CodeZen Backend

FastAPI + Celery backend for queued code execution with Docker sandboxing.

Visual architecture guides

• Open project_guide.html in a browser for a full beginner-friendly explanation of components.
• Open file_dependency.html in a browser for an in-depth breakdown of how files flow and depend on each other.
• Open real_world_restaurant.html in a browser for a deep real-life restaurant analogy with an interactive dependency graph built using Cytoscape.js (no Mermaid).

Supported languages

• python
• js (or javascript)
• java
• c
• cpp

Run with Docker Compose

1. Build sandbox image and start all services:

./run.sh

or on Windows PowerShell:

./run.ps1

2. API will be available at http://localhost:8000.

API endpoints

POST /submit

Queue execution and return task id.

Example body:

{
  "language": "python",
  "code": "print(input())",
  "stdin": "hello",
  "timeout_seconds": 5
}

GET /result/{task_id}

Get task status and result.

POST /execute

Queue and wait for completion for up to wait_timeout seconds.

Example body:

{
  "language": "cpp",
  "code": "#include <iostream>\nint main(){std::string s; std::getline(std::cin,s); std::cout<<s;}",
  "stdin": "hello cpp",
  "timeout_seconds": 5,
  "wait_timeout": 20
}