How meghayadavv/codepair works
Project Type and Purpose
• CodePairingPro is a full-stack single-page web application.
• Purpose: provide real-time code collaboration, AI-driven code suggestions, fixes, and explanations via OpenAI’s APIs.
File Structure
• Top-level folders:
• client/ – React + TypeScript SPA
• server/ – Node.js/TypeScript API server & real-time services
• shared/ – Shared TypeScript schema (Drizzle ORM models)
• Config files: Drizzle ORM config (drizzle.config.ts), Vite configs (vite.config.ts, server/vite.ts), tsconfig.json, tailwind.config.ts.
README for Stack Details
• Frontend: React, Vite, Tailwind CSS, custom UI library under client/src/components/ui. Uses PeerJS for video and WebSockets for real-time code sync.
• Backend: Node.js (server/index.ts), Express-style routes (server/routes.ts), wrappers for OpenAI (server/openai.ts), storage layer (server/storage.ts) using Drizzle ORM and relational DB.
• External Dependencies: OpenAI API, PeerJS signaling, database (PostgreSQL or SQLite), npm/yarn package manager.
Main Components
a. Frontend SPA (browser)
• React App (client/src/main.tsx → App.tsx)
• UI components: CodeEditor, Terminal, VideoChat, Dashboard, Navbar, Sidebar, modals, etc.
• Client libraries: openai.ts, peerjs.ts, websocket.ts, queryClient.ts.
b. Backend API Server
• HTTP REST endpoints (server/routes.ts) for code analysis, suggestions, fixes, explanations.
• WebSocket server (server/index.ts + websocket.ts) for real-time code sync and session management.
• PeerJS signaling (client/lib/peerjs.ts + server for NAT traversal).
• OpenAI service wrapper (server/openai.ts).
• Storage layer (server/storage.ts) interacting with DB via Drizzle ORM (shared/schema.ts).
c. Database
• Relational DB for persisting sessions, users, history (configured in drizzle.config.ts).
d. External Services
• OpenAI API
• PeerJS Signaling Service (or built-in signaling server)
e. Developer Tools / Build Tools
• Vite (client & server setup), Tailwind CSS, environment variables.
Relationships and Data Flows
• Browser → HTTP REST → Backend → OpenAI API → Backend → Browser
• Browser ↔ WebSocket → Backend (real-time code sync & session events)
• Browser ↔ PeerJS p2p connections (video/audio streams) via signaling server
• Backend ↔ Database (CRUD for sessions, users, histories)
Architectural Patterns & Principles
• Separation of concerns: UI layer vs. API layer vs. storage layer
• Client-side state management: TanStack Query (queryClient.ts) for caching
• Real-time collaboration: WebSockets + PeerJS p2p
• Modular code: shared schema for type safety across client and server
• API wrappers: isolation of OpenAI calls in lib modules