Skip to content

DOCKER.md

Latest commit

 

History

History
100 lines (67 loc) · 1.78 KB

DOCKER.md

File metadata and controls

100 lines (67 loc) · 1.78 KB

Docker Setup for SandboxEnvironment

This repository includes Docker configuration for development, testing, and deployment.

Quick Start

Run Tests

docker-compose run test

Interactive Development

docker-compose run dev

Then inside the container:

python calculator.py
pytest
ruff check .
black .

Run Linting and Formatting

docker-compose run lint

Build Images

Build the Docker image:

docker build -t sandbox:latest .

Run Individual Services

Test Service

docker-compose up test

Development Service (Interactive)

docker-compose run -it dev bash

Lint Service

docker-compose up lint

Dockerfile Details

The Dockerfile uses a multi-stage build:

  1. Builder stage: Installs Python dependencies
  2. Production stage: Strips down to essentials, adds security (non-root user), and includes a health check
  • Python 3.13-slim base image (minimal, ~150MB)
  • Non-root user (appuser) for security
  • Health check validates calculator functionality
  • Default command runs pytest

Docker Compose Services

Service Purpose
dev Interactive development shell with mounted volumes
test Runs pytest with verbose output
lint Runs ruff, black, and other checks

Cleaning Up

Remove containers and images:

docker-compose down
docker image rm sandbox:latest

Environment Variables

Set PYTHONUNBUFFERED=1 to see real-time output from Python applications.

Notes

  • Volume mounts allow live code editing reflected in containers
  • __pycache__ is excluded to prevent host/container conflicts
  • Multi-stage build reduces final image size
  • Health check ensures application stability in production