ThaiData UK

ThaiData UK — Restaurant Data & Claim System

https://thaidata.uk — Business intelligence and interactive tools for UK Thai restaurant operators. Free, no signup.

What's Here

• 1,594 Restaurant Pages — Auto-generated from FSA data, one per UK Thai restaurant
• 253 City Pages — Market data per UK city with FSA hygiene ratings
• 28 Chain Profiles — Detailed intelligence on Thai restaurant chains
• 8 Interactive Tools: Wage calculator, delivery profit calculator, hygiene lookup, staffing calculator, market reports, supplier price tracker, restaurant valuation, competitor radar
• 8 Content Guides: Opening a restaurant, chef visas, marketing, revenue benchmarks, menu pricing, opening timeline, UK overview, city breakdown
• 5 City Guides: Enriched landing pages for Manchester, Leeds, Birmingham, Edinburgh, Bristol
• 35-Term Glossary: Industry terms with cross-references
• Free API v1: Cities.json endpoint — no auth, no rate limit
• MCP Server: AI-callable tools at /mcp (JSON-RPC 2.0)
• Claim System: Restaurant owners can claim and enhance their listings

Quick Start

# Clone
git clone https://github.com/swifttiger888-commits/thaidata.git
cd thaidata

# Python deps (stdlib only currently)
pip install -r requirements.txt  # or skip — no external deps needed

# Copy env template
cp .env.example .env
# Edit .env with your values

# Copy wrangler config
cp wrangler.toml.example wrangler.toml
# Edit wrangler.toml with your D1 database ID

Architecture

                    ┌─────────────────────┐
                    │   FSA API (Gov UK)  │
                    └────────┬────────────┘
                             │ weekly cron (Mon 9am)
                             ▼
                    ┌─────────────────────┐
                    │ fetch_all_thai_data │──→ data/restaurants.json
                    │     .py (pipeline)  │──→ api/v1/cities.json
                    └────────┬────────────┘──→ api/v1/data.json
                             │
              ┌──────────────┼──────────────┐
              ▼              ▼              ▼
   generate_    generate_     generate_
   restaurant_   menu_price_  city_links_
   pages.py     page.py       .py
       │             │             │
       ▼             ▼             ▼
   r/[city]/    thai-menu-    cities/[city]
   [slug]/      price-index   .html
   index.html   .html
                              
   ┌──────────────────────────────────────────┐
   │     Cloudflare Pages (static deploy)     │
   │  git push → auto-build → thaidata.uk     │
   │                                          │
   │   functions/     ← API + MCP endpoints  │
   │     api/                                 │
   │       claim.js     POST /api/claim       │
   │       admin.js     POST/PUT /api/admin   │
   │       listing.js   GET/PUT /api/listing  │
   │       search.js    GET /api/search       │
   │       upload.js    POST /api/upload      │
   │       claimed.js   GET /api/claimed      │
   │       diag.js      GET /api/diag         │
   │     mcp.js         POST /mcp            │
   │     _middleware.js CORS                  │
   │                                          │
   │   D1 Database (SQLite)                   │
   │     claims, listings, auth_tokens        │
   └──────────────────────────────────────────┘

Project Structure

thaidata/
├── scripts/
│   ├── fetch_all_thai_data.py     # Main data pipeline (FSA API → JSON)
│   ├── generate_restaurant_pages.py  # 1,594 individual restaurant pages
│   ├── generate_menu_price_page.py   # Menu price index page
│   ├── collect_menu_prices.py        # Manual price survey tool
│   ├── add_city_restaurant_links.py  # Link city pages → restaurant pages
│   └── websub-ping.py                # WebSub/IndexNow notification
├── functions/
│   ├── _middleware.js             # CORS handler
│   ├── mcp.js                     # MCP server (AI tools)
│   └── api/
│       ├── admin.js               # Admin panel API
│       ├── claim.js               # Claim submission
│       ├── claimed.js             # Claimed listing queries
│       ├── diag.js                # Diagnostics
│       ├── listing.js             # Listing management
│       ├── search.js              # FSA proxy search
│       └── upload.js              # Photo upload
├── data/
│   ├── restaurants.json           # 1,594 restaurants (SINGLE SOURCE OF TRUTH)
│   └── postcode_geocodes.json     # Geocoded postcodes
├── api/v1/
│   ├── cities.json                # 253 city statistics
│   └── data.json                  # Overview data
├── schema.sql                     # D1 database schema
├── wrangler.toml                  # Cloudflare Pages config (LIVE)
├── wrangler.toml.example          # Template for new devs
├── .env.example                   # Environment variable template
├── .gitignore
└── requirements.txt               # Python dependencies

Data Pipeline

The weekly data refresh runs every Monday at 9:00 AM via cron job (Hermes Agent).

Pipeline steps:

1. scripts/fetch_all_thai_data.py — Fetches FSA FHRS + FHIS APIs
   • Searches by "Thai" keyword across all business types
   • Searches 11 named chains (Giggling Squid, Busaba, etc.) for non-"Thai" named restaurants
   • Deduplicates by FHRSID (first match wins)
   • Maps Local Authorities → cities via 85-entry lookup table
   • Computes city statistics (population density, avg hygiene rating)
   • Saves: data/restaurants.json, api/v1/cities.json, api/v1/data.json
2. scripts/generate_restaurant_pages.py — Generates individual /r/ pages
3. scripts/add_city_restaurant_links.py — Adds restaurant links to city pages
4. Push to git → Cloudflare Pages auto-deploys

Run manually:

python3 scripts/fetch_all_thai_data.py
python3 scripts/generate_restaurant_pages.py
python3 scripts/add_city_restaurant_links.py
git add -A && git commit -m "weekly data refresh" && git push origin main

Note: The pipeline script writes to /tmp/thaidata-repo/ by default. Edit REPO_DIR at the top of scripts/fetch_all_thai_data.py if your working directory is different.

Environment Variables

Set these in Cloudflare Dashboard (Workers & Pages → thaidata-uk → Settings → Variables):

Variable	Required	Description
ADMIN_PASSWORD	Yes	Password for /admin panel
MAGIC_LINK_SECRET	Yes	64+ char random key for claim token signing
RESEND_API_KEY	Yes	Resend.com API key (starts with re_) for magic link emails
SITE_URL	Yes	https://thaidata.uk

D1 Database Setup

# Create database
wrangler d1 create thaidata-db

# Update wrangler.toml with the returned database ID

# Initialize schema
wrangler d1 execute thaidata-db --file=schema.sql

# Bind in Cloudflare Dashboard:
# Workers & Pages → thaidata-uk → Settings → Functions → D1 database bindings
# Bind thaidata-db with variable name DB

Deploy

# Push to main — Cloudflare Pages auto-deploys
git add -A && git commit -m "your message" && git push origin main

Claim System Flow

1. Owner visits /claim/, searches for their restaurant
2. Submits claim with name, email, phone, verification notes
3. Admin reviews at /admin/ (password protected)
4. On approval: listing created, owner receives magic link email via Resend
5. Owner edits at /listing/manage?token=... — description, photos, hours, etc.

Security:

• Magic link tokens: single-use, SHA-256 hashed, 1-hour expiry
• Admin sessions: 24-hour expiry
• Rate-limited admin login: 5 attempts / 15 min per IP

Data Sources

• Food Standards Agency (FSA) — hygiene ratings and establishment data via FHRS API
• Food Hygiene Information Scheme (FHIS) — Scotland data
• Manual menu price surveys — 75 dish-price pairs across 5 cities
• ONS 2024 mid-year population estimates (for city per-capita calculations)

License

MIT — data sourced from FSA (Open Government Licence).