ToMenu  

Structured lunch menus from Slovak restaurants, served as a clean REST API.

ToMenu scrapes daily lunch menus from Slovak restaurants and exposes them via a JSON API. Filter by city, day, allergens, price, and delivery availability. Built to be self-hostable and easy to extend with new cities or scraper sources.

Currently covering: Levice 🇸🇰 — more cities as the project expands.

---

Project structure

tomenu/
├── api.py                      # FastAPI REST API + admin dashboard
├── main.py                     # CLI — key management & scrape log
├── scrapeAll.sh                # runs all scrapers in sequence
├── start.sh                    # container entrypoint
├── requirements.txt
├── Dockerfile
├── docker-compose.yml
├── .gitignore
├── scrapers/
│   ├── db.py                   # shared DB helpers for all scrapers
│   └── namenu.scrape.py        # namenu.sk scraper (multi-city, full week)
├── static/                     # static assets served by the API
└── webUI/
    ├── index.html              # admin dashboard UI
    ├── favicon.ico
    └── locales/
        ├── en.json
        ├── sk.json
        └── cs.json

---

How it works

On startup the container immediately runs a full scrape so the DB is never empty on first boot. After that, a cron job re-scrapes at 06:00, 12:00, 18:00, and 00:00 every day. You can also trigger a scrape manually via the admin dashboard or CLI.

Two separate SQLite databases are used:

File	Contains
main.db	API keys, scrape audit log
namenu.db	Cities, restaurants, menus, scrape runs

This keeps auth data separate from scraped data — you can wipe namenu.db without touching your keys.

---

API

Base URL: https://api.tomenu.sk

All requests require an Authorization header containing your API key.

GET /api/levice/menu
Authorization: your-api-key-here

Endpoints

Method	Path	Description
GET	/api/cities	List all available cities
GET	/api/{city}/week	Which days have data this week
GET	/api/{city}/restaurants	Restaurants for a city on a given date
GET	/api/{city}/restaurants/{slug}	Full menu for one restaurant
GET	/api/{city}/menu	All dishes, filterable

/api/{city}/menu query params

Param	Type	Description
date	YYYY-MM-DD	Menu date, defaults to today
type	soup | main | dessert	Dish type filter
delivery	bool	Delivery-only restaurants
max_price	float	Max price in EUR
exclude_allergens	1,7,14	EU allergen numbers to exclude (comma-separated)
limit	int	Max results (default 50, max 200)
offset	int	Pagination offset

Full API docs: tomenu.sk/api

---

Getting an API key

ToMenu uses API keys to prevent abuse. Keys are free and issued manually.

Visit tomenu.sk/api, scroll to the bottom, and use the email template button. Alternatively email contact@tomenu.sk directly.

---

Self-hosting with Docker

The recommended way to run ToMenu is via the published Docker image.

Quick start

docker run -d \
  --name tomenu-api \
  -p 2332:2332 \
  -e PORT=2332 \
  -e MAIN_DB=/app/data/main.db \
  -e NAMENU_DB=/app/data/namenu.db \
  -v tomenu_data:/app/data \
  ghcr.io/toomcis/tomenu:latest

With docker-compose (recommended)

services:
  tomenu:
    image: ghcr.io/toomcis/tomenu:latest
    restart: unless-stopped
    container_name: tomenu-api
    environment:
      - PORT=2332
      - MAIN_DB=/app/data/main.db
      - NAMENU_DB=/app/data/namenu.db
    ports:
      - "2332:2332"
    volumes:
      - tomenu_data:/app/data

volumes:
  tomenu_data:

docker compose up -d

On first boot, watch the logs for your admin API key:

docker logs tomenu-api

==============================
 ToMenu ready
==============================
API key created for 'admin':
  <your-key-here>
Save this — it won't be shown again.
==============================

Environment variables

Variable	Default	Description
PORT	8000	Port the API listens on
MAIN_DB	main.db	Path to the auth database
NAMENU_DB	namenu.db	Path to the menu data database

⚠️ Keep your image up to date

ToMenu is under active development. Always pull the latest image before reporting bugs.

docker compose pull && docker compose up -d

---

Admin dashboard

The admin dashboard is available at http://localhost:2332/ and is intentionally not exposed publicly. If proxying behind nginx or Caddy, restrict access to / and /admin/* to local IPs only.

Example nginx rule:

server {
    server_name api.tomenu.sk;

    location ~ ^/(admin.*|)$ {
        allow 127.0.0.1;
        allow 192.168.0.0/16;
        deny all;
    }

    location / {
        proxy_pass http://127.0.0.1:2332;
        proxy_set_header Host $host;
        proxy_set_header X-Real-IP $remote_addr;
    }
}

---

Local development

git clone https://github.com/toomcis/tomenu.git
cd tomenu

pip install -r requirements.txt

# create your first API key
python main.py --add-key "dev"

# scrape today's menus
python -X utf8 scrapers/namenu.scrape.py --today

# start the API
uvicorn api:app --reload
# → http://127.0.0.1:8000

Scraper commands

./scrapeAll.sh                        # full week, all sources
./scrapeAll.sh --today                # just today
./scrapeAll.sh --day pondelok         # specific day

python -X utf8 scrapers/namenu.scrape.py --today
python -X utf8 scrapers/namenu.scrape.py --day streda

Slovak day slugs: pondelok utorok streda stvrtok piatok

Key management (CLI)

python main.py --add-key "label"      # create a key
python main.py --list-keys            # list all keys
python main.py --revoke-key 3         # revoke key #3
python main.py --scrape-log           # show recent scrape audit log

---

Adding a new scraper source

1. Create scrapers/yoursite.scrape.py
2. Import the shared DB helpers:

   from scrapers.db import connect, init_db, get_or_create_city, upsert_restaurant, upsert_scrape_run

3. Set SOURCE = "yoursite" so runs are tracked separately
4. Add a line to scrapeAll.sh:

   python -X utf8 scrapers/yoursite.scrape.py $ARGS

---

Database schema

main.db — auth & audit

api_keys    id, key_hash, label, created_at, last_used, active
scrape_log  id, source, started_at, finished_at, status, items, error

namenu.db — menu data

cities        id, name, slug, url
restaurants   id, city_id, name, slug, url, address, phone, delivery, info
scrape_runs   id, city_id, source, scraped_at, day, date
              UNIQUE(city_id, source, date)
menu_items    id, restaurant_id, scrape_run_id, type, name, description,
              weight, price_eur, menu_price, allergens, nutrition, raw

---

Data & privacy

ToMenu currently does not collect any user data. It only stores scraped restaurant and menu information from public sources.

A future update will introduce opt-in anonymous usage data collection to enable better recommendations. This will always be opt-in, clearly documented, and when self-hosting, all data stays entirely on your own machine.

---

Roadmap

• Docker image + compose setup
• Multi-city support
• Admin dashboard with scrape history
• Allergen and price filtering
• EN / SK / CS localization
• User accounts with opt-in preferences
• Personalised feed + swipe discovery
• More cities
• Additional scraper sources
• Webhook support

---

Contributing

PRs welcome, especially for:

• Translations — improve or fix SK / CS / EN strings in webUI/locales/
• New cities — city portals with structured lunch menus
• New scraper sources — Slovak or Czech lunch aggregators
• Parser improvements — edge cases in the namenu scraper

Open an issue before starting anything large.

---

License

MIT — do whatever, just don't pretend you made it.

---

Made in Levice 🇸🇰 by toomcis