DepKit

DependencyManager

A flexible Python dependency manager that handles runtime dependencies and script imports.

Quick Start

from depkit import DependencyManager

# Simple usage
manager = DependencyManager(requirements=["requests>=2.31.0"])
manager.install()
import requests
# ... do your work ...
manager.uninstall()  # optional cleanup

Recommended Usage (Context Managers)

# Synchronous
with DependencyManager(requirements=["requests"]) as manager:
    import requests
    # ... do your work ...
    # cleanup happens automatically

# Asynchronous
async with DependencyManager(requirements=["requests"]) as manager:
    import requests
    # ... do your work ...
    # cleanup happens automatically

Features

• Simple install/uninstall methods for quick usage
• Context managers for proper resource management
• PEP 723 dependency declaration support
• Support for both pip and uv package managers
• Custom pip index URL support
• Temporary script importing
• Path management for imports

Installation

pip install depkit

Basic Usage

The DependencyManager supports both synchronous and asynchronous context managers:

Async Usage

from depkit import DependencyManager

async with DependencyManager(
    requirements=["requests>=2.31.0", "pandas"],
    prefer_uv=True
) as manager:
    import requests
    import pandas

Sync Usage

from depkit import DependencyManager

with DependencyManager(
    requirements=["requests>=2.31.0", "pandas"],
    prefer_uv=True
) as manager:
    import requests
    import pandas

Working with Scripts

The DependencyManager can handle scripts with PEP 723 dependency declarations:

# example_script.py
# /// script
# dependencies = [
#   "requests>=2.31.0",
#   "pandas>=2.0.0"
# ]
# requires-python = ">=3.12"
# ///

import requests
import pandas as pd

Load and use the script:

async with DependencyManager(
    scripts=["path/to/example_script.py"],
    extra_paths=["."]  # Add paths to Python's import path
) as manager:
    # Script's dependencies are installed automatically
    from example_script import some_function

Configuration Options

DependencyManager(
    prefer_uv: bool = False,          # Prefer uv over pip if available
    requirements: list[str] = None,   # List of PEP 508 requirement specifiers
    extra_paths: list[str] = None,    # Additional Python import paths
    scripts: list[str] = None,        # Scripts to load and process
    pip_index_url: str = None,        # Custom PyPI index URL
)

Features in Detail

UV Integration

The manager automatically detects and can use uv for faster package installation:

manager = DependencyManager(prefer_uv=True)

Custom Package Index

Specify a custom PyPI index:

manager = DependencyManager(
    requirements=["private-package>=1.0.0"],
    pip_index_url="https://private.pypi.org/simple"
)

Path Management

Add custom import paths:

manager = DependencyManager(
    extra_paths=[
        "./src",
        "./lib",
    ]
)

Error Handling

from depkit import DependencyError

try:
    async with DependencyManager(requirements=["nonexistent-package"]):
        pass
except DependencyError as e:
    print(f"Dependency management failed: {e}")

Best Practices

1. Use as a context manager to ensure proper cleanup
2. Specify exact version requirements when possible
3. Use PEP 723 for script dependencies
4. Handle DependencyError exceptions appropriately
5. Consider using uv in production for better performance

Limitations

• Requires Python 3.12 or higher
• Some features may not work on all platforms
• UV support requires uv to be installed separately