Setup development environment with conda

This creates an environment with Numpy, Scipy, Matplotlib, and Slycot, and in which changes you make to python-control are immediately available. "Immediately" is not quite true: in a continuously running Python environment, you will have to reload python-control to make any changes have an effect; you could use the IPython autoreload extension for that.

This method uses conda to get the latest releases of Numpy, Slycot, etc., in an isolated environment. You can create as many environments as you need, e.g., one for each different version of Python.

First, install conda-build in your base environment; it is need to run the final conda develop . command.

conda install conda-build

Next, create the environment. control-dev is the environment name, which you can change. The instructions below cause the environment to get packages from conda-forge, where Slycot is available.

conda create -n control-dev
conda activate control-dev
conda config --env --add channels conda-forge
conda config --env --set channel_priority strict 

Now install the packages you need. Depending on what you're doing and how you go about developing and testing, you could add other packages, e.g., IPython or Jupyter Lab.

conda install slycot scipy matplotlib pytest pytest-timeout

Finally, add your python-control source to the control-dev environment. This command must be run inside your python-control working tree:

conda develop .

You can check that everything is OK by running pytest in the root of the python-control working tree.