Setting the Python environment

1. Python requirements

Table 1. List of modules to compile Feel++ Python bindings

Module

Version

mpi4py

3.0

sympy

1.1

2. Using virtualenv for development

To develop Feel++ applications in Python, it is strongly suggested that you create a protected Python environment using e.g. virtualenv.

Virtualenv is a tool used to create an isolated Python environment. The environment has its own installation directories, the libraries are not shared with other virtualenv environments. It doesn’t access the globally installed libraries either.

virtualenv allows to create a custom user-based environment where experiments can be easily done during the development process.

The steps are as follows:

  • install virtualenv

    pip3 install virtualenv

    and check the full path of the environment

which virtualenv
  • create a virtual environment using python3

    > which python3
    /usr/bin/python3
    > mkdir pyfeelpp-env
    > virtualenv ~/pyfeelpp-env -p /usr/bin/python3

    The last command creates a local copy of your environment. While working on this pyFeel++, you should activate the local environment in order to make sure you’re working with the right versions of your tools and packages.

  • activate the environment

    > cd pyfeelpp-env
    > . bin/activate
  • configure Feel++ using cmake or configure with the environment activated so that the make install procedure installs pyFeel++, pyFeel++ Toolboxes and pyFeel++ MOR in the virtual environment where you can test it safely.