This repository contains a Jupyter widget for the GLVis finite element visualization tool based on the glvis-js JavaScript/WebAssembly library.

from glvis import glvis

glvis(data[, width=640, height=480])

# or assign if you want to update later
g = glvis(data)
# run a cell with `g` to show it
g

The data object and be one of:

• a str, in the format of *.saved files
• a Mesh, defined in PyMFEM
• a (Mesh, GridFunction) tuple, defined in PyMFEM

PyMFEM can be installed with pip install mfem --no-binary mfem.

Once you have a glvis object there are a few methods that can used to update the visualization:

# show a new Mesh/GridFunction, resets keys
g.display(data)
# show an updated visualization with the same `Mesh` and `GridFunction`
# dimensions, preserves keys
g.update(data)
# change the image size
g.set_size(width, height)
# force the widget to render. if the widget isn't the last statement in a cell it
# will not be shown without this. see ex9.ipynb
g.show()

See the examples directory for additional examples. To test those locally, start a Jupyter notebook server with

jupyter notebook

The GLVis Jupyter widget can be simply installed with pip:

pip install glvis

It order for the installation to be useful you must enable the extension for one or both of the Classic Notebook and Jupyter Lab, see the next two sections:

To use the widget with the basic Notebook enable it with jupyter nbextension enable:

jupyter nbextension enable --py glvis

After enabling the extension you can verify you're good to go with:

jupyter nbextension list

The output should be something like:

Known nbextensions:
  config dir: path/to/nbconfig
    notebook section
      glvis-jupyter/extension  enabled
      - Validating: OK
  <possibly a different config dir>
      jupyter-js-widgets/extension  enabled
      - Validating: OK

If glvis-jupyter and jupyter-js-widgets are not both listed, try the following:

jupyter nbextension install --user --py glvis
jupyter nbextension enable --user --py glvis
jupyter nbextension install --user --py widgetsnbextension
jupyter nbextension enable --user --py widgetsnbextension

You may also need to run these if you have upgraded to a newer version of the GLVis Jupyter widget.

JupyterLab requires another set of install commands:

jupyter labextension install @jupyter-widgets/jupyterlab-manager --no-build
jupyter labextension install glvis-jupyter

Development installation requires npm, after installing:

git clone https://github.com/glvis/pyglvis.git
cd pyglvis
pip install -e .

jupyter nbextension install --py --symlink --sys-prefix glvis
jupyter nbextension enable --py --sys-prefix glvis

jupyter labextension install @jupyter-widgets/jupyterlab-manager --no-build
# I believe you need node in the path Lab uses for this to work, I see an extension load error
# in a context where I don't have it:
# Failed to load resource: the server responded with a status of 500 (Internal Server Error)
#   lab/api/extensions?1610138347763
# Which is just a python stacktrace, ending with:
#   raise ValueError(msg)
#   ValueError: Please install Node.js and npm before continuing installation.
jupyter labextension link ./js

If you run into errors related to node/npm that aren't helpful try:

cd pyglvis
make clean
cd js
# fix errors in these steps, run `make -C .. clean` each time
npm install
npx webpack

• Update the required version of glvis in js/package.json

• Update the version in js/package.json

# clean out the `dist` and `node_modules` directories
git clean -fdx
npm install
npm publish

• Update glvis/_version.py

  • Set release version
  • Update extension_version to match js/package.json

• git add and git commit changes

  • glvis/_version.py, js/package.json, and js/package-lock.js

You will need twine to publish to PyPI, install with pip.

python setup.py sdist bdist_wheel
twine upload dist/*
git tag -a X.X.X -m 'comment'
git push --tags