How kedro-org/kedro-viz works
Main Components
a) Frontend Layer:
• React Application Components (/src/components)
• State Management (Redux store - /src/store)
• GraphQL Client (Apollo - /src/apollo)
• UI Components (/src/components/ui)
• Visualization Components (Flowchart, Minimap)
b) Backend Layer:
• API Layer (/package/kedro_viz/api)
• GraphQL API (/api/graphql)
• REST API (/api/rest)
• Data Access Layer (/package/kedro_viz/data_access)
• Integration Layer (/package/kedro_viz/integrations)
• Server Component (/package/kedro_viz/server.py)
c) Storage:
• SQLite Database (for session storage)
• File System (for data persistence)
Component Relationships
• Frontend to Backend communication via GraphQL/REST APIs
• Data flow between Data Access Layer and Storage
• Integration points with Kedro projects
• State management flow in the frontend
Specific Implementation Details
Frontend Section:
[React Components] → [Redux Store] → [Apollo Client]
• ↓ ↑ ↓
[UI Components] [State Management] [GraphQL/REST APIs]
Backend Section:
[GraphQL/REST APIs] → [Data Access Layer] → [Repositories]
• ↓ ↓ ↓
[Server Component] → [Integration Layer] → [Storage Layer]
Additional Elements
Include:
• Deployment options (AWS, Azure, GCP)
• Authentication flow (if applicable)
• Data processing pipeline visualization
• Experiment tracking components
• Modular pipeline structure
[Frontend Layer]
• ↓
[API Layer]
• ↓
[Backend Layer]
• ↓
[Storage Layer]
Notes for Implementation
• Version compatibility information
• Scaling points and extensibility options
• Security boundaries
• Public vs. private interfaces
• The separation between frontend and backend
• The role of GraphQL/REST APIs in communication
• Data flow through the system
• Integration points with Kedro projects
• Visualization capabilities and data processing