profiler-runit 0.1.4

Execute and profile cli commands

runit

A Python CLI tool to execute, monitor, and profile and visualize commands.

Project Purpose

This project was created to prove that advanced functionality is absolutely possible in the terminal. runit showcases real-time monitoring of child processes, delivers detailed resource reports, and is built with a strong emphasis on robust, well-tested code. The goal: bring a modern, delightful CLI experience to profiling command execution with charts and everything.

License

GPL v3. Please see the LICENSE file in the root of the repo for details.

Features

• Utilizes psutil for advanced resource tracking
• Reports execution time, system resource usage, threads, and child processes (including monitoring of all children)
• Terminal charts displaying CPU and memory usage
• Well tested

Demo

See runit in action:

Requirements

• Python 3.9+
• psutil
• plotext

Installation

Clone this repository and install with pip:

git clone https://github.com/notesofcliff/runit.git
cd runit
python -m venv .venv
source .venv/bin/activate  # Or (for Windows) .venv/Scripts/activate
pip install -e .
# Optionally run the tests
python test.py

Or, install directly from GitHub:

pip install git+https://github.com/notesofcliff/runit.git

After installation, use runit to run and analyze any command:

runit [args...] <command>

For example:

runit echo hello world

• If you run runit with no arguments, it will print help.

You'll receive a detailed resource usage report.

If plotext is installed and --plot is specified, terminal charts for CPU and memory usage over time will be displayed.

• Charts appear only if the child process runs long enough to gather data.

Example Output

=== runit report ===
Command: ['python', '-c', 'print(123)']
PID: 12345
Start Time: 2025-07-31 12:00:00.000000
End Time: 2025-07-31 12:00:01.000000
Duration: 0:00:01
Max RSS (bytes): 12345678
Max Threads: 4
Max Children: 0
Samples: 10

If plotext is installed, you'll also see beautiful terminal charts for CPU and memory usage.

Run the Benchmarks

To run the benchmarks:

runit python benchmark.py

CLI Options

• --log-level LEVEL: Set the logging level (default: INFO)
• --out-file FILE: Write the report and charts to a file (output is also shown in the terminal)
• --strip-ansi: Remove ANSI escape codes from output file (for plain text files)
• --plot: If specified, attempt to plot CPU and memory usage
• command: The command to run and monitor

When using --out-file, the output is both displayed in the terminal and written to the file. For plain text (no colors or terminal formatting), simply add --strip-ansi.

runit --out-file out.txt --strip-ansi python benchmark.py

You'll see the report and charts in your terminal, and a plain text version in out.txt.

Development & Extending

Project Structure

The codebase is organized for clarity and extensibility:

runit/
├── __init__.py
├── cli.py             # CLI handling: argparse setup and entrypoint
├── monitor.py         # Process launching + stat collection logic
├── report.py          # Summary formatter (text output)
├── plotting.py        # CPU/memory plots via plotext
├── utils.py           # Reusable helpers (ANSI stripping, extractors)
└── logging_config.py  # Centralized logging setup

How to Add Features or Extend

• Add new measurements:
  • Extend monitor.py to collect additional stats.
  • Add new extractors to utils.py.
• Change reporting:
  • Update or add new formatters in report.py.
• Add new CLI options:
  • Edit cli.py and its create_parser() function.
• Add/modify plotting:
  • Update plotting.py (plotting is optional and gracefully degrades if plotext is missing).
• Logging:
  • All modules use a named logger for granular control. Adjust logging config in logging_config.py. runit --out-file out.txt --strip-ansi python benchmark.py

Tips for Contributors

• Each module should do one thing (separation of concerns).
• Avoid global state; keep functions pure where possible.
• Add tests for new features in test.py.
• Use dependency injection for flexibility (e.g., pass formatters or flags).
• Optional features should degrade gracefully (never crash if a dependency is missing).