AdvGenPriceComparer

Combat Illusory Discounts with P2P Price Intelligence & AI

A Windows desktop application for tracking and comparing grocery prices across Australian supermarkets, powered by peer-to-peer data sharing and artificial intelligence.

🎯 Mission

Fight misleading "sale" prices and illusory discounts by building a transparent, community-driven grocery price tracking network. Share real-time price data across a P2P network and leverage AI to automatically extract pricing from supermarket catalogues.

✨ Key Features

🌐 Peer-to-Peer Price Sharing

• Decentralized Network: Share grocery prices directly with other users without central servers
• Server Discovery: Use servers.json configuration to find and connect to P2P nodes
• Regional Filtering: Connect to price-sharing nodes in your region (NSW, VIC, QLD, etc.)
• Real-time Sync: Automatic synchronization of price updates across the network
• Privacy-Focused: Direct peer connections, no central data collection
• Scalable Storage: For large-scale data, use AdvGenNoSqlServer (sister project)

🤖 AI-Powered Catalogue Processing

• LLM Integration: Automatically extract pricing from PDF catalogues using Large Language Models
• ML.Net Support: Machine learning capabilities for price prediction and analysis (planned)
• Smart Matching: AI-assisted product matching across different supermarket chains
• OCR Processing: Extract text from image-based catalogue PDFs

📊 Price Intelligence

• Historical Tracking: Track price changes over time to identify genuine vs. fake discounts
• Multi-Store Comparison: Compare prices across Coles, Woolworths, IGA, Aldi, and other chains
• Price Alerts & Notifications: Get notified when prices drop or deals are about to expire
• Discount Analysis: Identify illusory discounts by comparing current "sale" prices with historical data
• Reports & Analytics: Weekly specials digest and best deals tracking
• Import/Export: JSON import from Coles/Woolworths/Drakes, export with filters and compression

🛒 User Experience & Organization

• Shopping Lists: Create, manage, and track progress of your shopping lists
• Global Search: Search across all items, stores, and price records instantly
• Favorites: Pin your frequently purchased items for quick access
• Settings & Customization: Configure your experience and preferences

🏪 Comprehensive Coverage

• Coles
• Woolworths
• IGA
• Aldi
• Drakes
• Other Australian supermarkets

🏗️ Architecture

┌─────────────────────────────────────────────────────────────────┐
│           WPF Desktop App (WPF-UI Fluent Design)                │
│                     (UI Layer - .NET 9)                          │
│  ┌──────────────┐  ┌──────────────┐  ┌─────────────────────┐   │
│  │   Views      │  │  ViewModels  │  │   Converters        │   │
│  │  • ItemsPage │  │  • MainWindow│  │  • BoolToVisibility │   │
│  │  • StoresPage│  │  • ItemVM    │  │  • InverseBool      │   │
│  │  • Dashboard │  │  • ImportVM  │  │                     │   │
│  └──────────────┘  └──────────────┘  └─────────────────────┘   │
└─────────────────────┬───────────────────────────────────────────┘
                      │
┌─────────────────────┴───────────────────────────────────────────┐
│              AdvGenPriceComparer.Data.LiteDB                     │
│  ┌──────────────┐  ┌──────────────┐  ┌─────────────────────┐   │
│  │ Repositories │  │   Services   │  │     Entities        │   │
│  │  • Items     │  │  • JsonImport│  │  • ItemEntity       │   │
│  │  • Places    │  │  • Database  │  │  • PlaceEntity      │   │
│  │  • Prices    │  │  • Export    │  │  • PriceRecordEntity│   │
│  └──────────────┘  └──────────────┘  └─────────────────────┘   │
└─────────────────────┬───────────────────────────────────────────┘
                      │
┌─────────────────────┴───────────────────────────────────────────┐
│                   AdvGenPriceComparer.Core                       │
│  ┌──────────────┐  ┌──────────────┐  ┌─────────────────────┐   │
│  │   Models     │  │  Interfaces  │  │  NetworkManager     │   │
│  │  • Item      │  │  • IGrocery  │  │  • P2P Server       │   │
│  │  • Place     │  │  • IRepos    │  │  • P2P Client       │   │
│  │  • PriceRec  │  │  • Services  │  │  • Discovery        │   │
│  └──────────────┘  └──────────────┘  └─────────────────────┘   │
└─────────────────────┬───────────────────────────────────────────┘
                      │
┌─────────────────────┴───────────────────────────────────────────┐
│                LiteDB Embedded Database                          │
│    • Items Collection  • Places Collection  • PriceRecords      │
│    • Categories        • Alerts                                  │
└─────────────────────────────────────────────────────────────────┘

┌──────────────────────────────────────────────────────────────┐
│                  Python AI Processing                         │
│  • PDF Catalogue Extraction (LLM-powered)                    │
│  • OCR Processing (pdfplumber, PyPDF2)                       │
│  • Product Categorization                                    │
└──────────────────────────────────────────────────────────────┘

📈 Scalability & Sister Project

AdvGenNoSqlServer - Enterprise-Scale Price Storage

As your pricing data grows beyond the capacity of the embedded LiteDB database, you can seamlessly migrate to AdvGenNoSqlServer, our sister project designed for large-scale data management.

When to Use AdvGenNoSqlServer:

• Millions of price records: LiteDB is great for personal use, but enterprise deployments need more
• Multi-region deployments: Centralized NoSQL database for regional P2P hub servers
• Advanced analytics: Run complex queries across massive historical datasets
• Community hubs: Power community-run price-sharing nodes serving hundreds of users
• API services: Build public APIs for price data access

Features:

• High-performance NoSQL database (MongoDB, Cassandra, or CosmosDB)
• RESTful API for data access
• Horizontal scaling for millions of records
• Advanced indexing and query optimization
• Backup and replication support
• Docker deployment ready

Migration Path:

// Export from LiteDB
var exporter = new PriceDataExporter(groceryDataService);
await exporter.ExportToJson("price_data_export.json");

// Import to AdvGenNoSqlServer
var importer = new NoSqlImporter("https://your-nosql-server.com");
await importer.ImportFromJson("price_data_export.json");

Architecture with AdvGenNoSqlServer:

Desktop App (LiteDB) ──P2P──> Regional Hub (AdvGenNoSqlServer) ──P2P──> Other Hubs
      ↓                              ↓                                     ↓
 Personal Data              Regional Aggregation                   National Network

This hybrid approach lets individuals use lightweight P2P sharing while community leaders can run powerful regional hubs with AdvGenNoSqlServer for aggregated price intelligence.

📚 Documentation

For a comprehensive guide on how to use all features of AdvGenPriceComparer, please refer to our User Guide. It covers everything from navigating the dashboard and managing your shopping lists to importing/exporting data and troubleshooting common issues.

🚀 Getting Started

Prerequisites

For C# Application:

• Windows 10/11
• .NET 9.0 SDK
• Visual Studio 2022 (recommended)

For Python Scripts:

• Python 3.8+
• pip (Python package manager)

Installation

1. Clone the repository

   git clone https://github.com/yourusername/AdvGenPriceComparer.git
   cd AdvGenPriceComparer

2. Build the C# application

   cd AdvGenPriceComparer
   dotnet restore
   dotnet build -p:Platform=x64

3. Setup Python environment

   pip install -r requirements.txt

Running the Application

Option 1: Visual Studio (Recommended)

1. Open AdvGenPriceComparer.sln in Visual Studio 2022
2. Set platform to x64
3. Press F5 to run with debugging or Ctrl+F5 to run without debugging

Option 2: Packaged Deployment

cd AdvGenPriceComparer
dotnet publish -c Release -p:Platform=x64 --self-contained true
# Run from: bin\Release\net9.0-windows10.0.19041.0\win-x64\publish\

🌐 P2P Configuration

Server Discovery with servers.json

The application uses a servers.json configuration file to discover and connect to P2P price-sharing nodes. The file is stored at:

%AppData%\AdvGenPriceComparer\servers.json

Example Configuration:

[
  {
    "name": "AusPriceShare-Sydney",
    "host": "price.aus.example.com",
    "port": 8080,
    "isSecure": false,
    "region": "NSW",
    "description": "Australian grocery price sharing - Sydney region",
    "isActive": true
  },
  {
    "name": "LocalTestServer",
    "host": "localhost",
    "port": 8081,
    "isSecure": false,
    "region": "Local",
    "description": "Local testing server",
    "isActive": false
  }
]

Starting a P2P Server

var networkManager = new NetworkManager(groceryDataService);
await networkManager.StartServer(port: 8081);

Connecting to P2P Network

// Connect to specific server
await networkManager.ConnectToServer("AusPriceShare-Sydney");

// Auto-discover and connect to all active servers in region
await networkManager.DiscoverAndConnectToServers("NSW");

Sharing Prices

await networkManager.SharePrice(
    itemId: "item123",
    placeId: "place456",
    price: 3.99m,
    isOnSale: true,
    originalPrice: 5.99m,
    saleDescription: "50% Off - Special"
);

🤖 AI & Catalogue Processing

Extract Prices from PDF Catalogues

The Python scripts use LLM APIs to automatically extract product information and pricing from supermarket catalogue PDFs:

python pdf_catalog_extractor.py --input coles_catalogue.pdf --output coles_prices.json

Supported Catalogue Sources:

• Coles weekly specials
• Woolworths catalogues
• Aldi special buys
• IGA local catalogues

How AI Extraction Works

1. PDF Processing: Extract text and images from catalogue PDFs
2. LLM Analysis: Send pages to LLM (GPT-4, Claude, etc.) to identify products and prices
3. Structured Output: Generate JSON with product names, prices, brands, and categories
4. Import to Database: Load extracted data into LiteDB for tracking

🤖 ML.NET Integration (Planned)

Future ML.NET capabilities for Phase 9-11:

• Auto-Categorization: ML-powered product categorization during import
• Price Prediction: Forecast future price trends based on historical data
• Anomaly Detection: Identify suspicious price changes or fake discounts
• Smart Buying Recommendations: "Buy Now" vs "Wait" based on price forecasting

💾 Database Structure

LiteDB Collections

Items Collection

{
  "id": "item_001",
  "name": "Milk Full Cream",
  "brand": "Dairy Farmers",
  "category": "Dairy",
  "packageSize": "2L",
  "barcode": "9300632123456"
}

Places Collection

{
  "id": "place_001",
  "name": "Coles Chermside",
  "chain": "Coles",
  "suburb": "Chermside",
  "state": "QLD",
  "latitude": -27.3853,
  "longitude": 153.0356
}

PriceRecords Collection

{
  "id": "record_001",
  "itemId": "item_001",
  "placeId": "place_001",
  "price": 3.99,
  "isOnSale": true,
  "originalPrice": 5.99,
  "dateRecorded": "2026-02-25T10:30:00Z",
  "source": "p2p-network"
}

Database location: %AppData%\AdvGenPriceComparer\GroceryPrices.db

💡 Need More Scale? For large-scale deployments with millions of records, migrate to AdvGenNoSqlServer for enterprise-grade performance and scalability.

📁 Project Structure

AdvGenPriceComparer/
├── AdvGenPriceComparer.WPF/              # WPF Application (.NET 9, WPF-UI)
│   ├── Views/                            # XAML Views (Items, Stores, Dashboard)
│   ├── ViewModels/                       # MVVM ViewModels
│   ├── Services/                         # WPF-specific services
│   └── Converters/                       # XAML value converters
├── AdvGenPriceComparer.Core/             # Core Models & Interfaces
│   ├── Models/                           # Item, Place, PriceRecord
│   ├── Interfaces/                       # Repository interfaces
│   └── Helpers/                          # NetworkManager
├── AdvGenPriceComparer.Data.LiteDB/      # Data Access Layer
│   ├── Repositories/                     # LiteDB implementations
│   ├── Services/                         # JsonImportService, ExportService
│   └── Entities/                         # LiteDB entity classes
├── AdvGenPriceComparer.Tests/            # xUnit Test Suite (217+ tests)
│   ├── Services/                         # JsonImport, ServerConfig tests
│   ├── Repositories/                     # Repository layer tests
│   ├── ViewModels/                       # ViewModel tests
│   └── Integration/                      # End-to-end tests
├── TestConsole/                          # Console test application
├── NetworkTest/                          # P2P network testing
├── pdf_catalog_extractor.py              # LLM-powered PDF extraction
├── coles_catalogue_scraper.py            # Coles-specific scraper
├── woolworths_catalogue_parser.py        # Woolworths-specific parser
├── requirements.txt                      # Python dependencies
└── servers.json                          # Server configuration

🔧 Development

Building for Different Platforms

# Windows x64 (default)
dotnet publish -c Release -r win-x64

# Windows x86
dotnet publish -c Release -r win-x86

# Windows ARM64
dotnet publish -c Release -r win-arm64

Running Tests

dotnet test AdvGenPriceComparer.Tests

Python Environment Setup

# Install dependencies
pip install -r requirements.txt

# Key Python packages:
# - PyPDF2, pdfplumber: PDF processing
# - openai, anthropic: LLM API clients
# - pytesseract: OCR processing

🤝 Contributing

Contributions are welcome! This is a community-driven project to combat deceptive pricing practices.

Ways to contribute:

• Add support for more supermarket chains
• Improve AI extraction accuracy
• Set up regional P2P nodes with AdvGenNoSqlServer
• Add price alert features
• Enhance the UI/UX
• Report bugs and request features
• Help build community price-sharing hubs

📋 Roadmap

Completed ✅

• WinUI 3 Migrated to WPF with Fluent Design (WPF-UI)
• LiteDB integration with repository pattern
• JSON Import/Export (Coles, Woolworths, Drakes formats)
• 217+ xUnit tests with CI/CD pipeline
• P2P networking with server discovery
• LLM-powered catalogue extraction
• Sister project: AdvGenNoSqlServer for large-scale deployments
• Settings Service with JSON persistence
• Global Search functionality
• Price drop notifications and Deal expiration reminders
• Favorite items list
• Shopping list integration
• Reports and analytics (Best deals, weekly digest)

In Progress 🚧

• ML.NET Auto-Categorization (Phase 9)
• ML.NET Price Prediction & Forecasting (Phase 11)

Planned 📅

• AdvGenNoSqlServer integration and migration tools
• Mobile app (Android/iOS)
• Barcode scanning
• Browser extension for online shopping
• Public P2P node infrastructure with regional hubs

⚠️ Known Issues

COM Registration Error (0x80040154)

If you encounter this error, use .NET 8.0 or run via Visual Studio. See CLAUDE.md for detailed troubleshooting.

📝 License

[Add your license here]

🙏 Acknowledgments

• Built to combat misleading pricing practices in Australian supermarkets
• Inspired by consumer advocacy groups fighting illusory discounts
• Community-driven approach to transparent pricing

🔗 Related Projects

AdvGenNoSqlServer

Enterprise-scale NoSQL server for large pricing datasets. When your price data grows beyond embedded database limits, AdvGenNoSqlServer provides:

• Scalable storage for millions of price records
• RESTful API for distributed access
• Regional hub deployment for P2P networks
• Advanced analytics and reporting

Repository: AdvGenNoSqlServer

📞 Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions

---

Made with 💙 for Australian shoppers

Help us build a transparent grocery pricing network - because you deserve to know the real price!