execute({\n sql: `\n SELECT error\n FROM settlements\n WHERE site = ?\n ORDER BY dateLocal DESC\n LIMIT ?\n `,\n args: [site, limit],\n });\n const objs = rowsToObjects<{ error: any }>(res as any);\n return objs.map((o) => Number(o.error)).filter((x) => Number.isFinite(x));\n}\n\nfunction calibratedParams(errors: number[]) {\n // Conservative prior until you have real settlement history\n const priorSigma = 3.5; // °F\n const minSigma = 2.0;\n\n if (errors.length < 5) {\n return { muErr: 0, sigma: priorSigma, n: errors.length };\n }\n const mu = mean(errors);\n const sd = stddev(errors);\n\n // mild shrinkage + prevent overconfidence\n const muErr = mu * 0.7;\n const sigmaRaw = Number.isFinite(sd) ? sd : priorSigma;\n const sigma = clamp(\n Math.sqrt(sigmaRaw * sigmaRaw + priorSigma * priorSigma * 0.2),\n minSigma,\n 8,\n );\n\n return { muErr, sigma, n: errors.length };\n}\n\nasync function fetchJson(url: string, accept = \"application/geo+json\") {\n const r = await fetch(url, {\n headers: { \"User-Agent\": USER_AGENT, \"Accept\": accept },\n });\n if (!r.ok) throw new Error(`Fetch failed ${r.status} for ${url}`);\n return r.json();\n}\n\nasync function getNwsHourly(lat: number, lon: number) {\n const points = await fetchJson(\n `https://api.weather.gov/points/${lat},${lon}`,\n );\n const tz = points.properties.timeZone as string;\n const hourlyUrl = points.properties.forecastHourly as string;\n\n const hourly = await fetchJson(hourlyUrl);\n const periods = hourly.properties.periods as Array<\n { startTime: string; temperature: number; temperatureUnit: string }\n >;\n\n return {\n tz,\n hours: periods.map((p) => ({\n t: DateTime.fromISO(p.startTime),\n tempF: p.temperatureUnit === \"F\"\n ? p.temperature\n : (p.temperature * 9) / 5 + 32,\n })),\n };\n}\n\nasync function getLatestNwsObservation(stationId: string) {\n const j = await fetchJson(\n `https://api.weather.gov/stations/${stationId}/observations/latest`,\n );\n const ts = j.properties.timestamp as string;\n const tempC = j.properties.temperature?.value;\n if (!ts || tempC === null || tempC === undefined) return null;\n return {\n obsTimeUTC: DateTime.fromISO(ts).toUTC(),\n tempF: (tempC * 9) / 5 + 32,\n };\n}\n\nfunction highForLocalDate(\n hours: Array<{ t: DateTime; tempF: number }>,\n tz: string,\n localISODate: string,\n) {\n const todays = hours\n .map((h) => ({ ...h, local: h.t.setZone(tz) }))\n .filter((h) => h.local.toISODate() === localISODate);\n if (!todays.length) return null;\n return Math.max(...todays.map((h) => h.tempF));\n}\n\nfunction nearestForecastTemp(\n hours: Array<{ t: DateTime; tempF: number }>,\n whenUTC: DateTime,\n) {\n let best = hours[0];\n let bestAbs = Infinity;\n for (const h of hours) {\n const d = Math.abs(h.t.toUTC().diff(whenUTC).as(\"minutes\"));\n if (d < bestAbs) {\n bestAbs = d;\n best = h;\n }\n }\n return best.tempF;\n}\n\nfunction adjustedHighForToday(\n hours: Array<{ t: DateTime; tempF: number }>,\n tz: string,\n todayISO: string,\n obsUTC: DateTime,\n biasF: number,\n tauHours = 6,\n) {\n const obs = obsUTC.toUTC();\n const todays = hours\n .map((h) => ({ ...h, local: h.t.setZone(tz) }))\n .filter((h) => h.local.toISODate() === todayISO);\n\n let maxAdj = -Infinity;\n for (const h of todays) {\n const leadHours = Math.max(0, h.t.toUTC().diff(obs).as(\"hours\"));\n const w = Math.exp(-leadHours / tauHours);\n const tadj = h.tempF + biasF * w;\n if (tadj > maxAdj) maxAdj = tadj;\n }\n return maxAdj;\n}\n\nasync function getOpenMeteoHourly(\n lat: number,\n lon: number,\n tz: string,\n model: string,\n) {\n const url = new URL(\"https://api.open-meteo.com/v1/forecast\");\n url.searchParams.set(\"latitude\", String(lat));\n url.searchParams.set(\"longitude\", String(lon));\n url.searchParams.set(\"hourly\", \"temperature_2m\");\n url.searchParams.set(\"temperature_unit\", \"fahrenheit\");\n url.searchParams.set(\"timezone\", tz);\n url.searchParams.set(\"forecast_days\", \"2\");\n url.searchParams.set(\"models\", model);\n\n const r = await fetch(url.toString(), {\n headers: { \"User-Agent\": USER_AGENT },\n });\n if (!r.ok) {\n throw new Error(`Open-Meteo failed ${r.status} for model=${model}`);\n }\n const j = await r.json();\n\n const times: string[] = j.hourly?.time ?? [];\n const temps: number[] = j.hourly?.temperature_2m ?? [];\n return times.map((t, i) => ({\n t: DateTime.fromISO(t, { zone: tz }),\n tempF: temps[i],\n }));\n}\n\nasync function modelHighsForToday(\n lat: number,\n lon: number,\n tz: string,\n todayISO: string,\n) {\n const highs: Record = {};\n for (const model of OPEN_METEO_MODELS) {\n try {\n const hrs = await getOpenMeteoHourly(lat, lon, tz, model);\n const todays = hrs.filter((h) => h.t.toISODate() === todayISO);\n if (todays.length) highs[model] = Math.max(...todays.map((h) => h.tempF));\n } catch {\n // model not available / failed; ignore\n }\n }\n return highs;\n}\n\nexport default async function (req: Request): Promise {\n try {\n await initDb();\n\n const url = new URL(req.url);\n const site = (url.searchParams.get(\"site\")).toUpperCase();\n const strike = Number(url.searchParams.get(\"strike\") ?? \"0\");\n\n const cfg = SITES[site];\n if (!cfg) {\n return new Response(\n JSON.stringify({ error: `Unknown site ${site}` }, null, 2),\n { status: 400 },\n );\n }\n\n // Forecast basis: NWS hourly (already strong)\n const { tz, hours } = await getNwsHourly(cfg.lat, cfg.lon);\n const todayISO = DateTime.now().setZone(tz).toISODate()!;\n\n const nwsHigh = highForLocalDate(hours, tz, todayISO);\n if (nwsHigh === null) throw new Error(\"No NWS hourly data for today.\");\n\n // Observation correction: NWS latest station obs\n const obs = await getLatestNwsObservation(cfg.stationId);\n let bias = 0;\n let nwsAdjHigh = nwsHigh;\n if (obs) {\n const tNwsNow = nearestForecastTemp(hours, obs.obsTimeUTC);\n bias = obs.tempF - tNwsNow;\n nwsAdjHigh = adjustedHighForToday(\n hours,\n tz,\n todayISO,\n obs.obsTimeUTC,\n bias,\n 6,\n );\n }\n\n // Independent model signal(s) via Open-Meteo\n const modelHighs = await modelHighsForToday(cfg.lat, cfg.lon, tz, todayISO);\n const modelVals = Object.values(modelHighs).filter((x) =>\n Number.isFinite(x)\n );\n const modelMedian = modelVals.length\n ? modelVals.slice().sort((a, b) =>\n a - b\n )[Math.floor(modelVals.length / 2)]\n : null;\n\n // Blend\n const blended = modelMedian === null\n ? nwsAdjHigh\n : (0.65 * nwsAdjHigh + 0.35 * modelMedian);\n\n // Calibration against settlement: use recent settlement errors for this site\n const errors = await dbGetRecentErrors(site, 60);\n const { muErr, sigma, n } = calibratedParams(errors);\n\n // Mean prediction for settlement high (after bias correction learned from history)\n const meanHigh = blended + muErr;\n\n // Probability for Robinhood “High > strike”\n // Continuity correction: treat integer high; P(H > X) ~ 1 - Φ((X+0.5 - μ)/σ)\n const z = ((strike + 0.5) - meanHigh) / sigma;\n const pYes = 1 - normCdf(z);\n\n // Save today’s prediction (so the scheduled settler can score it tomorrow)\n await dbSavePrediction({\n site,\n dateLocal: todayISO,\n tz,\n issuedAtUTC: DateTime.utc().toISO()!,\n meanHigh,\n sigma,\n components: {\n nwsHigh,\n nwsAdjHigh,\n obsBiasF: bias,\n modelHighs,\n modelMedian,\n blended,\n calibN: n,\n muErr,\n },\n });\n\n const out = {\n site,\n dateLocal: todayISO,\n timeZone: tz,\n\n meanHighF: Math.round(meanHigh * 10) / 10,\n sigmaF: Math.round(sigma * 10) / 10,\n calibN: n,\n\n strikeF: strike,\n pHighGreaterThanStrike: Math.round(pYes * 10000) / 10000,\n fairPrice: Math.round(pYes * 10000) / 10000,\n\n debug: {\n nwsHighF: Math.round(nwsHigh * 10) / 10,\n nwsAdjHighF: Math.round(nwsAdjHigh * 10) / 10,\n obsBiasF: Math.round(bias * 10) / 10,\n modelHighs,\n },\n };\n\n return new Response(JSON.stringify(out, null, 2), {\n headers: { \"content-type\": \"application/json\" },\n });\n } catch (e: any) {\n return new Response(\n JSON.stringify({ error: String(e?.message ?? e) }, null, 2),\n {\n status: 500,\n headers: { \"content-type\": \"application/json\" },\n },\n );\n }\n}",[59],"4e1c47f2-e1ff-11f0-95b2-42dde27851f2","initial-works.ts","09916244-e1f4-11f0-8dec-42dde27851f2",23,["D",1766714709178],"import { DateTime } from \"https://esm.sh/luxon@3.5.0\";\n\nconst USER_AGENT = \"airport-high-forecast (jwa012@ciceroisback.com)\"; // required by NWS\n\n// Keep this list to ~10 airports. Hardcode lat/lon for simplicity & reliability.\nconst AIRPORTS: Record<\n string,\n { icao: string; iata?: string; lat: number; lon: number }\n> = {\n KLAX: { icao: \"KLAX\", iata: \"LAX\", lat: 33.9425, lon: -118.4081 },\n KHOU: { icao: \"KHOU\", iata: \"HOU\", lat: 29.6454, lon: -95.2789 },\n KSEA: { icao: \"KSEA\", iata: \"SEA\", lat: 47.4472, lon: -122.31361 },\n KMIA: { icao: \"KMIA\", iata: \"MIA\", lat: 25.79, lon: -80.32 },\n KMDW: { icao: \"KMDW\", iata: \"MDW\", lat: 41.78, lon: -87.65 },\n KAUS: { icao: \"KAUS\", iata: \"AUS\", lat: 30.18, lon: -97.68 },\n KDEN: { icao: \"KDEN\", iata: \"DEN\", lat: 39.85, lon: -104.66 },\n KSFO: { icao: \"KSFO\", iata: \"SFO\", lat: 37.61961, lon: -122.36558 },\n KDFW: { icao: \"KDFW\", iata: \"DFW\", lat: 32.9, lon: -97.02 },\n};\n\nfunction resolveStation(params: URLSearchParams) {\n const icaoParam = (params.get(\"icao\") || \"\").toUpperCase();\n const iataParam = (params.get(\"iata\") || \"\").toUpperCase();\n\n if (icaoParam && AIRPORTS[icaoParam]) return AIRPORTS[icaoParam];\n\n if (iataParam) {\n const match = Object.values(AIRPORTS).find((a) => a.iata === iataParam);\n if (match) return match;\n }\n return null;\n}\n\nasync function fetchJson(url: string) {\n const r = await fetch(url, {\n headers: { \"User-Agent\": USER_AGENT, \"Accept\": \"application/geo+json\" },\n });\n if (!r.ok) throw new Error(`Fetch failed ${r.status} for ${url}`);\n return r.json();\n}\n\nasync function getNwsHourly(lat: number, lon: number) {\n const points = await fetchJson(\n `https://api.weather.gov/points/${lat},${lon}`,\n );\n const tz = points.properties.timeZone as string;\n const hourlyUrl = points.properties.forecastHourly as string;\n\n const hourly = await fetchJson(hourlyUrl);\n const periods = hourly.properties.periods as Array<\n { startTime: string; temperature: number; temperatureUnit: string }\n >;\n\n // NWS hourly startTime is ISO with offset; Luxon will parse correctly.\n return {\n tz,\n hours: periods.map((p) => ({\n t: DateTime.fromISO(p.startTime),\n tempF: p.temperatureUnit === \"F\"\n ? p.temperature\n : (p.temperature * 9) / 5 + 32,\n })),\n };\n}\n\n// function parseAddsCsv(text: string): Array> {\n// // ADDS returns comment lines starting with #, then a header row.\n// const lines = text.split(\"\\n\").map((l) => l.trim()).filter((l) =>\n// l && !l.startsWith(\"#\")\n// );\n// if (lines.length < 2) return [];\n// const header = lines[0].split(\",\").map((s) => s.trim());\n// return lines.slice(1).map((line) => {\n// const cols = line.split(\",\");\n// const row: Record = {};\n// for (let i = 0; i < header.length; i++) {\n// row[header[i]] = (cols[i] ?? \"\").trim();\n// }\n// return row;\n// });\n// }\n\nasync function getLatestNwsObservation(stationId: string) {\n const url =\n `https://api.weather.gov/stations/${stationId}/observations/latest`;\n\n const r = await fetch(url, {\n headers: {\n \"User-Agent\": USER_AGENT,\n \"Accept\": \"application/geo+json\",\n },\n });\n if (!r.ok) {\n throw new Error(`NWS obs fetch failed ${r.status} for ${stationId}`);\n }\n const j = await r.json();\n\n const ts = j.properties.timestamp as string; // ISO time\n const tempC = j.properties.temperature?.value; // degC, may be null\n\n if (!ts || tempC === null || tempC === undefined) return null;\n\n return {\n obsTime: DateTime.fromISO(ts).toUTC(),\n tempF: (tempC * 9) / 5 + 32,\n raw: j.properties,\n };\n}\n\nfunction todayHighForLocalDate(\n hours: Array<{ t: DateTime; tempF: number }>,\n tz: string,\n localDateISO: string,\n) {\n const todays = hours\n .map((h) => ({ ...h, local: h.t.setZone(tz) }))\n .filter((h) => h.local.toISODate() === localDateISO);\n\n if (!todays.length) return null;\n\n let max = -Infinity;\n let maxTime: DateTime | null = null;\n for (const h of todays) {\n if (h.tempF > max) {\n max = h.tempF;\n maxTime = h.local;\n }\n }\n return { highF: max, atLocal: maxTime! };\n}\n\nfunction nearestForecastTemp(\n hours: Array<{ t: DateTime; tempF: number }>,\n whenUTC: DateTime,\n) {\n let best = hours[0];\n let bestAbs = Infinity;\n for (const h of hours) {\n const d = Math.abs(h.t.toUTC().diff(whenUTC).as(\"minutes\"));\n if (d < bestAbs) {\n bestAbs = d;\n best = h;\n }\n }\n return best.tempF;\n}\n\nfunction adjustedHigh(\n hours: Array<{ t: DateTime; tempF: number }>,\n tz: string,\n localDateISO: string,\n obsUTC: DateTime,\n biasF: number,\n tauHours = 6,\n) {\n const obs = obsUTC.toUTC();\n const todays = hours\n .map((h) => ({ ...h, local: h.t.setZone(tz) }))\n .filter((h) => h.local.toISODate() === localDateISO);\n\n let maxAdj = -Infinity;\n for (const h of todays) {\n const leadHours = Math.max(0, h.t.toUTC().diff(obs).as(\"hours\"));\n const weight = Math.exp(-leadHours / tauHours);\n const tadj = h.tempF + biasF * weight;\n if (tadj > maxAdj) maxAdj = tadj;\n }\n return maxAdj;\n}\n\nfunction clamp(x: number, a: number, b: number) {\n return Math.max(a, Math.min(b, x));\n}\n\nexport default async function (req: Request): Promise {\n try {\n const url = new URL(req.url);\n const airport = resolveStation(url.searchParams);\n if (!airport) {\n return new Response(\n JSON.stringify(\n {\n error: \"Pass eg ?icao=KLAX or ?iata=LAX.\",\n },\n null,\n 2,\n ),\n {\n status: 400,\n headers: { \"content-type\": \"application/json\" },\n },\n );\n }\n\n const { tz, hours } = await getNwsHourly(airport.lat, airport.lon);\n const nowLocal = DateTime.now().setZone(tz);\n const todayISO = nowLocal.toISODate()!;\n\n const base = todayHighForLocalDate(hours, tz, todayISO);\n if (!base) throw new Error(\"No NWS hourly data for today.\");\n\n const metar = await getLatestNwsObservation(airport.icao);\n\n let highAdj = base.highF;\n let bias = 0;\n\n if (metar) {\n const tNwsNow = nearestForecastTemp(hours, metar.obsTime);\n bias = metar.tempF - tNwsNow;\n highAdj = adjustedHigh(hours, tz, todayISO, metar.obsTime, bias, 6);\n }\n\n const range = clamp(\n 2 + 0.6 * Math.abs(highAdj - base.highF) + 0.3 * Math.abs(bias),\n 2,\n 8,\n );\n\n const out = {\n station: airport.icao,\n iata: airport.iata ?? null,\n dateLocal: todayISO,\n timeZone: tz,\n\n forecastHighF: Math.round(highAdj),\n confidencePlusMinusF: Math.round(range),\n\n debug: {\n nwsHighF: Math.round(base.highF),\n nwsHighTimeLocal: base.atLocal.toISO(),\n metarUsed: Boolean(metar),\n biasF: Math.round(bias * 10) / 10,\n metarObsTimeUTC: metar?.obsTime.toISO() ?? null,\n metarTempF: metar ? Math.round(metar.tempF * 10) / 10 : null,\n },\n };\n\n return new Response(JSON.stringify(out, null, 2), {\n headers: { \"content-type\": \"application/json\" },\n });\n } catch (e: any) {\n return new Response(\n JSON.stringify({ error: String(e?.message ?? e) }, null, 2),\n {\n status: 500,\n headers: { \"content-type\": \"application/json\" },\n },\n );\n }\n}",[67],{"_35":66,"_13":67,"_38":68,"_40":-5,"_41":42,"_43":69,"_45":46,"_47":83,"_17":49,"_50":63,"_52":71,"_54":55,"_56":-5,"_77":84,"_18":85},{"_35":58,"_13":59,"_38":60,"_40":-5,"_41":42,"_43":61,"_45":46,"_47":80,"_17":49,"_50":63,"_52":64,"_54":55,"_56":-5,"_77":81,"_18":82},{"_35":36,"_13":37,"_38":39,"_40":-5,"_41":42,"_43":44,"_45":46,"_47":76,"_17":49,"_50":51,"_52":53,"_54":55,"_56":-5,"_77":78,"_18":79},["D",1766715393167],"children",[],[37],["D",1766715540032],[],[59],["D",1766714709178],[],[67],"098fe1ee-e1f4-11f0-8dec-42dde27851f2",["D",1766709869715],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","imageUrl","userId","019aeb5a-8601-72dc-933c-5116c9b9ad52","pinnedFileId","pinPreview","pinMigrated","screenshotVersion","likeCount",0,"orgId","019aeb5a-8601-72dc-933c-5699b88cb012","tursoName","tursoHostname","tursoMeasuredAt","pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_11":12,"_139":135},"org",{"_11":12,"_133":-5,"_134":135,"_95":96,"_136":137,"_138":55},"branch",{"_17":61,"_13":16,"_35":42,"_125":86,"_126":-5,"_127":-5,"_128":-5,"_129":-5,"_130":-5,"_131":14,"_132":12},"isOwned","projectId","forkedBranchId","forkedBranchVersion","forkedProjectPrivacy","forkedProjectId","forkedProjectOrgId","projectName","userHandle","displayName","avatar","https://img.clerk.com/eyJ0eXBlIjoiZGVmYXVsdCIsImlpZCI6Imluc18yTm10cGhQUklwU0tqZHNiOHNnNGd2ZjJURFUiLCJyaWQiOiJ1c2VyXzM2T2tsUkc5c3FHYlNtOG1TamQxZ0pSZzdWWiJ9","tier","free","canRestrictVals","profileImageUrl","adoptOrgId",{"_35":86,"_13":14,"_47":183,"_88":-5,"_89":55,"_90":91,"_92":91,"_93":-5,"_94":-5,"_95":96,"_97":-5,"_98":55,"_99":-5,"_100":-5,"_101":102,"_103":104,"_105":-5,"_106":-5,"_107":-5,"_108":-5,"_109":110,"_111":110,"_112":113,"_114":113,"_115":113,"_116":110,"_117":110,"_118":184,"_120":185,"_122":186,"_124":55},[174,175,176],"files",[174,175,176],"counts",{"_164":102,"_165":166,"_167":102,"_31":168,"_169":170},[158,159,160],"upstreamProject","upstreamBranch","emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},{"_35":66,"_13":67,"_38":68,"_40":-5,"_41":42,"_43":69,"_45":46,"_47":70,"_17":49,"_50":63,"_52":71,"_54":55,"_56":-5,"_18":163},{"_35":58,"_13":59,"_38":60,"_40":-5,"_41":42,"_43":61,"_45":46,"_47":62,"_17":49,"_50":63,"_52":64,"_54":55,"_56":-5,"_18":162},{"_35":36,"_13":37,"_38":39,"_40":-5,"_41":42,"_43":44,"_45":46,"_47":48,"_17":49,"_50":51,"_52":53,"_54":55,"_56":-5,"_18":161},[37],[59],[67],"pulls","branches",1,"forks",{"_171":102,"_173":102},"connections",{"_171":102,"_172":102},"count","unfilledCount","emptyCount",{"_35":66,"_13":67,"_38":68,"_40":-5,"_41":42,"_43":69,"_45":46,"_47":83,"_17":49,"_50":63,"_52":71,"_54":55,"_56":-5,"_77":181,"_18":182},{"_35":58,"_13":59,"_38":60,"_40":-5,"_41":42,"_43":61,"_45":46,"_47":80,"_17":49,"_50":63,"_52":64,"_54":55,"_56":-5,"_77":179,"_18":180},{"_35":36,"_13":37,"_38":39,"_40":-5,"_41":42,"_43":44,"_45":46,"_47":76,"_17":49,"_50":51,"_52":53,"_54":55,"_56":-5,"_77":177,"_18":178},[],[37],[],[59],[],[67],["D",1766709869715],{"_11":12,"_139":135},{"_11":12,"_133":-5,"_134":135,"_95":96,"_136":137,"_138":55},{"_17":61,"_13":16,"_35":42,"_125":86,"_126":-5,"_127":-5,"_128":-5,"_129":-5,"_130":-5,"_131":14,"_132":12},"ctx",{"_118":-5,"_155":220,"_221":222,"_223":224,"_225":55,"_226":55,"_227":55,"_228":46,"_229":-5,"_230":-7,"_231":-7,"_232":102,"_233":55,"_234":55,"_235":-5,"_236":55,"_237":55,"_238":55,"_239":-5},"ENV",{"_191":192,"_193":194,"_195":16,"_196":197,"_198":199,"_200":201,"_202":203,"_204":205,"_206":207,"_208":209,"_210":211,"_212":213,"_214":215,"_216":217,"_218":219},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","e87eb04e707cf64e7e532c8e8ec395f209e17f15","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]

Code