hupper

What it is and what it does

Hupper is a process monitor designed for development workflows that automatically detects changes to imported Python files and restarts the process. It tracks all modules in sys.modules plus any custom paths you specify, making it useful for iterative development where you want code changes to take effect immediately without manual restarts.

You can use it either from the command line (hupper -m myapp) or programmatically by calling start_reloader() with an importable entry point. The reloader runs as a parent process that monitors file changes and spawns child processes to run your code, allowing you to add extra files to watch via the watch_files() method. It has no runtime dependencies and supports modern Python versions from 3.7 onward.

Use it for:

• Developing web applications where you want the server to reload automatically when you edit source files.
• Running long-lived daemons during development and having them pick up code changes without manual restarts.
• Monitoring configuration files alongside Python source and triggering restarts when either changes.
• Building development scripts that need to reload when dependencies or imported modules are modified.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Hupper monitors Python source files for changes and automatically restarts the process when files are modified, supporting both command-line and programmatic usage patterns.

Yes. Hupper is a stable, actively maintained tool with no dependencies and broad Python version support. It solves a genuine development friction point—automatic process restart on file changes—and is lightweight enough to add to any development workflow. No known security issues.

Install

hupper on PyPI

pip install hupper

uv add hupper

poetry add hupper

Installing hupper

Before you install

Low friction installation with no runtime dependencies. The package is actively maintained with recent commits and supports Python 3.7 through 3.12, including PyPy.

License in practice

MIT license permits use in commercial and private projects with minimal restrictions; attribution required.

Quickstart

pip install hupper

import hupper

reloader = hupper.start_reloader('myapp.scripts.serve.main')
reloader.watch_files(['foo.ini'])

Entry point must be an importable string path; the reloader only returns in a monitored subprocess.

Verify before relying

• Whether watch_files() supports glob patterns or directory recursion beyond explicit file paths.
• Performance characteristics when monitoring large numbers of files or deep directory structures.

Package facts

Evidence: hupper-1.12.1-py3-none-any.whl

Keywords: server, daemon, autoreload, reloader, hup, file, watch, process

Tags

More Monitoring packages