Metadata-Version: 2.3
Name: fdleaky
Version: 0.0.9
Summary: A simple IOC / Dependency injection micro framework for Python.
License: MIT
Author: Tim O'Farrell
Author-email: tofarr@gmail.com
Requires-Python: >=3.11,<4.0
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Requires-Dist: fastapi (>=0.115.11,<0.116.0)
Requires-Dist: uvicorn (>=0.34.0,<0.35.0)
Project-URL: Repository, https://github.com/tofarr/fdleaky
Description-Content-Type: text/markdown

# fdleaky

A Python utility for detecting file descriptor leaks in Python applications. fdleaky monitors file and socket operations in your application and reports any resources that remain open longer than expected.

Upon suspicion of unclosed File Descriptors being present, hit `p` in the terminal to print stack traces for the point at which each file descriptor was opened.

## Purpose

File descriptor leaks can be difficult to track down, especially in long-running applications. These leaks can lead to:
- Resource exhaustion
- "Too many open files" errors
- Degraded system performance
- Memory leaks

fdleaky helps you find these issues by:
- Tracking all file and socket operations
- Recording stack traces when resources are opened
- Detecting resources that remain open too long
- Reporting unclosed resources on shutdown

## Installation

```bash
pip install fdleaky
```

## Usage

Run your Python application with fdleaky monitoring:

```bash
# Run a module
python -m fdleaky your_module

# Run a Python file
python -m fdleaky your_script.py

# Run a package with __main__.py
python -m fdleaky your_package

# Run with uvicorn
poetry run python -m fdleaky uvicorn my_app:app
```

When a file descriptor remains open longer than the threshold (default 180 seconds), fdleaky will log:
- The type of resource (file/socket)
- The stack trace from when it was opened
- The time it has been open

## Development

This project uses poetry for dependency management. To get started:

```bash
# Install dependencies
poetry install

# Run tests
poetry run pytest
```

## How it Works

fdleaky works by:
1. Patching built-in file and socket operations
2. Tracking all open file descriptors
3. Monitoring for resources that remain open too long
4. Providing stack traces to help identify the source of leaks

