databooks was built to be used as a CLI application, you could also use the
databooks data models to manipulate notebooks in other scripts.
All the data models are based on Pydantic models. That means that we have a convenient way to read/write notebooks (JSON files) and ensure that the JSON is actually a valid notebook.
By using databooks to parse your notebooks, you can also count on IDE support.
Using databooks in python scripts is as simple as
from databooks import JupyterNotebook notebook = JupyterNotebook.parse_file(path="path/to/your/notebook") assert list(dict(notebook).keys()) == ['nbformat', 'nbformat_minor', 'metadata', 'cells']
For more information on the internal workings and what is implemented, see how it works.