yogsaa/repolens

Docker Overview

🌐 Live Demo: [***]

RepoLENS uses a multi-stage Docker architecture to package the React frontend and Flask backend into a single production-ready container. This approach reduces image size, simplifies deployment, and ensures a consistent runtime environment across local and cloud platforms.

---

Container Architecture

text
  [React Frontend] (Vite Production Build)
         ↓ (Serves Static HTML/JS/CSS Chunks)
  [Flask Backend] (WSGI Routing & Process Queue)
         ↓ (Executes Codebase & PDF Analysis)
  [LLM Runtime Layer] (Gemini / NVIDIA NIM Models)

---

Multi-Stage Build Process

The Dockerfile follows a two-stage build strategy to optimize the final image size and runtime performance.

Stage 1: Frontend Build

dockerfile
FROM node:20 AS frontend-builder

Responsibilities:

• Install Node.js dependencies
• Execute the Vite production build
• Optimize static assets
• Generate the frontend/dist directory

---

Stage 2: Backend Runtime

dockerfile
FROM python:3.11-slim

Responsibilities:

• Install Python dependencies
• Configure Flask services
• Copy compiled frontend assets
• Load environment variables
• Expose application port 5000

---

Build Workflow

text
  [Source Code] (Vite React + Python Flask code)
         ↓
  [Stage 1: Frontend Build] (Vite compilation in Node.js)
         ↓
  [Static Assets] (frontend/dist production bundles)
         ↓
  [Stage 2: Backend Build] (Assets copied to Flask directory)
         ↓
  [Production Image] (Unified server container exposed on Port 5000)

---

Container Features

• Multi-stage image optimization
• Single-port deployment for frontend and backend
• Environment-based LLM provider selection
• Production-ready static asset serving
• Independent Python and Node.js dependency management
• Portable execution across local and cloud environments

---

Build and Run

Build Docker Image

bash
docker build -t repolens:latest .

Run Container

bash
docker run -p 5000:5000 \
  --env-file ./backend/.env \
  repolens:latest

View Running Containers

bash
docker ps

View Container Logs

bash
docker logs <container_id>

Stop Container

bash
docker stop <container_id>

---

Environment Configuration

env
LLM_PROVIDER=nvidia

API_KEY=<your_api_key>
MODEL=<model_name>

# NVIDIA Configuration
NVIDIA_API_KEY=<your_api_key>
NVIDIA_MODEL=meta/llama-3.1-8b-instruct

---

Deployment Compatibility

The Docker-based deployment model supports:

• Render (Docker Runtime)
• Railway
• Google Cloud Run
• AWS ECS
• Azure Container Apps
• DigitalOcean App Platform
• Self-hosted Docker environments

---