🐳 Add GitHub Actions for Automated Docker Image Building

Problem Statement

Currently, the project requires manual Docker image building and distribution. Users need to build images locally using docker-compose.

Proposed Solution

Implement a comprehensive GitHub Actions workflow that automatically builds and publishes Docker images to GitHub Container Registry.

Features to Add

🏗️ Production Dockerfiles

• Dockerfile - Standard CPU-only image with multi-architecture support (AMD64/ARM64)
• Dockerfile.gpu - GPU-enabled image with CUDA 11.8 support (AMD64 only)
• Optimized for production with security best practices

⚙️ GitHub Actions Workflow

• Automated building on push to main/develop branches
• Build testing on pull requests (without publishing)
• Multi-architecture builds (AMD64 + ARM64 for CPU images)
• Automatic tagging based on Git events:
  • latest for main branch
  • develop for develop branch
  • v1.2.3 for version tags
  • pr-123 for pull requests

📦 Container Registry Integration

• Publish to GitHub Container Registry (ghcr.io)
• Proper image tagging and versioning
• Layer caching for faster builds
• Security scanning integration

🚀 Production Configurations

• docker-compose.prod.yml using pre-built images
• .env.prod.example template for production deployments
• Health check endpoints for container monitoring

📚 Documentation

• Comprehensive build guide (DOCKER_BUILD_GUIDE.md)
• Usage examples and troubleshooting
• Architecture support matrix

Benefits

1. 🚀 Faster Deployment: Users can pull pre-built images instead of building locally
2. 🔄 Consistent Builds: All images built in standardized GitHub environment
3. 🛡️ Security: Production hardened images with non-root users
4. 📈 Scalability: Multi-architecture support for different deployment targets
5. 🔧 Developer Experience: Simplified setup for contributors and users
6. 🎯 CI/CD Ready: Foundation for automated testing and deployment

Technical Details

Image Variants

• Standard Image: ghcr.io/catchthetornado/text-extract-api:latest

  • Python 3.10 slim base
  • CPU-only processing
  • Multi-architecture (AMD64/ARM64)

• GPU Image: ghcr.io/catchthetornado/text-extract-api-gpu:latest

  • NVIDIA CUDA 11.8 base
  • GPU-accelerated processing
  • AMD64 architecture only

Workflow Triggers

• Push to main → Build and publish latest images
• Push to develop → Build and publish develop images
• Git tags (v*) → Build and publish versioned images
• Pull requests → Build and test (no publish)

Security Features

• Non-root user execution (appuser)
• Minimal attack surface with slim base images
• Health check endpoints for monitoring
• Secure layer caching and optimization

Implementation Checklist

• Create production Dockerfile
• Create GPU-enabled Dockerfile.gpu
• Implement GitHub Actions workflow (.github/workflows/docker-build.yml)
• Add health check endpoint to FastAPI application
• Create production Docker Compose configuration
• Add comprehensive documentation
• Update .dockerignore for optimized builds
• Create environment template for production
• Test workflow with sample builds

Success Criteria

✅ Automated Builds: Images build automatically on code changes
✅ Multi-Platform: CPU images work on both Intel/AMD and Apple Silicon
✅ GPU Support: GPU images work with NVIDIA Docker runtime
✅ Documentation: Clear setup and usage instructions
✅ Testing: PR builds validate Docker configurations
✅ Production Ready: Images suitable for production deployment