Install Kernel Specifications into userspace

This code installs jupyter kernels for different languages into the user space. Once the kernel has been installed it can be used on the jupyter-hub server of DKRZ.

Installation

python3 -m pip install jupyter-kernel-install

Usage

Using the command line interface (cli):

jupyter-kernel-install --help

Usage: jupyter-kernel-install [-h] [--name NAME] [--display-name DISPLAY_NAME] [--version] language

Install jupyter kernel specs of different languages.

Positional Arguments:
  language              The programming language

Options:
  -h, --help            show this help message and exit
  --name, -n NAME       The name of the kernel (default: None)
  --display-name, -d DISPLAY_NAME
                        The display name of the kernel (default: None)
  --version, -V         Display version and exit

Alternatively you can use:

python -m jupyter_kernel_install --help

The following kernel specifications are supported:

• python3
• gnuR
• bash

Example for installing a gnuR kernel:

jupyter-kernel-install r --name r-regiklim --display-name "R for Regiklim"

Using the python library

Example for programmatically installing a bash kernel:

import jupyter_kernel_install as ki
kernel_path = ki.bash(name="bash-regiklim", display_name="bash kernel")

Contributing

Any contributions are welcome. To start developing we recommend creating a new mini conda environment.

conda env create -f environment.yml; conda activate install-kernelspec

Unit tests, building the documentation, type annotations and code style tests are done with tox. To run all tests, linting in parallel simply execute the following command:

tox -p 3

You can also run the each part alone, for example to only check the code style:

tox -e lint

available options are lint, types and test.

Tox runs in a separate python environment to run the tests in the current environment use:

pytest