Alternatives

A quick look at other ways to run external commands in Python, and when they might fit better than comrun.

Python standard library

• subprocess.run / subprocess.Popen: battle-tested and built in, but requires boilerplate. You manage streaming, decoding, and error handling yourself. This friction is what inspired the creation of comrun.
• asyncio.create_subprocess_exec: async-friendly subprocesses. You handle reading/writing streams and timeouts.

Use these when you want zero dependencies and are fine wiring up I/O, timeouts, and logging yourself.

Higher-level libraries

• Invoke / Fabric: task runners with decorators, remote execution, and SSH. Good for deployment scripts; heavier footprint than comrun.
• plumbum: shell-like DSL with pipelines and path abstractions; nice for Unix-like scripting, more opinionated than comrun.
• sh-style wrappers (e.g., sh, delegator.py): call executables as functions (git.status()), often shell=True; convenient, but can hide quoting concerns.

Choose these when you want richer task abstractions or DSL-like command composition.

When to choose comrun?

You want a small, synchronous runner with:

• Live output streaming (via Rich) and captured stdout/stderr/combined output.
• Structured results and simple success/failure helpers.
• Hooks for logging/metrics without adopting a heavier task runner.