Measurement site	Body of water	County	Date	Water temperature [°C]
Starnberg	StarnbergerSee	STA	19.07.2026 19:00	22.9
Ammerseeboje	Ammersee	LL	--	--
BadRow

`;\n\n// Mock date converter: returns the input unchanged so assertions are stable.\nconst mockConvert = (s: string) => s;\n\nconst readings = parseWaterTemperatures(html, 0, mockConvert);\n\nassertEqual(readings.length, 1, \"only the valid row is parsed\");\n\nassertEqual(\n readings[0] satisfies WaterTemperatureReading,\n {\n category_id: 0,\n measurement_site: \"Starnberg\",\n body_of_water: \"StarnbergerSee\",\n measured_at: \"19.07.2026 19:00\",\n water_temperature: 22.9,\n },\n \"valid row has correct fields\",\n);\n\n// Verify categoryId flows through for rivers (category_id=1).\nconst riverReadings = parseWaterTemperatures(html, 1, mockConvert);\nassertEqual(riverReadings[0].category_id, 1, \"rivers use category_id=1\");\n\nconsole.log(\"\\nAll tests passed.\");\n",[],[160,254],"41f2d670-1e36-4dee-b8ee-1f0b02714605","german-date.ts","d3cc2e9c-8396-11f1-b1dd-1607ee4eb77e",["D",1784481924665],"/**\n * Parse a German-format date string (\"24.05.2023 20:00\") as Berlin local time\n * and return the corresponding UTC ISO-8601 timestamp.\n *\n * Berlin observes CET (UTC+1) in winter and CEST (UTC+2) in summer. The offset\n * is resolved for the specific date via Intl timezone formatting, so DST\n * boundaries are handled correctly — no fixed UTC offset assumption.\n *\n * Used to convert gkd.bayern.de measurement timestamps (which are Berlin\n * wall-clock) into the UTC `measured_at` column stored in SQLite.\n */\nexport function germanDateToUtcIso(dateStr: string): string {\n const [day, month, year, hour, minute] = dateStr.split(/[.: ]/).map(Number);\n // Treat the components as if they were UTC to get a reference instant.\n const guess = new Date(Date.UTC(year, month - 1, day, hour, minute));\n // Determine Berlin's UTC offset (minutes) at that instant, then shift back.\n const offsetMin = berlinOffsetMinutes(guess);\n return new Date(guess.getTime() - offsetMin * 60000).toISOString();\n}\n\n/**\n * Return Berlin's UTC offset (in minutes) at the given instant. Positive means\n * Berlin is ahead of UTC (CEST = +120, CET = +60).\n *\n * Uses the same Intl trick as api/pools.ts's berlinDayUtcBounds: format the\n * instant as Berlin wall-clock, parse back as if UTC, and diff.\n */\nfunction berlinOffsetMinutes(at: Date): number {\n const s = at.toLocaleString(\"en-US\", {\n timeZone: \"Europe/Berlin\",\n hour12: false,\n year: \"numeric\",\n month: \"2-digit\",\n day: \"2-digit\",\n hour: \"2-digit\",\n minute: \"2-digit\",\n second: \"2-digit\",\n });\n // s like \"07/17/2026, 20:38:04\"\n const [d, t] = s.split(\", \");\n const [mo, da, yr] = d.split(\"/\").map(Number);\n const [h, mi, se] = t.split(\":\").map(Number);\n const asUtc = Date.UTC(yr, mo - 1, da, h, mi, se);\n return Math.round((asUtc - at.getTime()) / 60000);\n}\n",[],[160,261],"eab32498-06b9-4217-b196-00cee9e5ff83","german-date.test.ts","d3cc25d2-8396-11f1-b1dd-1607ee4eb77e",["D",1784481924665],"import { germanDateToUtcIso } from \"./german-date.ts\";\n\nfunction assertEqual(actual: unknown, expected: unknown, label: string): void {\n const a = JSON.stringify(actual);\n const e = JSON.stringify(expected);\n if (a !== e) {\n throw new Error(`FAIL: ${label}\\n expected: ${e}\\n actual: ${a}`);\n }\n console.log(`PASS: ${label} → ${a}`);\n}\n\n// Summer (CEST, UTC+2): 20:00 Berlin → 18:00 UTC\nassertEqual(\n germanDateToUtcIso(\"24.05.2023 20:00\"),\n \"2023-05-24T18:00:00.000Z\",\n \"summer CEST 20:00 Berlin → 18:00 UTC\",\n);\n\n// Winter (CET, UTC+1): 18:00 Berlin → 17:00 UTC\nassertEqual(\n germanDateToUtcIso(\"15.01.2023 18:00\"),\n \"2023-01-15T17:00:00.000Z\",\n \"winter CET 18:00 Berlin → 17:00 UTC\",\n);\n\n// Midnight Berlin → previous day 22:00 UTC (summer)\nassertEqual(\n germanDateToUtcIso(\"19.07.2026 00:00\"),\n \"2026-07-18T22:00:00.000Z\",\n \"midnight CEST → previous day 22:00 UTC\",\n);\n\n// Noon Berlin → 10:00 UTC (summer)\nassertEqual(\n germanDateToUtcIso(\"19.07.2026 12:00\"),\n \"2026-07-19T10:00:00.000Z\",\n \"noon CEST → 10:00 UTC\",\n);\n\n// DST transition spring: 26.03.2023 03:00 (just after spring-forward)\nassertEqual(\n germanDateToUtcIso(\"26.03.2023 03:00\"),\n \"2023-03-26T01:00:00.000Z\",\n \"after spring-forward 03:00 CEST → 01:00 UTC\",\n);\n\nconsole.log(\"\\nAll tests passed.\");\n",[],[160,268],"63212832-fac2-415b-9ff5-cddef63210d0","business-hours.ts","d556390a-820c-11f1-8a02-1607ee4eb77e",15,["D",1784312705542],"/**\n * Munich / German business-hours guard, ported from v1 cron/german-business-hours.js.\n * Returns true between 07:00 and 23:00 Europe/Berlin time. Used to skip pool\n * occupancy logging overnight when the baths are closed.\n */\nexport function isGermanBusinessHours(now: Date = new Date()): boolean {\n const germanTime = now.toLocaleString(\"en-US\", { timeZone: \"Europe/Berlin\" });\n const hours = new Date(germanTime).getHours();\n return hours >= 7 && hours < 23;\n}\n",[],[160,275],"b6b73857-61aa-45b6-8e94-b812c6db2676",["D",1784312705542],"directory",[182,183,184,185,186,187,188,189,190,191,192],[160],"c00dc9e1-a3d0-4af2-a91c-3413043f2d77","schema.ts","0c3d430c-820b-11f1-b505-1607ee4eb77e","caca4203-217f-44e5-9a2c-341ad96c7be7",["D",1784481924665],"import { sqlite } from \"https://esm.town/v/std/sqlite/main.ts\";\n\n/**\n * Create all tables + indexes. Idempotent: safe to run on every deploy.\n * All timestamps are ISO-8601 UTC strings, e.g. \"2024-06-01T18:00:00Z\".\n */\nexport async function migrate() {\n await sqlite.batch([\n // --- Water temperatures (lakes, rivers, feringasee) ---\n `CREATE TABLE IF NOT EXISTS water_temperatures (\n id INTEGER PRIMARY KEY AUTOINCREMENT,\n category_id INTEGER NOT NULL, -- 0 = lakes, 1 = rivers\n measurement_site TEXT NOT NULL,\n body_of_water TEXT NOT NULL,\n measured_at TEXT NOT NULL, -- ISO-8601 UTC\n water_temperature REAL NOT NULL,\n updated_at TEXT NOT NULL -- ISO-8601 UTC\n )`,\n // Upsert key: one reading per site+body+time.\n `CREATE UNIQUE INDEX IF NOT EXISTS idx_water_temperatures_unique\n ON water_temperatures (category_id, measurement_site, body_of_water, measured_at)`,\n // Read path: latest readings per body of water within a category.\n `CREATE INDEX IF NOT EXISTS idx_water_temperatures_read\n ON water_temperatures (category_id, body_of_water, measured_at)`,\n\n // --- Bodies of water catalog (presentation metadata for the UI) ---\n // Auto-populated by the water-temperature scraper (INSERT OR IGNORE by\n // category_id + name). `geopath` is curated by hand afterwards.\n `CREATE TABLE IF NOT EXISTS bodies_of_water (\n id INTEGER PRIMARY KEY AUTOINCREMENT,\n category_id INTEGER NOT NULL, -- 0 = lakes, 1 = rivers\n name TEXT NOT NULL, -- matches water_temperatures.body_of_water\n slug TEXT NOT NULL, -- generated from name\n geopath TEXT, -- optional frontend asset, e.g. \"isar.svg\"\n created_at TEXT NOT NULL -- ISO-8601 UTC\n )`,\n `CREATE UNIQUE INDEX IF NOT EXISTS idx_bodies_of_water_unique\n ON bodies_of_water (category_id, name)`,\n `CREATE INDEX IF NOT EXISTS idx_bodies_of_water_slug\n ON bodies_of_water (category_id, slug)`,\n\n // --- Bath areas catalog (SWM pool occupancy) ---\n // Roster comes from the SWM visitorCount feed itself (locations -> areas),\n // so this is auto-populated by the pools cron, not hand-seeded. Each area\n // (Hallenbad / Freibad / Sauna / ...) belongs to a location.\n `CREATE TABLE IF NOT EXISTS bath_areas (\n area_id INTEGER PRIMARY KEY, -- SWM area_id\n area_name TEXT NOT NULL,\n location_id INTEGER NOT NULL, -- SWM location_id\n location_name TEXT NOT NULL,\n active INTEGER NOT NULL DEFAULT 1, -- hide areas without deleting history\n updated_at TEXT NOT NULL -- ISO-8601 UTC\n )`,\n `CREATE INDEX IF NOT EXISTS idx_bath_areas_location\n ON bath_areas (location_id)`,\n\n // --- Pool occupancy history (area-level time series) ---\n `CREATE TABLE IF NOT EXISTS occupancy_log (\n id INTEGER PRIMARY KEY AUTOINCREMENT,\n area_id INTEGER NOT NULL, -- -> bath_areas.area_id\n customer_amount INTEGER NOT NULL, -- current occupancy\n customer_amount_max INTEGER NOT NULL, -- capacity\n recorded_at TEXT NOT NULL -- ISO-8601 UTC\n )`,\n `CREATE INDEX IF NOT EXISTS idx_occupancy_log_read\n ON occupancy_log (area_id, recorded_at)`,\n ]);\n}\n\n/**\n * Slugify a body-of-water name. Ported verbatim from v1 src/lib/slugify.js so\n * generated slugs match the frontend's expectations (ä→ae, ö→oe, ü→ue, spaces→-).\n */\nexport function slugify(text: string): string {\n const replacements: Record = { \"ä\": \"ae\", \"ö\": \"oe\", \"ü\": \"ue\" };\n text = text.toLowerCase();\n for (const [umlaut, replacement] of Object.entries(replacements)) {\n text = text.split(umlaut).join(replacement);\n }\n return text.replace(/\\s+/g, \"-\");\n}\n\n/**\n * Register any newly-seen bodies of water in the catalog. Called by the water\n * scraper with the (category_id, name) pairs it just observed. INSERT OR IGNORE\n * keeps existing rows (and their curated `geopath`) untouched.\n */\nexport async function upsertBodiesOfWater(\n bodies: { category_id: number; name: string }[],\n): Promise {\n if (bodies.length === 0) return;\n // De-dupe within this batch so we don't issue redundant inserts.\n const seen = new Set();\n const unique = bodies.filter((b) => {\n const key = `${b.category_id}\\u0000${b.name}`;\n if (seen.has(key)) return false;\n seen.add(key);\n return true;\n });\n const now = new Date().toISOString();\n await sqlite.batch(\n unique.map((b) => ({\n sql: `INSERT OR IGNORE INTO bodies_of_water (category_id, name, slug, created_at)\n VALUES (?, ?, ?, ?)`,\n args: [b.category_id, b.name, slugify(b.name), now],\n })),\n );\n}\n\n/**\n * Upsert water-temperature readings. On conflict (same category+site+body+time),\n * update the temperature and `updated_at` timestamp so re-scraping is idempotent.\n * Mirrors the upsertBathAreas pattern.\n */\nexport async function upsertWaterTemperatures(\n readings: {\n category_id: number;\n measurement_site: string;\n body_of_water: string;\n measured_at: string;\n water_temperature: number;\n }[],\n): Promise {\n if (readings.length === 0) return;\n const now = new Date().toISOString();\n await sqlite.batch(\n readings.map((r) => ({\n sql: `INSERT INTO water_temperatures\n (category_id, measurement_site, body_of_water, measured_at, water_temperature, updated_at)\n VALUES (?, ?, ?, ?, ?, ?)\n ON CONFLICT(category_id, measurement_site, body_of_water, measured_at) DO UPDATE SET\n water_temperature = excluded.water_temperature,\n updated_at = excluded.updated_at`,\n args: [r.category_id, r.measurement_site, r.body_of_water, r.measured_at, r.water_temperature, now],\n })),\n );\n}\n\n/**\n * Keep the bath_areas roster in sync with the SWM feed. Upserts area/location\n * names by area_id (so renames propagate) without touching the `active` flag\n * once a row exists — you can deactivate areas by hand and it will stick.\n */\nexport async function upsertBathAreas(\n areas: { area_id: number; area_name: string; location_id: number; location_name: string }[],\n): Promise {\n if (areas.length === 0) return;\n const now = new Date().toISOString();\n await sqlite.batch(\n areas.map((a) => ({\n sql: `INSERT INTO bath_areas (area_id, area_name, location_id, location_name, active, updated_at)\n VALUES (?, ?, ?, ?, 1, ?)\n ON CONFLICT(area_id) DO UPDATE SET\n area_name = excluded.area_name,\n location_id = excluded.location_id,\n location_name = excluded.location_name,\n updated_at = excluded.updated_at`,\n args: [a.area_id, a.area_name, a.location_id, a.location_name, now],\n })),\n );\n}\n",[],[295,288],"db","976a6e28-b4f0-46ab-825c-6a3aaea2870a",["D",1784311938655],[180],[295],"755c2c95-93fd-423d-ada2-9d7f8d33e54b","water-temperatures.ts","d3cc1614-8396-11f1-b1dd-1607ee4eb77e","d0f4549d-bbf6-4e80-b297-f7fc8a3e7f9c",["D",1784481924665],"interval","import { fetch as stdFetch } from \"https://esm.town/v/std/fetch\";\nimport { migrate, upsertBodiesOfWater, upsertWaterTemperatures } from \"../db/schema.ts\";\nimport { parseWaterTemperatures } from \"../lib/parse-water-temperatures.ts\";\nimport { germanDateToUtcIso } from \"../lib/german-date.ts\";\n\nconst SOURCES = [\n { category_id: 0, url: \"https://www.gkd.bayern.de/en/lakes/watertemperature/tables\" },\n { category_id: 1, url: \"https://www.gkd.bayern.de/en/rivers/watertemperature/tables\" },\n] as const;\n\nasync function fetchHtml(url: string): Promise {\n const res = await stdFetch(url, {\n headers: { accept: \"text/html\", \"user-agent\": \"Mozilla/5.0\" },\n });\n if (!res.ok) throw new Error(`HTTP ${res.status} for ${url}`);\n return await res.text();\n}\n\n/**\n * One cron pass: fetch both lakes and rivers tables from gkd.bayern.de, parse\n * them into readings, register any newly-seen bodies of water in the catalog\n * (slug auto-generated), and upsert the readings. Idempotent — re-scraping the\n * same measurement just refreshes `updated_at`.\n */\nexport async function recordWaterTemperatures(): Promise<{\n lakes: number;\n rivers: number;\n}> {\n await migrate();\n const counts: Record = {};\n for (const source of SOURCES) {\n const html = await fetchHtml(source.url);\n const readings = parseWaterTemperatures(html, source.category_id, germanDateToUtcIso);\n const bodies = readings.map((r) => ({ category_id: r.category_id, name: r.body_of_water }));\n await upsertBodiesOfWater(bodies);\n await upsertWaterTemperatures(readings);\n counts[source.category_id] = readings.length;\n }\n return { lakes: counts[0] ?? 0, rivers: counts[1] ?? 0 };\n}\n\nexport default async function (interval: Interval) {\n const result = await recordWaterTemperatures();\n console.log(\n `Scraped ${result.lakes} lake readings and ${result.rivers} river readings.`,\n );\n}\n",[],[309,301],"cron","6d8aef11-7bc2-48fe-bff3-d7c9f3c1dfa3","pools-capacity.ts","de2a3e3c-820c-11f1-aaf3-1607ee4eb77e",16,["D",1784312720344],"import { fetch as stdFetch } from \"https://esm.town/v/std/fetch\";\nimport { sqlite } from \"https://esm.town/v/std/sqlite/main.ts\";\nimport { migrate, upsertBathAreas } from \"../db/schema.ts\";\nimport { isGermanBusinessHours } from \"../lib/business-hours.ts\";\n\nconst VISITOR_COUNT_URL = \"https://www.swm.de/.rest/bath/visitorCount\";\n\ninterface SwmArea {\n area_id: number;\n area_name: string;\n customer_amount: number;\n customer_amount_max: number;\n}\ninterface SwmLocation {\n location_id: number;\n location_name: string;\n area_list: SwmArea[];\n}\ninterface SwmResponse {\n return_code: string;\n data: SwmLocation[];\n}\n\n/**\n * Fetch the full SWM bath occupancy feed. A single request (no query params)\n * returns every location with its nested areas and live counts.\n */\nasync function fetchVisitorCounts(): Promise {\n const res = await stdFetch(VISITOR_COUNT_URL, {\n headers: { accept: \"*/*\", \"user-agent\": \"Mozilla/5.0\" },\n });\n if (!res.ok) throw new Error(`SWM visitorCount HTTP ${res.status}`);\n const json = (await res.json()) as SwmResponse;\n if (json.return_code !== \"success\" || !Array.isArray(json.data)) {\n throw new Error(`SWM visitorCount unexpected response: ${JSON.stringify(json).slice(0, 200)}`);\n }\n return json.data;\n}\n\n/**\n * One cron pass: fetch the feed, sync the area roster, and log area-level\n * occupancy. Skips logging outside Munich business hours (baths closed), but\n * still refreshes the roster so new areas are discovered even overnight.\n */\nexport async function recordOccupancy(now: Date = new Date()): Promise<{\n skipped: boolean;\n areas: number;\n rows: number;\n}> {\n await migrate();\n\n const locations = await fetchVisitorCounts();\n\n // Flatten locations -> areas, carrying the parent location metadata.\n const flat = locations.flatMap((loc) =>\n loc.area_list.map((a) => ({\n area_id: a.area_id,\n area_name: a.area_name,\n location_id: loc.location_id,\n location_name: loc.location_name,\n customer_amount: a.customer_amount,\n customer_amount_max: a.customer_amount_max,\n }))\n );\n\n // Keep the roster fresh regardless of the hour.\n await upsertBathAreas(flat);\n\n if (!isGermanBusinessHours(now)) {\n return { skipped: true, areas: flat.length, rows: 0 };\n }\n\n const recorded_at = now.toISOString();\n await sqlite.batch(\n flat.map((a) => ({\n sql: `INSERT INTO occupancy_log (area_id, customer_amount, customer_amount_max, recorded_at)\n VALUES (?, ?, ?, ?)`,\n args: [a.area_id, a.customer_amount, a.customer_amount_max, recorded_at],\n })),\n );\n\n return { skipped: false, areas: flat.length, rows: flat.length };\n}\n\nexport default async function (interval: Interval) {\n const result = await recordOccupancy();\n console.log(\n result.skipped\n ? `Outside business hours — refreshed ${result.areas} areas, logged 0 rows.`\n : `Logged ${result.rows} occupancy rows across ${result.areas} areas.`,\n );\n}\n",[],[309,311],"1c4f882b-7bd1-41a6-a7b8-c77844d91d66","feringasee.ts","bc7812fe-83a2-11f1-a9b7-1607ee4eb77e",32,["D",1784487378812],"import { fetch as stdFetch } from \"https://esm.town/v/std/fetch\";\nimport {\n migrate,\n upsertBodiesOfWater,\n upsertWaterTemperatures,\n} from \"../db/schema.ts\";\nimport { germanDateToUtcIso } from \"../lib/german-date.ts\";\n\nconst URL = \"https://xn--wasserwacht-unterfhring-plc.de/aktuelles-wetter/\";\n\n// \"vom 19.07.2026 20:10: 25.80 Grad Celius im Wasser und 24.40 Grad Celsius an der Luft.\"\n// \"Celius\" (sic) is a typo in the source's water-temperature wording — preserved verbatim.\nconst READING_REGEX =\n /vom (\\d{2}\\.\\d{2}\\.\\d{4}\\s\\d{2}:\\d{2}):\\s([\\d.]+) Grad Celius im Wasser und ([\\d.]+) Grad Celsius an der Luft\\./;\n\nasync function fetchHtml(url: string): Promise {\n const res = await stdFetch(url, {\n headers: { accept: \"text/html\", \"user-agent\": \"Mozilla/5.0\" },\n });\n if (!res.ok) throw new Error(`HTTP ${res.status} for ${url}`);\n return await res.text();\n}\n\n/**\n * One cron pass: fetch the Wasserwacht Unterföhring Feringasee weather page,\n * extract the latest water+air temperature reading, register Feringasee in the\n * bodies-of-water catalog if new, and upsert the reading. Idempotent —\n * re-scraping the same measurement just refreshes `updated_at`.\n *\n * Returns null when no reading is found on the page (sensors down, page\n * restructured, etc.), otherwise the parsed reading.\n */\nexport async function recordFeringasee(): Promise<{\n water_temperature: number;\n air_temperature: number;\n measured_at: string;\n} | null> {\n await migrate();\n const html = await fetchHtml(URL);\n // The page wraps the temperature values in tags (e.g. \"25.80 Grad\n // Celius\"). Strip all tags so the regex sees plain text, matching the\n // original v1 scraper's use of node-html-parser's .textContent.\n const text = html.replace(/<[^>]+>/g, \"\");\n const match = text.match(READING_REGEX);\n if (!match) return null;\n\n const measured_at = germanDateToUtcIso(match[1]);\n const water_temperature = parseFloat(match[2]);\n const air_temperature = parseFloat(match[3]);\n\n await upsertBodiesOfWater([{ category_id: 0, name: \"Feringasee\" }]);\n await upsertWaterTemperatures([\n {\n category_id: 0,\n measurement_site: \"Feringasee\",\n body_of_water: \"Feringasee\",\n measured_at,\n water_temperature,\n },\n ]);\n\n return { water_temperature, air_temperature, measured_at };\n}\n\nexport default async function (interval: Interval) {\n const result = await recordFeringasee();\n if (!result) {\n console.log(\"No Feringasee reading found on the page.\");\n return;\n }\n console.log(\n `Feringasee: ${result.water_temperature}°C water, ${result.air_temperature}°C air at ${result.measured_at}.`,\n );\n}\n",[],[309,319],"7eb4bc66-14c6-4aa3-a9b9-5469b6fab177",["D",1784312720344],[176,177,178],[309],"c44215fc-edd8-4463-badf-bede58c65dcf","d3cbe34c-8396-11f1-b1dd-1607ee4eb77e","5dee9817-c87b-4937-82db-c075261bbdf5",["D",1785062610268],14,"import { sqlite } from \"https://esm.town/v/std/sqlite/main.ts\";\nimport { line, curveNatural } from \"https://esm.sh/d3-shape@3\";\nimport { weekTimeValue } from \"../lib/week-time-value.ts\";\nimport { sparklineAltText } from \"../lib/sparkline-alt-text.ts\";\nimport { slugify } from \"../db/schema.ts\";\n\n/**\n * GET /?category=lakes|rivers\n *\n * Returns the latest water-temperature reading per body of water / measurement\n * site pair within the requested category, plus a 7-day time series and a\n * precomputed d3 SVG path (y = temperature mapped to 0–35 °C range). `category` defaults to \"lakes\".\n *\n * The SVG viewBox is 0 0 200 110 (matching AreaCard.svelte). X-axis: 7-day\n * window → [0, 200]. Y-axis: 0 °C → y=100, 35 °C → y=0.\n *\n * Public + CORS-friendly. Cached for 10 minutes.\n */\n\ninterface Row {\n body_of_water: string;\n slug: string;\n measurement_site: string;\n water_temperature: number;\n measured_at: string;\n}\n\ninterface WaterDataPoint {\n measured_at: string;\n water_temperature: number;\n}\n\ninterface WaterBody {\n body_of_water: string;\n slug: string;\n measurement_site: string;\n water_temperature: number;\n measured_at: string;\n path: string;\n trend_text: string;\n}\n\nconst CATEGORIES: Record = { lakes: 0, rivers: 1 };\n\nconst MAX_TEMP = 35;\nconst WEEK_MS = 7 * 24 * 60 * 60 * 1000;\n\nfunction tempToY(temp: number): number {\n return 100 - (temp / MAX_TEMP) * 100;\n}\n\nexport default async function (req: Request): Promise {\n const url = new URL(req.url);\n const categoryParam = url.searchParams.get(\"category\") ?? \"lakes\";\n const categoryId = CATEGORIES[categoryParam];\n if (categoryId === undefined) {\n return Response.json(\n { error: `Invalid category '${categoryParam}'. Use 'lakes' or 'rivers'.` },\n { status: 400 },\n );\n }\n\n const now = new Date();\n const weekAgo = new Date(now.getTime() - WEEK_MS).toISOString();\n\n const { rows } = (await sqlite.execute({\n sql: `\n SELECT w.body_of_water, b.slug, w.measurement_site, w.water_temperature, w.measured_at\n FROM water_temperatures w\n JOIN bodies_of_water b\n ON b.category_id = w.category_id AND b.name = w.body_of_water\n WHERE w.category_id = ?\n AND w.measured_at >= ?\n ORDER BY w.body_of_water ASC, w.measurement_site ASC, w.measured_at ASC\n `,\n args: [categoryId, weekAgo],\n })) as unknown as { rows: Row[] };\n\n // Group rows by body_of_water + measurement_site, preserving ASC time order for path generation.\n const bodyMap = new Map();\n for (const r of rows) {\n const key = `${r.body_of_water}\\u0000${r.measurement_site}`;\n let body = bodyMap.get(key);\n if (!body) {\n body = { row: r, data: [] };\n bodyMap.set(key, body);\n }\n body.data.push({\n measured_at: r.measured_at,\n water_temperature: r.water_temperature,\n });\n // Keep the latest reading as the \"current\" reading.\n body.row = r;\n }\n\n const pathGen = line<[number, number]>()\n .x((d) => d[0])\n .y((d) => d[1])\n .curve(curveNatural);\n\n const bodies: WaterBody[] = [...bodyMap.values()].map(({ row, data }) => {\n const points: [number, number][] = data.map((d) => [\n weekTimeValue(d.measured_at, now),\n tempToY(d.water_temperature),\n ]);\n return {\n body_of_water: row.body_of_water,\n slug: `${row.slug}-${slugify(row.measurement_site)}`,\n measurement_site: row.measurement_site,\n water_temperature: row.water_temperature,\n measured_at: row.measured_at,\n path: pathGen(points) ?? \"\",\n trend_text: sparklineAltText(\n data.map((d) => ({ timestamp: d.measured_at, value: d.water_temperature })),\n row.water_temperature,\n \"°\",\n ),\n };\n });\n\n return Response.json(\n { category: categoryParam, bodies },\n { headers: { \"Cache-Control\": \"public, max-age=600\" } },\n );\n}\n","muenchen-plantscht-water",[],[339,301],"api","72f25944-6d05-4c97-b13a-e3d32598547b","pools.ts","b4a36988-820e-11f1-b7fc-1607ee4eb77e",44,["D",1785058040952],6,"import { sqlite } from \"https://esm.town/v/std/sqlite/main.ts\";\nimport { line, curveNatural } from \"https://esm.sh/d3-shape@3\";\nimport { timeValue } from \"../lib/time-value.ts\";\nimport { sparklineAltText } from \"../lib/sparkline-alt-text.ts\";\n\n/**\n * GET /?date=YYYY-MM-DD (date optional, defaults to today in Europe/Berlin)\n *\n * Returns pool occupancy for a single day, grouped location -> areas. Each area\n * carries its intraday series, a `latest` snapshot, and a precomputed d3 SVG\n * path (y = % free) so a dumb client/widget can render it directly.\n *\n * Public + CORS-friendly (Val Town adds Access-Control-Allow-Origin: * by\n * default as long as we don't set CORS headers ourselves).\n */\n\ninterface Row {\n area_id: number;\n area_name: string;\n location_id: number;\n location_name: string;\n customer_amount: number;\n customer_amount_max: number;\n recorded_at: string;\n}\n\n/** % free, v1's metric: 100 - occupancy%. Clamped to [0,100]. */\nfunction capacityFreePct(count: number, max: number): number {\n if (!max || max <= 0) return 0;\n const pct = 100 - Math.round((count / max) * 100);\n return Math.max(0, Math.min(100, pct));\n}\n\n/** Today's date (YYYY-MM-DD) in Europe/Berlin. */\nfunction todayBerlin(): string {\n // en-CA gives YYYY-MM-DD\n return new Date().toLocaleDateString(\"en-CA\", { timeZone: \"Europe/Berlin\" });\n}\n\nfunction isValidDate(s: string): boolean {\n return /^\\d{4}-\\d{2}-\\d{2}$/.test(s);\n}\n\n/**\n * Given a Berlin calendar day (YYYY-MM-DD), return the UTC instant bounds\n * [startUtc, endUtc) that cover it. DST-correct: derives Berlin's offset for\n * that specific date instead of assuming a fixed +1/+2.\n */\nfunction berlinDayUtcBounds(date: string): { start: string; end: string } {\n // Offset (minutes) that Berlin is ahead of UTC at a given instant.\n const berlinOffsetMinutes = (at: Date): number => {\n // Format the same instant as Berlin wall-clock, parse back as if UTC,\n // and diff — yields the zone offset for that instant (handles DST).\n const s = at.toLocaleString(\"en-US\", {\n timeZone: \"Europe/Berlin\",\n hour12: false,\n year: \"numeric\",\n month: \"2-digit\",\n day: \"2-digit\",\n hour: \"2-digit\",\n minute: \"2-digit\",\n second: \"2-digit\",\n });\n // s like \"07/17/2026, 20:38:04\"\n const [d, t] = s.split(\", \");\n const [mo, da, yr] = d.split(\"/\").map(Number);\n const [h, mi, se] = t.split(\":\").map(Number);\n const asUtc = Date.UTC(yr, mo - 1, da, h, mi, se);\n return Math.round((asUtc - at.getTime()) / 60000);\n };\n\n // Berlin local midnight of `date`, expressed as a first-guess UTC instant,\n // then corrected by that instant's actual offset.\n const [y, m, d] = date.split(\"-\").map(Number);\n const guess = Date.UTC(y, m - 1, d, 0, 0, 0);\n const offset = berlinOffsetMinutes(new Date(guess));\n const startUtc = guess - offset * 60000;\n const endUtc = startUtc + 24 * 60 * 60 * 1000;\n return { start: new Date(startUtc).toISOString(), end: new Date(endUtc).toISOString() };\n}\n\nexport default async function (req: Request): Promise {\n const url = new URL(req.url);\n const dateParam = url.searchParams.get(\"date\");\n const date = dateParam && isValidDate(dateParam) ? dateParam : todayBerlin();\n\n // Select the day's rows for active areas. `recorded_at` is stored UTC; the\n // \"day\" is a Berlin calendar day, so filter by that day's UTC bounds\n // (DST-correct, no fixed offset).\n const { start, end } = berlinDayUtcBounds(date);\n const { rows } = (await sqlite.execute({\n sql: `\n SELECT o.area_id, b.area_name, b.location_id, b.location_name,\n o.customer_amount, o.customer_amount_max, o.recorded_at\n FROM occupancy_log o\n JOIN bath_areas b ON b.area_id = o.area_id\n WHERE b.active = 1\n AND o.recorded_at >= ?\n AND o.recorded_at < ?\n ORDER BY b.location_name, b.area_name, o.recorded_at ASC\n `,\n args: [start, end],\n })) as unknown as { rows: Row[] };\n\n // Group location -> area -> series.\n const locMap = new Map;\n }>();\n\n for (const r of rows) {\n let loc = locMap.get(r.location_id);\n if (!loc) {\n loc = { location_id: r.location_id, location_name: r.location_name, areas: new Map() };\n locMap.set(r.location_id, loc);\n }\n let area = loc.areas.get(r.area_id);\n if (!area) {\n area = { area_id: r.area_id, area_name: r.area_name, data: [] };\n loc.areas.set(r.area_id, area);\n }\n area.data.push({\n recorded_at: r.recorded_at,\n customer_amount: r.customer_amount,\n customer_amount_max: r.customer_amount_max,\n capacity_free_pct: capacityFreePct(r.customer_amount, r.customer_amount_max),\n });\n }\n\n const pathGen = line<[number, number]>()\n .x((d) => d[0])\n .y((d) => d[1])\n .curve(curveNatural);\n\n const locations = [...locMap.values()]\n .map((loc) => ({\n location_id: loc.location_id,\n location_name: loc.location_name,\n areas: [...loc.areas.values()].map((a) => {\n const points: [number, number][] = a.data.map((d) => [\n timeValue(d.recorded_at),\n d.capacity_free_pct,\n ]);\n const last = a.data[a.data.length - 1];\n return {\n area_id: a.area_id,\n area_name: a.area_name,\n latest: last\n ? {\n recorded_at: last.recorded_at,\n customer_amount: last.customer_amount,\n customer_amount_max: last.customer_amount_max,\n capacity_free_pct: last.capacity_free_pct,\n }\n : null,\n trend_text: sparklineAltText(\n a.data.map((d) => ({ timestamp: d.recorded_at, value: d.capacity_free_pct })),\n last?.capacity_free_pct ?? null,\n \"% free\",\n ),\n path: pathGen(points) ?? \"\",\n };\n }),\n }))\n .sort((a, b) => a.location_name.localeCompare(b.location_name, \"de\"));\n\n return Response.json(\n { date, locations },\n { headers: { \"Cache-Control\": \"public, max-age=600\" } },\n );\n}\n","muenchen-plantscht-pools",[],[339,341],"5ed2bc11-5bb2-4e1a-9f83-852a4c175573",20,["D",1784313509669],[173,174],[339],["D",1784311508215],["D",1789559472742],{"_11":12,"_129":76},{"_11":12,"_125":-5,"_75":76,"_30":74,"_126":127,"_128":24},{"_38":116,"_13":16,"_32":35,"_117":78,"_118":-5,"_119":-5,"_120":-5,"_121":-5,"_122":-5,"_123":14,"_124":12},"ctx",{"_52":-5,"_148":393,"_394":395,"_396":397,"_398":24,"_399":24,"_400":24,"_401":90,"_402":-5,"_403":-7,"_404":-7,"_405":26,"_406":24,"_407":24,"_408":-5,"_409":24,"_410":24,"_411":24,"_412":-5},"ENV",{"_364":365,"_366":367,"_368":16,"_369":370,"_371":372,"_373":374,"_375":376,"_377":378,"_379":380,"_381":382,"_383":384,"_385":386,"_387":388,"_389":390,"_391":392},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","fdb67791413f4386b4dbf18f354029a617f6160f","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"]

History · week-time-value.ts