MyTime: Initialstand (scripts/scrape-bridge-days.mjs)

scripts/scrape-bridge-days.mjs

@@ -0,0 +1,139 @@
+#!/usr/bin/env node
+/**
+ * Yearly bridge-days refresh from ferien-feiertage.de.
+ *
+ * Fetches `/bundesland/<slug>/<year>/brueckentage/` for every German state
+ * and writes the result to `src/data/bridge-days.generated.ts`, which the
+ * Brückentage tab reads (offline-friendly, bundled at build time).
+ *
+ * Run once a year, e.g.:
+ *   npm run scrape:bridge
+ *   npm run scrape:bridge -- --years=2027 --states=HH,BY
+ *
+ * Please be polite: the script pauses between requests and sends an
+ * identifying User-Agent. robots.txt currently allows crawling (/).
+ */
+import { readFileSync, writeFileSync } from "node:fs";
+import { parseBridgeDaysTable } from "./parse-bridge-days.mjs";
+
+const BASE = "https://ferien-feiertage.de";
+const USER_AGENT =
+  "MyUrlaubZeit/1.0 (personal holiday planner; yearly refresh; contact via repo owner)";
+const OUT = new URL("../src/data/bridge-days.generated.ts", import.meta.url);
+const STORE = new URL("../src/data/bridge-days.json", import.meta.url);
+
+const STATES = [
+  ["BW", "Baden-Württemberg"],
+  ["BY", "Bayern"],
+  ["BE", "Berlin"],
+  ["BB", "Brandenburg"],
+  ["HB", "Bremen"],
+  ["HH", "Hamburg"],
+  ["HE", "Hessen"],
+  ["MV", "Mecklenburg-Vorpommern"],
+  ["NI", "Niedersachsen"],
+  ["NW", "Nordrhein-Westfalen"],
+  ["RP", "Rheinland-Pfalz"],
+  ["SL", "Saarland"],
+  ["SN", "Sachsen"],
+  ["ST", "Sachsen-Anhalt"],
+  ["SH", "Schleswig-Holstein"],
+  ["TH", "Thüringen"],
+];
+
+export function slugOf(name) {
+  return name
+    .toLowerCase()
+    .replace(/ä/g, "ae")
+    .replace(/ö/g, "oe")
+    .replace(/ü/g, "ue")
+    .replace(/ß/g, "ss")
+    .replace(/\s+/g, "-");
+}
+
+function parseArgs(argv) {
+  const args = { years: null, states: null, delay: 600 };
+  for (const arg of argv) {
+    if (arg.startsWith("--years=")) {
+      args.years = arg.slice("--years=".length).split(",").map(Number);
+    } else if (arg.startsWith("--states=")) {
+      args.states = arg.slice("--states=".length).split(",").map((s) => s.toUpperCase());
+    } else if (arg.startsWith("--delay=")) {
+      args.delay = Number(arg.slice("--delay=".length));
+    }
+  }
+  return args;
+}
+
+const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms));
+
+async function fetchTable(slug, year) {
+  const url = `${BASE}/bundesland/${slug}/${year}/brueckentage/`;
+  const response = await fetch(url, { headers: { "User-Agent": USER_AGENT } });
+  if (!response.ok) {
+    throw new Error(`HTTP ${response.status} for ${url}`);
+  }
+  return parseBridgeDaysTable(await response.text(), year);
+}
+
+async function main() {
+  const args = parseArgs(process.argv.slice(2));
+  const now = new Date().getFullYear();
+  const years = args.years ?? [now, now + 1];
+  const states = args.states
+    ? STATES.filter(([code]) => args.states.includes(code))
+    : STATES;
+  if (states.length === 0) {
+    throw new Error("No matching states");
+  }
+  const data = {};
+  try {
+    Object.assign(data, JSON.parse(readFileSync(STORE, "utf-8")));
+    console.log(`merged ${Object.keys(data).length} existing keys`);
+  } catch {
+    console.log("no existing store, starting fresh");
+  }
+  const missing = [];
+  for (const [code, name] of states) {
+    const slug = slugOf(name);
+    for (const year of years) {
+      const key = `${code}-${year}`;
+      try {
+        data[key] = await fetchTable(slug, year);
+        console.log(`ok   ${key}: ${data[key].length} entries`);
+      } catch (error) {
+        missing.push(`${key} (${error.message})`);
+        console.error(`fail ${key}: ${error.message}`);
+      }
+      await sleep(args.delay);
+    }
+  }
+  const fetchedAt = new Date().toISOString();
+  writeFileSync(STORE, JSON.stringify(data, null, 2) + "\n");
+  const storedYears = [...new Set(Object.keys(data).map((key) => Number(key.slice(-4))))].sort();
+  const storedStates = [...new Set(Object.keys(data).map((key) => key.slice(0, -5)))].sort();
+  const payload =
+    `// AUTO-GENERATED by scripts/scrape-bridge-days.mjs — do not edit by hand.\n` +
+    `// Source: ${BASE} (robots.txt allows crawling). Refresh yearly via \`npm run scrape:bridge\`.\n` +
+    `export interface ScrapedBridgeDay {\n` +
+    `  start: string;\n` +
+    `  end: string;\n` +
+    `  occasion: string;\n` +
+    `  vacationDays: number;\n` +
+    `  freeDays: number;\n` +
+    `  freeStart: string;\n` +
+    `  freeEnd: string;\n` +
+    `}\n\n` +
+    `export const SCRAPED_BRIDGE_META = ${JSON.stringify({ source: BASE, fetchedAt, years: storedYears, states: storedStates }, null, 2)} as const;\n\n` +
+    `export const SCRAPED_BRIDGE_DAYS: Record<string, ScrapedBridgeDay[]> = ${JSON.stringify(data, null, 2)};\n`;
+  writeFileSync(OUT, payload);
+  console.log(`wrote ${OUT.pathname} (${Object.keys(data).length} keys)`);
+  if (missing.length > 0) {
+    console.error(`missing: ${missing.join("; ")}`);
+    process.exitCode = 1;
+  }
+}
+
+if (import.meta.url === `file://${process.argv[1]}`) {
+  await main();
+}