.. _sec-install-matlab:

MATLAB installation
===================

  Main authors: `Fabrice Le Bars <https://www.ensta-bretagne.fr/lebars/>`_, `Simon Rohou <https://www.simon-rohou.fr/research/>`_

Since 2019, MATLAB allows you to import Python packages into its environment. Codac offers a Python package dedicated to MATLAB, with the necessary adaptations to run in this environment (for example: indices starting at ``1``, alternatives to unsupported ``&`` and ``|`` operations, *etc.*). This Python package dedicated to MATLAB is called ``codac4matlab``. See `the related PyPI page <https://pypi.org/project/codac4matlab>`_.

1. **Dependencies**: Ensure you have MATLAB R2019b or later.
2. **Python integration**: Follow the instructions `here <https://www.mathworks.com/help/matlab/matlab_external/install-supported-python-implementation.html>`_ to set up Python integration.
3. **Install the Python package**:

   .. code-block:: bash
   
      pip install codac4matlab

Once Codac is installed, you can use it in your MATLAB scripts using the following import command:

.. tabs::
  
  .. code-tab:: matlab

   import py.codac4matlab.*

   % Your code...

.. py.importlib.import_module('codac')


Potential problem with VIBes
----------------------------

One of the graphical outputs supported in Codac is `the VIBes viewer <https://enstabretagnerobotics.github.io/VIBES/>`_.
VIBes uses a client/server architecture and reads its configuration from the ``.vibes.json`` file located in your home directory (typically ``$HOME``). This file is automatically generated.

If you launch MATLAB with ``sudo``, the ``HOME`` variable points to ``/root``, and VIBes will not find the ``.vibes.json`` file, causing the display to fail.

A solution is to avoid using ``sudo``. If required, run MATLAB with the correct ``HOME``:

.. code::

  sudo HOME=/home/your_username matlab