feat: RCDB-backed roller coaster filter with fuzzy name matching

- Add lib/park-meta.ts to manage data/park-meta.json (rcdb_id + coaster lists)
- Add lib/scrapers/rcdb.ts to scrape operating coaster names from RCDB park pages
- discover.ts now seeds park-meta.json with skeleton entries for all parks
- scrape.ts now refreshes RCDB coaster lists (30-day staleness) for parks with rcdb_id set
- fetchLiveRides() accepts a coasterNames Set; isCoaster uses normalize() on both sides
  to handle trademark symbols, 'THE ' prefixes, and punctuation differences between
  Queue-Times and RCDB names — applies correctly to both land rides and top-level rides
- Commit park-meta.json so it ships in the Docker image (fresh volumes get it automatically)
- Update .gitignore / .dockerignore to exclude only *.db files, not all of data/
- Dockerfile copies park-meta.json into image before VOLUME declaration
- README: document coaster filter setup and correct staleness window (72h not 7d)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

.dockerignore

@@ -2,10 +2,9 @@
 .gitea
 .next
 node_modules
-data/
+data/*.db
-*.db
+data/*.db-shm
-*.db-shm
+data/*.db-wal
-*.db-wal
 .env*
 npm-debug.log*
 .DS_Store

.gitignore

@@ -27,7 +27,9 @@ yarn-debug.log*
 yarn-error.log*
 
 # scraped data — local only, not committed
-/data/
+/data/*.db
+/data/*.db-shm
+/data/*.db-wal
 
 # env files
 .env*

Dockerfile

@@ -44,8 +44,13 @@ COPY --from=deps --chown=nextjs:nodejs /app/node_modules ./node_modules
 RUN npx playwright install --with-deps chromium && \
     chown -R nextjs:nodejs /app/.playwright
 
-# SQLite data directory — mount a named volume here for persistence
+# Seed data directory with park-meta.json (RCDB coaster lists + rcdb_id mappings).
+# Must be copied before VOLUME so Docker initialises a fresh named volume with
+# this file already present. Existing volumes retain their own copy.
 RUN mkdir -p /app/data && chown nextjs:nodejs /app/data
+COPY --from=builder --chown=nextjs:nodejs /app/data/park-meta.json ./data/park-meta.json
+
+# SQLite database lives here — mount a named volume for persistence
 VOLUME ["/app/data"]
 
 USER nextjs

README.md

@@ -31,6 +31,13 @@ The park detail page shows ride open/closed status using a two-tier approach:
 
 2. **Schedule fallback (Six Flags API)** — the Six Flags operating-hours API drops the current day from its response once a park opens. When Queue-Times data is unavailable, the app falls back to the nearest upcoming date from the Six Flags schedule API as an approximation.
 
+### Roller Coaster Filter
+
+When live data is shown, a **Coasters only** toggle appears if roller coaster data has been populated for that park. Coaster lists are sourced from [RCDB](https://rcdb.com) and stored in `data/park-meta.json`. To populate them:
+
+1. Open `data/park-meta.json` and set `rcdb_id` for each park to the numeric RCDB park ID (visible in the URL: `https://rcdb.com/4529.htm` → `4529`).
+2. Run `npm run scrape` — coaster lists are fetched from RCDB and stored in the JSON file. They refresh automatically every 30 days on subsequent scrapes.
+
 ---
 
 ## Local Development
@@ -111,4 +118,6 @@ docker run --rm -v sixflagssupercalendar_park_data:/app/data \
 
 ## Data Refresh
 
-The scraper skips any park + month combination scraped within the last 7 days. Run `npm run scrape` on a weekly schedule to keep data current. Parks or months not yet in the database show a `—` placeholder; parks with no open days in the displayed week are hidden from the calendar automatically.
+The scraper skips any park + month already scraped within the last 72 hours. The nightly Docker scraper service handles this automatically. Parks or months not yet in the database show a `—` placeholder; parks with no open days in the displayed week are hidden from the calendar automatically.
+
+Roller coaster lists (from RCDB) are refreshed every 30 days on each `npm run scrape` run, for parks with a configured `rcdb_id`.

app/park/[id]/page.tsx

@@ -5,6 +5,7 @@ import { openDb, getParkMonthData, getApiId } from "@/lib/db";
 import { scrapeRidesForDay } from "@/lib/scrapers/sixflags";
 import { fetchLiveRides } from "@/lib/scrapers/queuetimes";
 import { QUEUE_TIMES_IDS } from "@/lib/queue-times-map";
+import { readParkMeta, getCoasterSet } from "@/lib/park-meta";
 import { ParkMonthCalendar } from "@/components/ParkMonthCalendar";
 import { LiveRidePanel } from "@/components/LiveRidePanel";
 import type { RideStatus, RidesFetchResult } from "@/lib/scrapers/sixflags";
@@ -46,11 +47,14 @@ export default async function ParkPage({ params, searchParams }: PageProps) {
 
   // ── Ride data: try live Queue-Times first, fall back to schedule ──────────
   const queueTimesId = QUEUE_TIMES_IDS[id];
+  const parkMeta = readParkMeta();
+  const coasterSet = getCoasterSet(id, parkMeta);
+
   let liveRides: LiveRidesResult | null = null;
   let ridesResult: RidesFetchResult | null = null;
 
   if (queueTimesId) {
-    liveRides = await fetchLiveRides(queueTimesId);
+    liveRides = await fetchLiveRides(queueTimesId, coasterSet);
   }
 
   // Only hit the schedule API as a fallback when live data is unavailable

data/park-meta.json

@@ -0,0 +1,416 @@
+{
+  "greatadventure": {
+    "rcdb_id": 4534,
+    "coasters": [
+      "Superman - Ultimate Flight",
+      "El Toro",
+      "Dark Knight",
+      "Joker",
+      "Jersey Devil Coaster",
+      "Lil' Devil Coaster",
+      "Flash: Vertical Velocity",
+      "Batman The Ride",
+      "Skull Mountain",
+      "Runaway Mine Train",
+      "Medusa",
+      "Harley Quinn Crazy Train",
+      "Nitro"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:40:09.731Z"
+  },
+  "magicmountain": {
+    "rcdb_id": 4532,
+    "coasters": [
+      "Ninja",
+      "New Revolution",
+      "Batman The Ride",
+      "Viper",
+      "Gold Rusher",
+      "Riddler's Revenge",
+      "Canyon Blaster",
+      "Goliath",
+      "X2",
+      "Scream!",
+      "Tatsu",
+      "Apocalypse the Ride",
+      "Road Runner Express",
+      "Speedy Gonzales Hot Rod Racers",
+      "Full Throttle",
+      "Twisted Colossus",
+      "West Coast Racers",
+      "Wonder Woman Flight of Courage"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:45:43.666Z"
+  },
+  "greatamerica": {
+    "rcdb_id": 4530,
+    "coasters": [
+      "Demon",
+      "Batman The Ride",
+      "American Eagle",
+      "Viper",
+      "Whizzer",
+      "Sprocket Rockets",
+      "Raging Bull",
+      "Flash: Vertical Velocity",
+      "Superman - Ultimate Flight",
+      "Dark Knight",
+      "Little Dipper",
+      "Goliath",
+      "X-Flight",
+      "Joker",
+      "Maxx Force",
+      "Wrath of Rakshasa"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:29:24.092Z"
+  },
+  "overgeorgia": {
+    "rcdb_id": 4535,
+    "coasters": [
+      "Blue Hawk",
+      "Great American Scream Machine",
+      "Dahlonega Mine Train",
+      "Batman The Ride",
+      "Georgia Scorcher",
+      "Superman - Ultimate Flight",
+      "Joker Funhouse Coaster",
+      "Goliath",
+      "Dare Devil Dive",
+      "Twisted Cyclone",
+      "Riddler Mindbender",
+      "Georgia Gold Rusher"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:29:26.121Z"
+  },
+  "overtexas": {
+    "rcdb_id": 4531,
+    "coasters": [
+      "Pandemonium",
+      "New Texas Giant",
+      "Joker",
+      "Aquaman: Power Wave",
+      "Shock Wave",
+      "Judge Roy Scream",
+      "Runaway Mine Train",
+      "Runaway Mountain",
+      "Mini Mine Train",
+      "Mr. Freeze",
+      "Batman The Ride",
+      "Titan",
+      "Wile E. Coyote's Grand Canyon Blaster"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:45:45.715Z"
+  },
+  "stlouis": {
+    "rcdb_id": 4536,
+    "coasters": [
+      "Ninja",
+      "River King Mine Train",
+      "Mr. Freeze Reverse Blast",
+      "Batman The Ride",
+      "Screamin' Eagle",
+      "Boss",
+      "Pandemonium",
+      "American Thunder",
+      "Boomerang",
+      "Rookie Racer"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:45:47.770Z"
+  },
+  "fiestatexas": {
+    "rcdb_id": 4538,
+    "coasters": [
+      "Batgirl Coaster Chase",
+      "Road Runner Express",
+      "Poltergeist",
+      "Boomerang Coast to Coaster",
+      "Superman Krypton Coaster",
+      "Pandemonium",
+      "Chupacabra",
+      "Iron Rattler",
+      "Batman The Ride",
+      "Wonder Woman Golden Lasso Coaster",
+      "Dr. Diabolical's Cliffhanger"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:45:49.819Z"
+  },
+  "newengland": {
+    "rcdb_id": 4565,
+    "coasters": [
+      "Joker",
+      "Thunderbolt",
+      "Great Chase",
+      "Riddler Revenge",
+      "Superman the Ride",
+      "Flashback",
+      "Catwoman's Whip",
+      "Pandemonium",
+      "Batman - The Dark Knight",
+      "Wicked Cyclone",
+      "Gotham City Gauntlet Escape from Arkham Asylum"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:45:51.866Z"
+  },
+  "discoverykingdom": {
+    "rcdb_id": 4711,
+    "coasters": [
+      "Roadrunner Express",
+      "Medusa",
+      "Cobra",
+      "Flash: Vertical Velocity",
+      "Kong",
+      "Boomerang",
+      "Superman Ultimate Flight",
+      "Joker",
+      "Batman The Ride",
+      "Sidewinder Safari"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:45:53.909Z"
+  },
+  "mexico": {
+    "rcdb_id": 4629,
+    "coasters": [
+      "Tsunami",
+      "Superman Krypton Coaster",
+      "Batgirl Batarang",
+      "Batman The Ride",
+      "Superman el Último Escape",
+      "Dark Knight",
+      "Joker",
+      "Medusa Steel Coaster",
+      "Wonder Woman",
+      "Speedway Stunt Coaster"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:45:55.963Z"
+  },
+  "greatescape": {
+    "rcdb_id": 4596,
+    "coasters": [
+      "Comet",
+      "Steamin' Demon",
+      "Flashback",
+      "Canyon Blaster",
+      "Frankie's Mine Train",
+      "Bobcat"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:45:58.013Z"
+  },
+  "darienlake": {
+    "rcdb_id": 4581,
+    "coasters": [
+      "Predator",
+      "Viper",
+      "Mind Eraser",
+      "Boomerang",
+      "Ride of Steel",
+      "Hoot N Holler",
+      "Moto Coaster",
+      "Tantrum"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:00.042Z"
+  },
+  "cedarpoint": {
+    "rcdb_id": 4529,
+    "coasters": [
+      "Raptor",
+      "Rougarou",
+      "Magnum XL-200",
+      "Blue Streak",
+      "Corkscrew",
+      "Gemini",
+      "Wilderness Run",
+      "Woodstock Express",
+      "Millennium Force",
+      "Iron Dragon",
+      "Cedar Creek Mine Ride",
+      "Maverick",
+      "GateKeeper",
+      "Valravn",
+      "Steel Vengeance",
+      "Top Thrill 2",
+      "Wild Mouse",
+      "Siren’s Curse"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:02.082Z"
+  },
+  "knotts": {
+    "rcdb_id": 4546,
+    "coasters": [
+      "Jaguar!",
+      "GhostRider",
+      "Xcelerator",
+      "Silver Bullet",
+      "Sierra Sidewinder",
+      "Pony Express",
+      "Coast Rider",
+      "HangTime",
+      "Snoopy’s Tenderpaw Twister Coaster"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:04.120Z"
+  },
+  "canadaswonderland": {
+    "rcdb_id": 4539,
+    "coasters": [
+      "Flight Deck",
+      "Dragon Fyre",
+      "Mighty Canadian Minebuster",
+      "Wilde Beast",
+      "Ghoster Coaster",
+      "Thunder Run",
+      "Bat",
+      "Vortex",
+      "Taxi Jam",
+      "Fly",
+      "Silver Streak",
+      "Backlot Stunt Coaster",
+      "Behemoth",
+      "Leviathan",
+      "Wonder Mountain's Guardian",
+      "Yukon Striker",
+      "Snoopy's Racing Railway",
+      "AlpenFury"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:06.152Z"
+  },
+  "carowinds": {
+    "rcdb_id": 4542,
+    "coasters": [
+      "Carolina Cyclone",
+      "Woodstock Express",
+      "Carolina Goldrusher",
+      "Hurler",
+      "Vortex",
+      "Wilderness Run",
+      "Afterburn",
+      "Flying Cobras",
+      "Thunder Striker",
+      "Fury 325",
+      "Copperhead Strike",
+      "Snoopy’s Racing Railway",
+      "Ricochet",
+      "Kiddy Hawk"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:08.185Z"
+  },
+  "kingsdominion": {
+    "rcdb_id": 4544,
+    "coasters": [
+      "Racer 75",
+      "Woodstock Express",
+      "Grizzly",
+      "Flight of Fear",
+      "Reptilian",
+      "Great Pumpkin Coaster",
+      "Apple Zapple",
+      "Backlot Stunt Coaster",
+      "Dominator",
+      "Pantherian",
+      "Twisted Timbers",
+      "Tumbili",
+      "Rapterra"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:10.223Z"
+  },
+  "kingsisland": {
+    "rcdb_id": 4540,
+    "coasters": [
+      "Flight of Fear",
+      "Beast",
+      "Racer",
+      "Adventure Express",
+      "Woodstock Express",
+      "Bat",
+      "Great Pumpkin Coaster",
+      "Invertigo",
+      "Diamondback",
+      "Banshee",
+      "Orion",
+      "Mystic Timbers",
+      "Snoopy's Soap Box Racers",
+      "Woodstock’s Air Rail",
+      "Queen City Stunt Coaster"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:12.251Z"
+  },
+  "valleyfair": {
+    "rcdb_id": 4552,
+    "coasters": [
+      "High Roller",
+      "Corkscrew",
+      "Excalibur",
+      "Wild Thing",
+      "Mad Mouse",
+      "Steel Venom",
+      "Renegade",
+      "Cosmic Coaster"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:14.298Z"
+  },
+  "worldsoffun": {
+    "rcdb_id": 4533,
+    "coasters": [
+      "Timber Wolf",
+      "Cosmic Coaster",
+      "Mamba",
+      "Spinning Dragons",
+      "Patriot",
+      "Prowler",
+      "Zambezi Zinger",
+      "Boomerang"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:16.328Z"
+  },
+  "miadventure": {
+    "rcdb_id": 4578,
+    "coasters": [
+      "Corkscrew",
+      "Wolverine Wildcat",
+      "Zach's Zoomer",
+      "Shivering Timbers",
+      "Mad Mouse",
+      "Thunderhawk",
+      "Woodstock Express"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:18.370Z"
+  },
+  "dorneypark": {
+    "rcdb_id": 4588,
+    "coasters": [
+      "Thunderhawk",
+      "Steel Force",
+      "Wild Mouse",
+      "Woodstock Express",
+      "Talon",
+      "Hydra the Revenge",
+      "Possessed",
+      "Iron Menace"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:20.413Z"
+  },
+  "cagreatamerica": {
+    "rcdb_id": 4541,
+    "coasters": [
+      "Demon",
+      "Grizzly",
+      "Woodstock Express",
+      "Patriot",
+      "Flight Deck",
+      "Lucy's Crabbie Cabbies",
+      "Psycho Mouse",
+      "Gold Striker",
+      "RailBlazer"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:22.465Z"
+  },
+  "frontiercity": {
+    "rcdb_id": 4559,
+    "coasters": [
+      "Silver Bullet",
+      "Wildcat",
+      "Diamondback",
+      "Steel Lasso",
+      "Frankie's Mine Train"
+    ],
+    "coasters_scraped_at": "2026-04-04T17:46:24.519Z"
+  }
+}

lib/park-meta.ts

@@ -0,0 +1,86 @@
+/**
+ * park-meta.json — persisted alongside the SQLite DB in data/
+ *
+ * This file stores per-park metadata that doesn't belong in the schedule DB:
+ *   - rcdb_id: user-supplied RCDB park ID (fills into https://rcdb.com/{id}.htm)
+ *   - coasters: list of operating roller coaster names scraped from RCDB
+ *   - coasters_scraped_at: ISO timestamp of last RCDB scrape
+ *
+ * discover.ts: ensures every park has a skeleton entry (rcdb_id null by default)
+ * scrape.ts:   populates coasters[] for parks with a known rcdb_id (30-day staleness)
+ */
+
+import fs from "fs";
+import path from "path";
+
+const META_PATH = path.join(process.cwd(), "data", "park-meta.json");
+
+export interface ParkMeta {
+  /** RCDB park page ID — user fills this in manually after discover creates the skeleton */
+  rcdb_id: number | null;
+  /** Operating roller coaster names scraped from RCDB */
+  coasters: string[];
+  /** ISO timestamp of when coasters was last scraped from RCDB */
+  coasters_scraped_at: string | null;
+}
+
+export type ParkMetaMap = Record<string, ParkMeta>;
+
+export function readParkMeta(): ParkMetaMap {
+  try {
+    return JSON.parse(fs.readFileSync(META_PATH, "utf8")) as ParkMetaMap;
+  } catch {
+    return {};
+  }
+}
+
+export function writeParkMeta(meta: ParkMetaMap): void {
+  fs.mkdirSync(path.dirname(META_PATH), { recursive: true });
+  fs.writeFileSync(META_PATH, JSON.stringify(meta, null, 2) + "\n");
+}
+
+/** Default skeleton entry for a park that has never been configured. */
+export function defaultParkMeta(): ParkMeta {
+  return { rcdb_id: null, coasters: [], coasters_scraped_at: null };
+}
+
+const COASTER_STALE_MS = 30 * 24 * 60 * 60 * 1000; // 30 days
+
+/** Returns true when the coaster list needs to be re-scraped from RCDB. */
+export function areCoastersStale(entry: ParkMeta): boolean {
+  if (!entry.coasters_scraped_at) return true;
+  return Date.now() - new Date(entry.coasters_scraped_at).getTime() > COASTER_STALE_MS;
+}
+
+/**
+ * Normalize a ride name for fuzzy matching between data sources.
+ *
+ * Queue-Times uses branded names (BATMAN™ The Ride, THE JOKER™ Funhouse Coaster)
+ * while RCDB uses clean names (Batman The Ride, Joker Funhouse Coaster).
+ *
+ * Normalization steps:
+ *   1. Strip trademark/copyright symbols (™ ® ©)
+ *   2. Strip leading "THE " / "THE" prefix
+ *   3. Replace punctuation (- : ' ") with spaces
+ *   4. Collapse runs of whitespace
+ *   5. Lowercase and trim
+ */
+export function normalizeRideName(name: string): string {
+  return name
+    .replace(/[™®©]/g, "")
+    .replace(/^the\s+/i, "")
+    .replace(/[-:'".]/g, " ")
+    .replace(/\s+/g, " ")
+    .toLowerCase()
+    .trim();
+}
+
+/**
+ * Returns a Set of normalized coaster names for fast membership checks.
+ * Returns null when no coaster data exists for the park.
+ */
+export function getCoasterSet(parkId: string, meta: ParkMetaMap): Set<string> | null {
+  const entry = meta[parkId];
+  if (!entry || entry.coasters.length === 0) return null;
+  return new Set(entry.coasters.map(normalizeRideName));
+}

lib/scrapers/queuetimes.ts

@@ -9,6 +9,20 @@
 
 const BASE = "https://queue-times.com/parks";
 
+/**
+ * Normalize a ride name for fuzzy matching between Queue-Times and RCDB.
+ * Strips trademark symbols, leading "THE ", and punctuation before comparing.
+ */
+function normalize(name: string): string {
+  return name
+    .replace(/[™®©]/g, "")
+    .replace(/^the\s+/i, "")
+    .replace(/[-:'".]/g, " ")
+    .replace(/\s+/g, " ")
+    .toLowerCase()
+    .trim();
+}
+
 const HEADERS = {
   "User-Agent":
     "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 " +
@@ -21,7 +35,7 @@ export interface LiveRide {
   isOpen: boolean;
   waitMinutes: number;
   lastUpdated: string; // ISO 8601
-  /** True when Queue-Times placed this ride in a "Coasters" land category. */
+  /** True when the ride name appears in the RCDB coaster list for this park. */
   isCoaster: boolean;
 }
 
@@ -58,11 +72,16 @@ interface QTResponse {
  * - The request fails
  * - The response contains no rides
  *
+ * Pass coasterNames (from RCDB static data) to classify rides accurately.
+ * Matching is case-insensitive. When coasterNames is null no ride is
+ * classified as a coaster and the "Coasters only" toggle is hidden.
+ *
  * Pass revalidate (seconds) to control Next.js ISR cache lifetime.
  * Defaults to 300s (5 min) to match Queue-Times update frequency.
  */
 export async function fetchLiveRides(
   queueTimesId: number,
+  coasterNames: Set<string> | null = null,
   revalidate = 300,
 ): Promise<LiveRidesResult | null> {
   const url = `${BASE}/${queueTimesId}/queue_times.json`;
@@ -78,10 +97,7 @@ export async function fetchLiveRides(
 
     const rides: LiveRide[] = [];
 
-    // Rides are nested inside lands. Queue-Times labels coaster sections
-    // with names like "Coasters", "Steel Coasters", "Wooden Coasters", etc.
     for (const land of json.lands ?? []) {
-      const isCoaster = land.name.toLowerCase().includes("coaster");
       for (const r of land.rides ?? []) {
         if (!r.name) continue;
         rides.push({
@@ -89,7 +105,7 @@ export async function fetchLiveRides(
           isOpen: r.is_open,
           waitMinutes: r.wait_time ?? 0,
           lastUpdated: r.last_updated,
-          isCoaster,
+          isCoaster: coasterNames ? coasterNames.has(normalize(r.name)) : false,
         });
       }
     }
@@ -102,7 +118,7 @@ export async function fetchLiveRides(
         isOpen: r.is_open,
         waitMinutes: r.wait_time ?? 0,
         lastUpdated: r.last_updated,
-        isCoaster: false,
+        isCoaster: coasterNames ? coasterNames.has(normalize(r.name)) : false,
       });
     }
 

lib/scrapers/rcdb.ts

@@ -0,0 +1,91 @@
+/**
+ * RCDB (Roller Coaster DataBase) scraper.
+ *
+ * Fetches a park's RCDB page (https://rcdb.com/{id}.htm) and extracts the
+ * names of operating roller coasters from the "Operating Roller Coasters"
+ * section.
+ *
+ * RCDB has no public API. This scraper reads the static HTML page.
+ * Please scrape infrequently (30-day staleness window) to be respectful.
+ */
+
+const BASE = "https://rcdb.com";
+
+const HEADERS = {
+  "User-Agent":
+    "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 " +
+    "(KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36",
+  Accept: "text/html,application/xhtml+xml",
+  "Accept-Language": "en-US,en;q=0.9",
+};
+
+/**
+ * Scrape operating roller coaster names for a park.
+ *
+ * Returns an array of coaster names on success, or null when the page
+ * cannot be fetched or contains no operating coasters.
+ */
+export async function scrapeRcdbCoasters(rcdbId: number): Promise<string[] | null> {
+  const url = `${BASE}/${rcdbId}.htm`;
+  try {
+    const res = await fetch(url, { headers: HEADERS });
+    if (!res.ok) {
+      console.error(`  RCDB ${rcdbId}: HTTP ${res.status}`);
+      return null;
+    }
+    const html = await res.text();
+    return parseOperatingCoasters(html);
+  } catch (err) {
+    console.error(`  RCDB ${rcdbId}: ${err}`);
+    return null;
+  }
+}
+
+/**
+ * Parse operating roller coaster names from RCDB park page HTML.
+ *
+ * RCDB park pages list coasters in sections bounded by <section> tags.
+ * The operating section heading looks like:
+ *   <h4>Operating Roller Coasters: <a href="...">16</a></h4>
+ *
+ * Each coaster is an <a> link to its detail page with an unquoted href:
+ *   <td data-sort="Batman The Ride"><a href=/5.htm>Batman The Ride</a>
+ *
+ * We extract only those links (href=/DIGITS.htm) from within the
+ * operating section, stopping at the next <section> tag.
+ */
+function parseOperatingCoasters(html: string): string[] {
+  // Find the "Operating Roller Coasters" section heading.
+  const opIdx = html.search(/Operating\s+Roller\s+Coasters/i);
+  if (opIdx === -1) return [];
+
+  // The section ends at the next <section> tag (e.g. "Defunct Roller Coasters").
+  const after = html.slice(opIdx);
+  const nextSection = after.search(/<section\b/i);
+  const sectionHtml = nextSection > 0 ? after.slice(0, nextSection) : after;
+
+  // Extract coaster names from links to RCDB detail pages.
+  // RCDB uses unquoted href attributes: href=/1234.htm
+  // General links (/g.htm, /r.htm, /location.htm, etc.) won't match \d+\.htm.
+  const names: string[] = [];
+  const linkPattern = /<a\s[^>]*href=["']?\/(\d+)\.htm["']?[^>]*>([^<]+)<\/a>/gi;
+  let match: RegExpExecArray | null;
+
+  while ((match = linkPattern.exec(sectionHtml)) !== null) {
+    const name = decodeHtmlEntities(match[2].trim());
+    if (name) names.push(name);
+  }
+
+  // Deduplicate while preserving order
+  return [...new Set(names)];
+}
+
+function decodeHtmlEntities(text: string): string {
+  return text
+    .replace(/&amp;/g, "&")
+    .replace(/&lt;/g, "<")
+    .replace(/&gt;/g, ">")
+    .replace(/&quot;/g, '"')
+    .replace(/&#(\d+);/g, (_, code) => String.fromCharCode(parseInt(code, 10)))
+    .replace(/&[a-z]+;/gi, "");
+}

scripts/discover.ts

@@ -17,6 +17,7 @@ import { chromium } from "playwright";
 import { openDb, getApiId, setApiId, type DbInstance } from "../lib/db";
 import { PARKS } from "../lib/parks";
 import { fetchParkInfo, isMainThemePark } from "../lib/scrapers/sixflags";
+import { readParkMeta, writeParkMeta, defaultParkMeta } from "../lib/park-meta";
 
 const CLOUDFRONT_PATTERN = /operating-hours\/park\/(\d+)/;
 
@@ -124,11 +125,39 @@ async function main() {
     await new Promise((r) => setTimeout(r, 2000));
   }
 
+  // ── Ensure park-meta.json has a skeleton entry for every park ────────────
+  // Users fill in rcdb_id manually; scrape.ts populates coasters[] from RCDB.
+  const meta = readParkMeta();
+  let metaChanged = false;
+
+  for (const park of PARKS) {
+    if (!meta[park.id]) {
+      meta[park.id] = defaultParkMeta();
+      metaChanged = true;
+    }
+  }
+  // Remove entries for parks no longer in the registry
+  for (const id of Object.keys(meta)) {
+    if (!PARKS.find((p) => p.id === id)) {
+      delete meta[id];
+      metaChanged = true;
+    }
+  }
+
+  if (metaChanged) {
+    writeParkMeta(meta);
+    console.log("\nUpdated data/park-meta.json");
+    console.log("  → Set rcdb_id for each park to enable the coaster filter.");
+    console.log("    Find a park's RCDB ID from: https://rcdb.com (the number in the URL).");
+  }
+
   // Print summary
   console.log("\n── Discovered IDs ──");
   for (const park of PARKS) {
     const id = getApiId(db, park.id);
-    console.log(`  ${park.id.padEnd(30)} ${id ?? "NOT FOUND"}`);
+    const rcdbId = meta[park.id]?.rcdb_id;
+    const rcdbStr = rcdbId ? `rcdb:${rcdbId}` : "rcdb:?";
+    console.log(`  ${park.id.padEnd(30)} api:${String(id ?? "?").padEnd(8)} ${rcdbStr}`);
   }
 
   db.close();

scripts/scrape.ts

@@ -10,6 +10,8 @@
 import { openDb, upsertDay, getApiId, isMonthScraped } from "../lib/db";
 import { PARKS } from "../lib/parks";
 import { scrapeMonth, RateLimitError } from "../lib/scrapers/sixflags";
+import { readParkMeta, writeParkMeta, areCoastersStale } from "../lib/park-meta";
+import { scrapeRcdbCoasters } from "../lib/scrapers/rcdb";
 
 const YEAR = 2026;
 const MONTHS = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12];
@@ -99,6 +101,42 @@ async function main() {
   if (totalErrors > 0) console.log("  Re-run to retry failed months.");
 
   db.close();
+
+  // ── RCDB coaster scrape (30-day staleness) ────────────────────────────────
+  const meta = readParkMeta();
+  const rcdbParks = PARKS.filter((p) => {
+    const entry = meta[p.id];
+    return entry?.rcdb_id && (FORCE || areCoastersStale(entry));
+  });
+
+  if (rcdbParks.length === 0) {
+    console.log("\nCoaster data up to date.");
+    return;
+  }
+
+  console.log(`\n── RCDB coaster scrape — ${rcdbParks.length} park(s) ──`);
+
+  for (const park of rcdbParks) {
+    const entry = meta[park.id];
+    const rcdbId = entry.rcdb_id!;
+    process.stdout.write(`  ${park.shortName.padEnd(30)} `);
+
+    const coasters = await scrapeRcdbCoasters(rcdbId);
+    if (coasters === null) {
+      console.log("FAILED");
+      continue;
+    }
+
+    entry.coasters = coasters;
+    entry.coasters_scraped_at = new Date().toISOString();
+    console.log(`${coasters.length} coasters`);
+
+    // Polite delay between RCDB requests
+    await new Promise((r) => setTimeout(r, 2000));
+  }
+
+  writeParkMeta(meta);
+  console.log("  Saved to data/park-meta.json");
 }
 
 main().catch((err) => {