Routine - AI-Powered Nutrition Tracking

A beautiful iOS app for tracking meals, calories, and water intake with AI food recognition.

Features

• Sign in with Apple - Secure authentication
• AI Food Recognition - Scan meals with your camera
• Real-time Sync - Data syncs across all your devices
• Nutrition Tracking - Calories, protein, carbs, and fat
• Water Intake - Track daily hydration
• Meal Photos - Visual food diary

Tech Stack

• SwiftUI - Modern declarative UI
• Firebase - Backend services
  • Authentication (Sign in with Apple)
  • Firestore (Real-time database)
  • Storage (Food photos)
  • Analytics & Crashlytics
• MVVM Architecture - Clean separation of concerns

Firebase Setup Instructions

1. Create Firebase Project

1. Go to Firebase Console
2. Click "Add project"
3. Name it "Routine" (or your preferred name)
4. Follow the setup wizard

2. Register iOS App

1. In Firebase Console, click "Add app" → iOS
2. Enter bundle ID: com.yourcompany.Routine (or your actual bundle ID)
3. Download GoogleService-Info.plist
4. Add the file to your Xcode project (drag into Routine/ folder)
5. Make sure "Copy items if needed" is checked

Note: GoogleService-Info.plist is gitignored for security. A template is available at GoogleService-Info.plist.template for reference.

3. Add Firebase SDK via Swift Package Manager

1. In Xcode: File → Add Package Dependencies
2. Enter URL: https://github.com/firebase/firebase-ios-sdk
3. Click "Add Package"
4. Select these products:
   • FirebaseAuth
   • FirebaseFirestore
   • FirebaseFirestoreSwift
   • FirebaseStorage
   • FirebaseAnalytics
   • FirebaseCrashlytics
   • FirebaseFunctions (optional, for server-side food recognition)

4. Enable Sign in with Apple

In Firebase Console:

1. Go to Authentication → Sign-in method
2. Enable "Apple" provider
3. Add your Apple Developer Team ID
4. Configure Service ID and Key (from Apple Developer portal)

In Xcode:

1. Select project → Signing & Capabilities
2. Click "+ Capability"
3. Add "Sign in with Apple"

In Apple Developer Portal:

1. Go to Certificates, Identifiers & Profiles
2. Create a Service ID for Sign in with Apple
3. Configure return URL: https://routine-xxxxx.firebaseapp.com/__/auth/handler
4. Create and download private key

5. Set Up Firestore Database

1. In Firebase Console, go to Firestore Database
2. Click "Create database"
3. Start in Test mode (for development)
4. Choose a location

Security Rules (for production):

rules_version = '2';
service cloud.firestore {
  match /databases/{database}/documents {
    match /users/{userId} {
      allow read, write: if request.auth != null && request.auth.uid == userId;
    }
    
    match /foodEntries/{entryId} {
      allow read, write: if request.auth != null && 
        request.auth.uid == resource.data.userId;
      allow create: if request.auth != null && 
        request.auth.uid == request.resource.data.userId;
    }
    
    match /waterEntries/{entryId} {
      allow read, write: if request.auth != null && 
        request.auth.uid == resource.data.userId;
      allow create: if request.auth != null && 
        request.auth.uid == request.resource.data.userId;
    }
  }
}

6. Set Up Firebase Storage

1. In Firebase Console, go to Storage
2. Click "Get started"
3. Start in Test mode (for development)

Security Rules (for production):

rules_version = '2';
service firebase.storage {
  match /b/{bucket}/o {
    match /{userId}/food/{allPaths=**} {
      allow read, write: if request.auth != null && request.auth.uid == userId;
    }
  }
}

7. Uncomment Firebase Code

Once Firebase SDK is added, uncomment the Firebase imports and code in:

Routine/RoutineApp.swift:

import FirebaseCore

// In init():
FirebaseApp.configure()

Security Notes

• GoogleService-Info.plist is gitignored and should never be committed to public repositories
• This file contains your Firebase project configuration and API keys
• While the API key in this file is a public identifier (not a secret), it's best practice to keep it private
• Firebase security comes from Security Rules in Firestore and Storage, not from hiding this file
• For production apps, configure Security Rules to restrict access to authenticated users only (examples provided above)

Routine/Services/FirebaseManager.swift:

• Uncomment the Firebase implementation
• Remove the stub class

Routine/Services/AuthService.swift:

• Uncomment the Firebase Auth implementation
• Remove the stub class

Routine/Services/FirestoreService.swift:

• Replace stub methods with actual Firestore calls

Routine/Services/StorageService.swift:

• Replace stub methods with actual Storage calls

Project Structure

Routine/
├── RoutineApp.swift
├── ContentView.swift
├── Models/
│   ├── User.swift
│   ├── FoodEntry.swift
│   ├── WaterEntry.swift
│   └── DailySummary.swift
├── ViewModels/
│   ├── AuthViewModel.swift
│   ├── FoodLogViewModel.swift
│   └── WaterViewModel.swift
├── Views/
│   └── Auth/
│       └── LoginView.swift
├── Services/
│   ├── FirebaseManager.swift
│   ├── AuthService.swift
│   ├── FirestoreService.swift
│   ├── StorageService.swift
│   └── FoodRecognitionService.swift
└── GoogleService-Info.plist (you add this)

Running the App

Without Firebase (Demo Mode)

The app runs with mock data for UI development and testing.

With Firebase (Full Features)

1. Complete all setup steps above
2. Uncomment Firebase code
3. Build and run on device or simulator
4. Sign in with Apple (requires physical device for full test)

Food Recognition

The app includes real AI-powered food recognition using OpenAI's GPT-4 Vision API!

Quick Start

1. Get an OpenAI API key from https://platform.openai.com/api-keys
2. Copy Routine/Config.plist.template to Routine/Config.plist
3. Add your API key to Config.plist
4. Build and run - take a photo of food and get instant nutrition analysis!

Features

• Real AI Analysis: GPT-4 Vision identifies foods and estimates portions
• Detailed Nutrition: Calories, protein, carbs, fat, sugar, fiber, sodium
• Ingredient Breakdown: See each component of your meal
• Smart Analysis: Get health insights and suggestions
• Firebase Persistence: All data synced to Firestore and Storage

Setup Options

Option 1: OpenAI Direct (Development)

• Quick setup with Config.plist
• API key stored locally (gitignored)
• See PHOTO_ANALYSIS_SETUP.md for details

Option 2: Firebase Functions (Production)

• Secure server-side processing
• API key never exposed to clients
• Better for production apps
• See PHOTO_ANALYSIS_SETUP.md for deployment guide

Option 3: Mock Mode

• No API key required
• Uses sample data for testing
• Automatically falls back if API unavailable

For detailed setup instructions, see PHOTO_ANALYSIS_SETUP.md.

Alternative Services

You can also integrate:

1. Firebase ML Kit - On-device image labeling
2. Google Cloud Vision API - Via Firebase Functions
3. Edamam Food Database - Nutrition data API
4. Nutritionix API - Food database with NLP
5. Clarifai - Food recognition API

Requirements

• iOS 17.0+
• Xcode 15.0+
• Swift 5.9+
• Apple Developer Account (for Sign in with Apple)

License

MIT License - feel free to use this project as a starting point for your own apps.

Support

For issues with:

• Firebase Setup: Check Firebase documentation
• Sign in with Apple: Verify Apple Developer configuration
• App Code: Open an issue on GitHub