It is strongly recommended to install NumPy and Matplotlib before installing FluidDyn.

FluidDyn also depends on psutil, but it will be automatically installed during the installation of FluidDyn so you do not actually care.

FluidDyn also used some other packages for some particular tasks, as in particular Scipy and h5py. Since they can be difficult to get for some small hardware, they are not considered as real dependencies, but be prepared to get ImportError if you try to do something using these packages without them.

Basic installation

FluidDyn can be installed from the Python Package Index by the command:

pip install fluiddyn

You can also download the source-code and run:

python install

Install in development mode

Fluiddyn has been designed to be used by scientists-developers. Thus, you may have to modify the code of the FluidDyn packages. So I would advice to work “as a developer”, i.e. to get the source code and to use revision control and the development mode of the Python installer.

For FluidDyn, I use the revision control software Mercurial and the main repository is hosted here in Bitbucket. I would advice to fork this repository (click on “Fork”) and to clone your newly created repository to get the code on your computer (click on “Clone” and run the command that will be given). If you are new with Mercurial and Bitbucket, you can also read this short tutorial:

If you really don’t want to use Mercurial, you can also just manually download the package from the Bitbucket page or from the PyPI page.

The development mode is often very convenient. From the root directory of the project, run:

python develop

After the installation, it is a good practice to run the unit tests by running python -m unittest discover from the root directory or from any of the “test” directories.