A Jupyter kernel for Scilab

Prerequisites

Jupyter Notebook, and Scilab.

Installation

To install using pip:

pip install scilab_kernel

Add --user to install in the user-level environment instead of the system environment.

This kernel needs the Scilab executable to be run, it which will be searched in this order:

• Using environment variable SCILAB_EXECUTABLE,
• Under Windows, based on registry,
• Under macOS, based on Spotlight database,
• Using the PATH environment variable.

Use the scilab-adv-cli executable if using a Posix-like OS, and WScilex-cli.exe if using Windows.

Usage

To use the kernel, run one of:

jupyter notebook  # or ``jupyter lab``, if available
# In the notebook interface, select Scilab from the 'New' menu
jupyter qtconsole --kernel scilab
jupyter console --kernel scilab

If jupyter executable is not found in your PATH, try python -m notebook instead.

This kernel is based on MetaKernel, which means it features a standard set of magics (such as %%html). For a full list of magics, run %lsmagic in a cell.

A sample notebook is available online.

Configuration

The kernel can be configured by adding an scilab_kernel_config.py file to the jupyter config path (for example ~/.jupyter/scilab_kernel_config.py. The ScilabKernel class offers plot_settings as a configurable traits. The available plot settings are:

• 'format': 'svg' (default), 'png', 'jpg',
• 'backend': 'inline',
• 'size': '<width>,<height>' ('560,420' by default),
• 'antialiasing': for 'svg' backend only, True by default.

c.ScilabKernel.plot_settings = dict(format='svg', backend='inline', size='560,420', antialiasing=False)

Scilab default behavior is setup using lines(0, 800) and mode(0). You can change these behaviors using scilab code on cells.

Files ending with .sci in the current directory are loaded.

Troubleshooting

Kernel Times Out While Starting

If the kernel does not start, run the following command from a terminal:

python -m scilab_kernel.check

This can help diagnose problems with setting up integration with Scilab. If in doubt, create an issue with the output of that command.

Kernel is Not Listed

If the kernel is not listed as an available kernel, first try the following command:

python -m scilab_kernel install --user

If the kernel is still not listed, verify that the following point to the same version of python:

which python  # use "where" if using cmd.exe
which jupyter

Advanced Installation Notes

We automatically install a Jupyter kernelspec when installing the python package. This location can be found using jupyter kernelspec list. If the default location is not desired, you can remove the directory for the scilab kernel, and install using python -m scilab_kernel install. See python -m scilab_kernel install --help for available options.