ProbablyTasty

An intelligent recipe management application with AI-powered search, self-contained HTML exports, and smart kitchen tools.

✨ Features

Recipe Management

• 📚 Full CRUD Operations - Create, read, update, and delete recipes
• 🏷️ Tag-Based Organization - Organize recipes with custom tags
• 🌳 Tree View Navigation - Browse recipes by categories and tags
• 📸 Image Import - Extract recipes from photos using vision AI
• 🔗 URL Import - Import recipes from websites automatically
• 📤 HTML Export Format - Self-contained HTML

AI-Powered Features

• 🔍 Natural Language Search - Find recipes with conversational queries
• 🤖 Multi-LLM Support - OpenAI, Anthropic, Google, and Ollama
• 📷 Vision AI - Extract recipes from images and PDFs
• 🧠 Smart Ingredient Parsing - Automatically extracts quantities, units, and names

Kitchen Tools

• 🛒 Smart Shopping Lists - Auto-consolidate ingredients across recipes
• ⚖️ Recipe Scaling - Adjust servings with automatic quantity recalculation
• 🔄 Unit Conversion - Convert between metric and imperial systems
• 🖨️ Print-Friendly - Clean printing layout for recipes

Self-Contained HTML Exports

• 🌐 Universal Format - Open in any browser, no app needed
• 📱 Fully Interactive - Built-in servings scaler and unit converter
• 📤 Shareable - Email or cloud-share as single file
• 💾 Re-importable - Import HTML recipes back into the app

User Experience

• 🎨 Dark/Light Themes - Beautiful themes with live preview
• 🖼️ Application Icon - Professional branding
• ⚡ Fast & Responsive - PySide6-powered desktop interface
• 💻 Cross-Platform - Works on Windows, macOS, and Linux
• ⚙️ Easy Configuration - All settings in-app, no config files needed

� System Requirements

Windows

• Microsoft Visual C++ Redistributable - Required for running the application
  • Download the latest version: Microsoft Visual C++ Redistributable
  • Install both x64 and x86 versions if unsure which you need

All Platforms

• Python 3.8 or higher (for running from source)
• Active internet connection (for AI features and URL import)

�🚀 Quick Start

1. Clone the repository

   git clone <your-repo-url>
   cd probablytasty

2. Create a virtual environment

   Linux/Mac:

   python -m venv venv
   source venv/bin/activate

   Windows (Command Prompt):

   python -m venv venv
   venv\Scripts\activate.bat

   Windows (PowerShell):

   python -m venv venv
   venv\Scripts\Activate.ps1

   Note: If PowerShell gives an execution policy error, run:

   Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser

3. Install dependencies

   pip install -r requirements.txt

4. Initialize the database

   python -m src.init_db

5. Run the application

   python -m src.main

   Note: Configure API keys and LLM settings in the app via Settings → Preferences

📁 Project Structure

probablytasty/
├── src/
│   ├── models/           # SQLAlchemy database models
│   ├── services/         # Business logic (recipes, search, LLM)
│   ├── ui/               # PySide6 UI components
│   │   ├── themes/       # Dark/Light QSS stylesheets
│   │   └── dialogs/      # Recipe editor, settings, etc.
│   ├── templates/        # HTML export templates
│   ├── utils/            # Import/export utilities
│   └── main.py           # Application entry point
├── icons/                # Application icons
│   ├── applebiter.png    # Main app icon
│   └── hicolor/          # Multi-resolution icons
├── tests/                # Test suite
├── example_recipe.html   # Sample self-contained recipe
├── BUILD.md              # Build and packaging instructions
├── probablytasty.spec    # PyInstaller configuration
├── installer.iss         # Inno Setup script (Windows)
└── requirements.txt      # Python dependencies

⚙️ Configuration

All configuration is done in-app via Settings → Preferences:

• LLM Provider - Choose OpenAI, Anthropic, Google, or Ollama
• API Keys - Enter your API keys for cloud providers
• Ollama Settings - Configure local Ollama server URL and models
• Theme - Switch between Dark and Light themes (live preview)
• Font Size - Adjust interface text size

Settings are stored in ~/.probablytasty/settings.json

🎯 Usage Examples

Natural Language Search

"quick vegetarian dinner recipes"
"desserts with chocolate"
"recipes under 30 minutes"

Self-Contained HTML Exports

1. Select a recipe
2. File → Export Recipes → Choose "HTML Recipe File"
3. Share the file - recipients can open in any browser!
4. Features include:
   • Interactive servings scaler
   • Unit converter (metric ↔ imperial)
   • Print-friendly layout
   • Embedded recipe data (re-importable)

Shopping List Generation

1. Tools → Generate Shopping List
2. Select multiple recipes
3. Auto-consolidated ingredients by category
4. Copy to clipboard or print

📦 Building Installers

See BUILD.md for detailed instructions on creating:

• Windows installer (.exe) using Inno Setup
• Linux package (.deb) using dpkg
• Standalone executables with PyInstaller

Quick build:

# Install PyInstaller
pip install pyinstaller

# Build executable
pyinstaller probablytasty.spec

# Windows: Create installer with Inno Setup
# Linux: Create .deb with dpkg-deb

🧪 Development

Run tests:

pytest

Format code:

black src/ tests/

License

MIT License - See LICENSE file for details