coex 0.1.19

Execute code snippets in isolated Docker environments with multi-language support and security protection

coex - Code Execution Library

coex is a Python library that executes code snippets using subprocess calls on the host system. It provides secure, multi-language code execution with comprehensive validation and testing capabilities.

Features

• Host System Execution: Uses subprocess to run language tools directly on the host
• Security First: Protection against destructive file operations and malicious code
• Multi-Language Support: Python, JavaScript, Java, C++, C, Go, Rust, and more
• Multiple Execution Modes: Input/output validation, function comparison, simple execution
• High Performance: Direct subprocess execution without container overhead
• Robust Error Handling: Comprehensive error handling and timeout management
• Comprehensive Testing: Extensive test suite with security and integration tests

Quick Start

Installation

pip install coex

Prerequisites

• Python 3.8 or higher
• Language tools installed on host system:
  • Java: javac and java (OpenJDK 8+ or Oracle JDK)
  • JavaScript: node (Node.js 14+)
  • C/C++: gcc and g++ (GCC 7+)
  • Go: go (Go 1.16+)
  • Rust: rustc (Rust 1.50+)
• Docker (optional, for sandboxed execution)

Basic Usage

import coex

# Method 1: Input/Output validation (explicit mode)
result = coex.execute(
    mode="answer",
    inputs=[0, 1, 2, 3],
    outputs=[1, 2, 3, 4],
    code="def add_one(x): return x + 1",
    language="python"
)
# Returns: [1, 1, 1, 1] (boolean array indicating pass/fail for each test case)

# Method 2: Function comparison (explicit mode)
result = coex.execute(
    mode="function",
    answer_fn="def say_hello(): return 'hello world'",
    code="def hello(): return 'hello world'",
    language="python"
)
# Returns: [1] (boolean indicating if functions return same value)

# Method 3: Backward compatibility (auto-detection)
result = coex.execute(
    inputs=[1, 2, 3],
    outputs=[2, 4, 6],
    code="def double(x): return x * 2",
    language="python"
)
# Returns: [1, 1, 1] (auto-detects "answer" mode)

# Method 4: Simple execution
result = coex.execute(code="print('Hello, World!')", language="python")
# Returns: [1] (boolean indicating successful execution)

# Docker cleanup
coex.rm_docker()  # Remove cached Docker containers

Supported Languages

Language	File Extension	Docker Image	Aliases
Python	.py	python:3.11-slim	py
JavaScript	.js	node:18-slim	js
Java	.java	openjdk:11-slim
C++	.cpp	gcc:latest	c++, cxx
C	.c	gcc:latest
Go	.go	golang:1.19-slim
Rust	.rs	rust:slim	rs

API Reference

coex.execute()

Execute code snippets in isolated Docker environments.

Parameters:

• inputs (List[Any], optional): List of input values for testing
• outputs (List[Any], optional): List of expected output values
• code (str, optional): Code to execute
• answer_fn (str, optional): Reference function code for comparison
• language (str, default="python"): Programming language
• timeout (int, optional): Execution timeout in seconds
• mode (str, optional): Execution mode ("answer" for input/output validation, "function" for function comparison, None for auto-detection)

Returns:

• List[int]: List of integers (0 or 1) indicating pass/fail for each test case

Raises:

• SecurityError: If dangerous code is detected
• ValidationError: If input validation fails
• ExecutionError: If code execution fails
• DockerError: If Docker operations fail

New in v0.1.0:

• Explicit Mode Parameter: Use mode="answer" or mode="function" for explicit execution modes
• 3-Second Timeout: Each test case has a 3-second timeout limit
• Unique File Naming: Automatic unique file naming for batch processing support

coex.rm_docker()

Remove all cached Docker containers.

Parameters: None

Returns: None

Execution Modes

1. Input/Output Validation Mode

Test code against specific input/output pairs:

inputs = [1, 2, 3, 4, 5]
outputs = [1, 4, 9, 16, 25]
code = """
def square(x):
    return x * x
"""

result = coex.execute(inputs=inputs, outputs=outputs, code=code)
# Tests: square(1)==1, square(2)==4, square(3)==9, etc.

2. Function Comparison Mode

Compare two functions to see if they produce the same output:

answer_fn = """
def fibonacci(n):
    if n <= 1:
        return n
    return fibonacci(n-1) + fibonacci(n-2)
"""

code = """
def fibonacci(n):
    a, b = 0, 1
    for _ in range(n):
        a, b = b, a + b
    return a
"""

result = coex.execute(answer_fn=answer_fn, code=code)
# Compares if both functions return the same value

3. Simple Execution Mode

Execute code and check if it runs without errors:

code = """
import math
print(f"Pi is approximately {math.pi:.2f}")
"""

result = coex.execute(code=code, language="python")
# Returns [1] if execution succeeds, [0] if it fails

Security Features

coex includes comprehensive security protection:

Blocked Operations

• File system operations (rm, mkdir, chmod, etc.)
• Network operations (wget, curl, ssh, etc.)
• System operations (sudo, su, etc.)
• Dangerous imports (os, subprocess, sys, etc.)
• Code evaluation (eval, exec, __import__, etc.)

Example Security Protection

dangerous_code = """
import os
os.system("rm -rf /")  # This will be blocked!
"""

result = coex.execute(code=dangerous_code)
# Returns [0] - execution blocked for security

Multi-Language Examples

Python

code = """
def greet(name):
    return f"Hello, {name}!"
"""
result = coex.execute(code=code, language="python")

JavaScript

code = """
function greet(name) {
    return `Hello, ${name}!`;
}
"""
result = coex.execute(code=code, language="javascript")

Java

code = """
public class Greeter {
    public String greet(String name) {
        return "Hello, " + name + "!";
    }
}
"""
result = coex.execute(code=code, language="java")

C++

code = """
#include <iostream>
#include <string>

std::string greet(std::string name) {
    return "Hello, " + name + "!";
}
"""
result = coex.execute(code=code, language="cpp")

Configuration

Environment Variables

• COEX_DOCKER_TIMEOUT: Docker operation timeout (default: 300 seconds)
• COEX_DOCKER_MEMORY_LIMIT: Container memory limit (default: "512m")
• COEX_EXECUTION_TIMEOUT: Code execution timeout (default: 30 seconds)
• COEX_TEMP_DIR: Temporary directory for files (default: "/tmp/coex")
• COEX_DISABLE_SECURITY: Disable security checks (default: False)

Custom Configuration

from coex.config.settings import settings

# Modify settings
settings.set("execution.timeout", 60)
settings.set("docker.memory_limit", "1g")
settings.set("security.enable_security_checks", False)

Error Handling

coex provides comprehensive error handling:

try:
    result = coex.execute(code="invalid syntax", language="python")
except coex.SecurityError as e:
    print(f"Security violation: {e}")
except coex.ValidationError as e:
    print(f"Validation error: {e}")
except coex.ExecutionError as e:
    print(f"Execution failed: {e}")
except coex.DockerError as e:
    print(f"Docker error: {e}")

Performance Tips

1. Container Reuse: Containers are cached automatically for better performance
2. Batch Operations: Process multiple test cases in single calls when possible
3. Timeout Management: Set appropriate timeouts for your use case
4. Memory Limits: Adjust memory limits based on your code requirements
5. Cleanup: Use coex.rm_docker() to clean up containers when done

Development

Running Tests

# Install development dependencies
pip install -e ".[dev]"

# Run all tests
pytest

# Run specific test categories
pytest tests/test_security.py -v
pytest tests/test_integration.py -v -m integration

# Run with coverage
pytest --cov=coex --cov-report=html

Code Quality

# Format code
black coex/ tests/

# Lint code
flake8 coex/ tests/

# Type checking
mypy coex/

Contributing

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Make your changes
4. Add tests for your changes
5. Ensure all tests pass (pytest)
6. Commit your changes (git commit -m 'Add amazing feature')
7. Push to the branch (git push origin feature/amazing-feature)
8. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Changelog

v0.1.0 (Initial Release)

• Docker-based code execution
• Multi-language support (Python, JavaScript, Java, C++, C, Go, Rust)
• Security protection against dangerous operations
• Input/output validation mode
• Function comparison mode
• Simple execution mode
• Comprehensive test suite
• Container caching for performance
• Robust error handling

Support

• 📖 Documentation
• 🐛 Issue Tracker
• 💬 Discussions

Acknowledgments

• Docker for containerization technology
• The Python community for excellent tooling and libraries
• Contributors and testers who helped make this project possible