Installing IMAS-ParaView

If the IMAS-ParaView module is available on your HPC environment, you can ignore the further steps below and simply load the module as follows:

module load IMAS-ParaView
# Run paraview (add vglrun to enable hardware acceleration)
vglrun paraview
# Open the "Sources" tab in the top left, and ensure you see "IMAS Tools" in
# the drop down menu.
# The command-line interface can be started using:
imas-paraview --help

Installation from PyPI

IMAS-ParaView is published on PyPI and can be installed into a virtual environment that uses ParaView's own Python interpreter (pvpython).

  1. Download ParaView 5.13.3 from the ParaView website: https://www.paraview.org/download/?version=v5.13&filter=Linux

  2. Unzip the ParaView package and enter the extracted directory:

    tar -xzvf ParaView-5.13.3-MPI-Linux-Python3.10-x86_64.tar.gz
cd ParaView-5.13.3-MPI-Linux-Python3.10-x86_64

  3. Create a virtual environment using ParaView's bundled pvpython interpreter and install IMAS-ParaView from PyPI. This example uses uv (pip install uv), but python -m venv and pip work too.

    uv venv -p ./bin/pvpython
uv pip install imas-paraview

  4. Activate the virtual environment and set the required environment variables.

    . .venv/bin/activate
export PYTHONPATH="$(python -c 'import sysconfig; print(sysconfig.get_path("purelib"))')"
export PV_PLUGIN_PATH="$(python -c 'import imas_paraview; print(imas_paraview.__path__[0])')/plugins"
# This LD_PRELOAD is required when loading data with the imas_core HDF5 backend.
# Note that it may break any built-in ParaView HDF5 support...
export LD_PRELOAD="$(ls "$PYTHONPATH"/imas_core.libs/libhdf5-*.so.*)"

  5. Run ParaView:

    ./bin/paraview
# Open the "Sources" tab in the top left, and ensure you see "IMAS Tools" in
# the drop down menu.

    The command-line interface is available in the activated environment as well:

    imas-paraview --help

Development installation on SDCC

This section provides instructions for installing and using a development version of the IMAS-Paraview plugins on the ITER SDCC cluster.

  • Setup a project folder and clone git repository

mkdir projects
cd projects
git clone git@github.com:iterorganization/IMAS-ParaView.git
cd IMAS-ParaView

  • To run a plugin in Paraview, run the following at the root of the project directory.

# Load compatible IMAS-Python and ParaView modules, like:
module load IMAS-Python/2.3.0-foss-2023b ParaView/5.12.0-foss-2023b
# export environment variables, this assumes the current
# working directory is the root of the repository
export PV_PLUGIN_PATH=$PWD/imas_paraview/plugins:$PV_PLUGIN_PATH
export PYTHONPATH=$PWD:$PYTHONPATH
# Run paraview (add vglrun to enable hardware acceleration)
vglrun paraview
# Open the "Sources" tab in the top left, if you see "IMAS Tools"
# in the drop down, it is installed correctly.

  • To use the command-line interface, setup a python virtual environment and install python dependencies

# Load compatible IMAS-Python and ParaView modules, like:
module load IMAS-Python/2.3.0-foss-2023b ParaView/5.12.0-foss-2023b
# create virtual environment and install dependencies
python3 -m venv ./venv
. venv/bin/activate
pip install --upgrade pip
pip install --upgrade wheel setuptools
# For development install in editable mode
pip install -e .[all]
# Run CLI with help information
imas-paraview --help
# If you see the help page of IMAS-ParaView, it is installed correctly.

  • Every time that a new session is started, ensure the correct modules are loaded, the python virtual environment is activated, and the environment variables are set.

# Load the required modules
module load IMAS-Python/2.3.0-foss-2023b ParaView/5.12.0-foss-2023b
# Export the environment variables
export PV_PLUGIN_PATH=$PWD/imas_paraview/plugins:$PV_PLUGIN_PATH
export PYTHONPATH=$PWD:$PYTHONPATH
# And activate the Python virtual environment
. venv/bin/activate
# Validate if it is working as intended
imas-paraview --version

  • To run the unit and integration tests, make sure the install is working using the code block above. Also ensure the optional test dependencies are pip installed (or simply use all, to install all optional dependencies).

# The integration tests require X virtual framebuffer to be installed
module load Xvfb/21.1.9-GCCcore-13.2.0

# The tests require downloading external IDSs from Zenodo
wget -P data/ -i zenodo_datasets.txt

# Run the tests
python -m pytest

  • Alternatively, if you want to skip running the integration tests and the tests that contain external data:

python -m pytest -m "not integration and not external_data"

  • To build the IMAS-ParaView documentation, ensure the optional docs dependencies are pip installed (or simply use all, to install all optional dependencies).

make -C docs html
# You can now open ./docs/_build/html/index.html
Last update: 2026-07-16