run-ipynb 0.1.0

A tool to run Jupyter notebooks as Python scripts

run-ipynb

Run Jupyter notebooks as plain Python scripts from the command line.

This tool converts a .ipynb to .py, removes IPython-only lines, ensures a non-interactive Matplotlib backend, optionally adjusts sys.path when os.chdir(...) is used, executes the script, and cleans up the generated file by default.

Installation

• Requirements: Python >= 3.7

Install directly from Git using pip:

# HTTPS
pip install "git+https://github.com/dschaub95/run-ipynb"

After installation, a CLI named run-ipynb will be available.

Usage

run-ipynb path/to/notebook.ipynb [--keep] [--yes]

• notebook_path: Path to the .ipynb notebook to run
• -k, --keep: Keep the generated .py file after execution
• -y, --yes: Automatically overwrite an existing .py file without prompting

Examples

# Basic: convert, run, and remove the temporary .py file
run-ipynb notebooks/analysis.ipynb

# Keep the .py file for inspection and overwrite if it already exists
run-ipynb notebooks/analysis.ipynb --keep --yes

What it does

• Converts the notebook with jupyter nbconvert --to script
• Removes get_ipython().run_line_magic lines
• Sets Matplotlib backend to Agg for headless environments
• If os.chdir(...) is detected, appends os.getcwd() to sys.path right after the change
• Executes the resulting script with your current Python interpreter
• Deletes the generated .py file unless --keep is passed

Notes

• jupyter is installed as a dependency and must be available on your PATH for nbconvert to work.