Backend Noga 🎇

Backend service untuk aplikasi analisis nutrisi makanan menggunakan OCR dan AI. Service ini menangani upload gambar label makanan dan koordinasi dengan Python AI service untuk pemrosesan OCR.

Tech Stack

• Runtime: Node.js
• Framework: Express.js
• Database: MySQL + Prisma ORM
• File Upload: Multer
• Environment: dotenv

Instalasi

1. Clone repository dan masuk ke direktori:

cd Noga-BE

2. Install dependencies:

npm install

3. Setup environment variables:

cp .env.example .env

Edit file .env dan sesuaikan:

PORT=3000
DATABASE_URL=mysql://user:password@host:port/database

4. Buat database MySQL:

PENTING: Database MySQL harus dibuat manual terlebih dahulu. Prisma hanya auto-create tables, bukan database.

-- Login ke MySQL
mysql -u root -p

-- Buat database
CREATE DATABASE nama_database;

5. Setup akan otomatis saat pertama kali run (skip ke step Menjalankan Server)

Menjalankan Server

Development (Auto-setup tables)

Otomatis generate Prisma client dan sync schema ke database, lalu start server:

npm run dev

Production

npm start

Server akan berjalan di http://localhost:3000

Catatan: Saat pertama kali npm install, Prisma client akan otomatis ter-generate (postinstall hook)

API Endpoints

POST /api/image/packaged-food

Upload gambar untuk makanan kemasan berlabel.

Form Data:

• composition: File gambar komposisi/ingredient
• nutrition_info: File gambar informasi nutrisi
• sessionid: Session ID (string)

Response:

{
  "composition": "http://localhost:3000/uploads/composition/composition-xxx.jpg",
  "nutrition_info": "http://localhost:3000/uploads/nutrition_info/nutrition_info-xxx.jpg",
  "sessionid": "session-123"
}

POST /api/image/prepared-food

Upload gambar untuk makanan siap saji tanpa label.

Form Data:

• foods: File gambar makanan
• sessionid: Session ID (string)

Response:

{
  "foods": "http://localhost:3000/uploads/foods/foods-xxx.jpg",
  "sessionid": "session-123"
}

Struktur Database

Table ocr_table:

• sessionid (String, Primary Key): ID unik untuk setiap sesi
• status (String): Status pemrosesan
• ingredients (Text): JSON data ingredient
• nutrition_info (Text): JSON data informasi nutrisi

Struktur Project

Noga-BE/
├── routes/          # Definisi endpoint API
├── controller/      # Logic handling request
├── function/        # Utility functions
├── prisma/          # Database schema
├── uploads/         # Temporary file storage
├── app.js           # Express configuration
└── bin/www          # Server startup

Database Commands

# Generate Prisma client
npx prisma generate

# Create migration
npx prisma migrate dev --name nama_migrasi

# Open database GUI
npx prisma studio

# Push schema tanpa migration
npx prisma db push

Docker

Build Docker Image

docker build -t noga-backend .

Run Docker Container

docker run -p 3000:3000 \
  -e DATABASE_URL="mysql://user:password@host:port/database" \
  -e PORT=3000 \
  noga-backend

Docker Compose (Optional)

Buat file docker-compose.yml:

version: '3.8'
services:
  app:
    build: .
    ports:
      - "3000:3000"
    environment:
      - DATABASE_URL=${DATABASE_URL}
      - PORT=3000
    volumes:
      - ./uploads:/app/uploads

Run dengan:

docker-compose up

Deploy ke Railway

Setup Database MySQL di Railway

1. Buka Railway dan login

2. Klik "New Project" → pilih "Provision MySQL"

3. Setelah MySQL service dibuat, klik service tersebut

4. Buka tab "Variables" → copy nilai DATABASE_URL

   Contoh format:

   mysql://root:RKgrphVeECnkSZVEopgoeMmbKNKSAkSB@maglev.proxy.rlwy.net:20862/railway
   

Deploy Backend

1. Push code ke GitHub repository

2. Di Railway project yang sama, klik "New" → "GitHub Repo"

3. Pilih repository Noga-BE dan connect

4. Tambahkan Environment Variables:

   • DATABASE_URL: Paste connection string dari MySQL service
   • PORT: Railway akan auto-set (tidak perlu diisi manual)

5. Railway akan otomatis detect Dockerfile dan deploy

6. Setelah deploy selesai, klik service → tab "Settings" → "Generate Domain" untuk mendapatkan public URL

Tips Railway:

• Railway otomatis detect Dockerfile
• Bisa pakai Railway MySQL add-on atau external MySQL
• Environment variables diset di Settings → Variables
• Auto-deploy setiap push ke branch main/master
• Check logs di dashboard jika ada error

Resources

• Dokumentasi Endpoint: Link
• Task Flow: Link