/nu_jupyter

An experimental juptyer notebook kernel for Nushell

Primary LanguageJupyter NotebookMIT LicenseMIT

Jupyter Notebook support for Nushell

Example of Jupyter running Nushell

This is a very(!) experimental kernel for Jupyter for running Nushell. It currently requires 0.11.1 or later.

To install and run the kernel:

> cd nu_jupyter
nu_jupyter> jupyter kernelspec install ../nu_jupyter --user
nu_jupyter> jupyter lab

Limitations:

  • State is not kept between runs. This is something that will change in future versions.

How to create the python egg

This step is only needed if you've made changes to the nushell kernelspec.

Creating the python egg

To create the egg file do the following:

> cd nu_jupyter
nu_jupyter> python.exe setup.py bdist_egg

This creates an egg file in the dist folder that pipenv can utilize.

Installing using pipenv

From within your pipenv environment folder, mine is called me/jup.

jup> pipenv install -e git+https://github.com/nushell/nu_jupyter.git#egg=dist

This should add the nu_jupyter kernelspec and insert entries in your pipfile and pipfile.lock file.

Running with pipenv

From within your pipenv env folder, mine is called /me/jup.

# To activate your pipenv environment, run
jup> pipenv shell

# To add the new jupyter kernel
jup> jupyter kernelspec install ../nu_jupyter --user

# To launch jupyter lab, run
jup> jupyter lab

After launching you should have jupyter lab looking similar to this. Jupyter Lab The last thing is to ensure that NuShell is somewhere in your path. Jupyter requires access to it in order to run the commands in the notebook.