Installation

The PyWAsP system is distributed through the conda package, dependency, and environment manager via DTU Wind Energy’s WAsP Conda Channel.

Note

Before you can download the software, you will need an access token for the software, and access to the DTU Wind Energy license server.

Currently, PyWAsP can be installed on Linux and Windows systems. However, the windows installation is considered experimental at this stage, so we recommend to install PyWAsP on windows via WSL. In the future we expect full support for installation on Windows and MacOS as well.

conda/mamba

  1. To install PyWAsP on Linux, you first need a package manager. We recommend Miniforge:

Miniforge provides both conda and the faster mamba commands:

Linux

wget https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-Linux-x86_64.sh
bash Miniforge3-Linux-x86_64.sh

Follow the instructions and restart your terminal to activate the base environment.

Windows

Go to https://github.com/conda-forge/miniforge and download and run the latest .exe installer for windows.

Complete the installation using the “default” option and follow the instructions. Then open the “miniforge” prompt which is installed with the miniforge.

If you prefer to use powershell, run the command:

conda init powershell

Now open powershell, which should now have an activate conda environment.

  1. Install PyWAsP using mamba (recommended for speed) or conda by creating a new environment. Replace <env_name> with your desired environment name:

mamba create -c https://<user>:<token>@conda.windenergy.dtu.dk/channel/wasp -n <env_name>  pywasp

Here, <user> and <token> are your credentials from the license email.

Note

The Python source code and binaries are installed into a folder in the distribution you installed. For example, if you installed Miniforge, the source code is located in <path>/miniforge/envs/<env_name>. Where <path> is the path to the folder where you installed Miniforge, and <env_name> is the name of the environment you created.

  1. Optionally, you can add the WAsP Conda channel to your conda configuration to simplify future installations.:

Replace <user> and <token> with the credentials from your license email:

mamba config --add channels https://<user>:<token>@conda.windenergy.dtu.dk/channel/wasp

Verify your channels are configured correctly:

mamba config --show channels

You should see:

channels:
 - conda-forge
 - https://<user>:<token>@conda.windenergy.dtu.dk/channel/wasp

Now you can install PyWAsP in any new environment without specifying the channel again:

mamba create -n <env_name> pywasp

If other channels exist and you run into issues, you can try to remove those additional channels to prevent package conflicts:

mamba config --remove channels <channel_name>

  1. Activate your environment:

mamba activate <env_name>

  1. Verify your installation:

python -c "import pywasp; print(pywasp.__version__)"

  1. Optionally, configure PyWAsP interactively by running:

python -c "import pywasp as pw; pw.user_config.create_config_interactively()"

You’ll be prompted for licensing details. For more information, see the User Config section.

If you’re new to PyWAsP, explore our Tutorials and Examples.

pixi

Pixi is a modern package manager that provides a fast and reliable way to manage dependencies and environments.

  1. First, install pixi by following the instructions at https://pixi.sh/.

  2. Create a new pixi project or add PyWAsP to an existing project:

    mkdir my-pywasp-project  # (if you don't have a project yet)
cd my-pywasp-project
pixi init

  3. Add the WAsP conda channel and install PyWAsP:

    pixi project channel add https://<user>:<token>@conda.windenergy.dtu.dk/channel/wasp
pixi add pywasp

    Here, <user> and <token> are your credentials from the license email.

  4. Run PyWAsP in the pixi environment:

    pixi run python -c "import pywasp; print(pywasp.__version__)"

  5. To work interactively with PyWAsP, start a shell in the pixi environment:

    pixi shell

  6. Configure PyWAsP interactively (optional):

    pixi run python -c "import pywasp as pw; pw.user_config.create_config_interactively()"

    For more information, see the User Config section.

MacOS

Warning

Installation on MacOS is currently not supported. But, look for it in future releases!

Installing a newer version of PyWAsP

The PyWAsP package is updated regularly. To update your installation to the latest version we recommend installing a new environment with the latest version of PyWAsP by following the installation instructions above and giving the new environment a new name.

If you use pixi, you can update your existing project by running:

pixi update

This will update all packages in your project to the latest versions available in the channels you have configured.

Optional configuration for custom or self-signed Certificate Authority (CA) certificate.

In order to validate the licensing, PyWAsP must connect to a licensing server. This connection uses TLS for secure communication. By default, it uses the system’s root certificates to validate peers. If you’re operating in a private network with a custom certificate infrastructure, need to setup a custom or self-signed Certificate Authority (CA) certificate.

If you run PyWAsP in this use case, you may get an error that looks like

ssl.SSLCertVerificationError: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: self-signed certificate in certificate chain

Or

pywasp._errors.LicenseError: DTU License Error 123: The license server is not trusted. Additional root certificates may be needed.

To setup the self-signed CA certificate, you need to add the path to the .pem file in your local machine, to the pywasp config file. Your IT administrator should provide you this file. You can find the path to the pywasp config file by opening a python CLI and running the command

pywasp.user_config.get_config_filepath()

Then, open the file with a text editor and add the following line to the [licensing] section.

root_ca_filepath = /path/to/file.pem

See User Config for more information on the pywasp config file.

PyWAsP versions and backwards compatibility

PyWAsP versions are numbered according to the EffVer approach, which is versioning scheme which signals the effort required to upgrade to a new version. Generally, we try to avoid breaking changes, but we dont promise it, and recommend pinning the version of PyWAsP you use in your projects to avoid unexpected changes in behavior.