Dr.Lt.Data
573742e800
The package installed last provides cv2. The repair ranked the packages with the GUI above the headless ones, so on a Linux without the graphics libraries it replaced a working headless cv2 with one that cannot be imported. - Install a headless package after one with the GUI, and within each kind the one with the contrib modules last: opencv-python, opencv-contrib-python, opencv-python-headless, opencv-contrib-python-headless - The package that ends up providing cv2 follows from this order alone, not from which package was the older one - Update the tests and correct the CHANGELOG sentence
ComfyUI-Manager: Core Backend (glob)
This directory contains the Python backend modules that power ComfyUI-Manager, handling the core functionality of node management, downloading, security, and server operations.
Directory Structure
- glob/ - code for new cacheless ComfyUI-Manager
- legacy/ - code for legacy ComfyUI-Manager
Core Modules
- manager_core.py: The central implementation of management functions, handling configuration, installation, updates, and node management.
- manager_server.py: Implements server functionality and API endpoints for the web interface to interact with the backend.
Specialized Modules
- share_3rdparty.py: Manages integration with third-party sharing platforms.
Architecture
The backend follows a modular design pattern with clear separation of concerns:
- Core Layer: Manager modules provide the primary API and business logic
- Utility Layer: Helper modules provide specialized functionality
- Integration Layer: Modules that connect to external systems
Security Model
The system implements a comprehensive security framework with multiple levels:
- Block: Highest security - blocks most remote operations
- High: Allows only specific trusted operations
- Middle: Standard security for most users
- Normal-: More permissive for advanced users
- Weak: Lowest security for development environments
Implementation Details
- The backend is designed to work seamlessly with ComfyUI
- Asynchronous task queuing is implemented for background operations
- The system supports multiple installation modes
- Error handling and risk assessment are integrated throughout the codebase
API Integration
The backend exposes a REST API via manager_server.py that enables:
- Custom node management (install, update, disable, remove)
- Model downloading and organization
- System configuration
- Snapshot management
- Workflow component handling