A Jupyter Widget for Niivue based on anywidget.
pip install ipyniivue
In a Jupyter environment:
from ipyniivue import NiiVue
nv = NiiVue()
nv.load_volumes([{"path": "images/mni152.nii.gz"}])
nv
See the basic demo to learn more.
ipyniivue uses the
recommended hatchling
build-system, which is convenient to use via the hatch
CLI. We recommend installing hatch
globally
(e.g., via pipx
) and running the various commands defined within
pyproject.toml
. hatch
will take care of creating and synchronizing a
virtual environment with all dependencies defined in pyproject.toml
.
All commands are run from the root of the project, from a terminal:
Command | Action |
---|---|
hatch run format |
Format project with ruff format . and apply linting with ruff --fix . |
hatch run lint |
Lint project with ruff check . . |
hatch run test |
Run unit tests with pytest |
Alternatively, you can develop ipyniivue by manually creating a virtual
environment and managing installation and dependencies with pip
.
python3 -m venv .venv && source .venv/bin/activate
pip install -e ".[dev]"
This is an anywidget project, which means
the code base is hybrid Python and JavaScript. The JavaScript part is developed
under js/
and uses esbuild to bundle the code.
Any time you make changes to the JavaScript code, you need to rebuild the files
under src/ipyniivue/static
. This can be done in two ways:
npm run build
which will build the JavaScript code once, or you can start a development server:
npm run dev
which will start a development server that will automatically rebuild the code as you make changes. We recommend the latter approach, as it is more convenient.
Once you have the development server running, you can start the JupyterLab
or VS Code to develop the widget. When finished, you can stop the development
server with Ctrl+C
.
NOTE: In order to have anywidget automatically apply changes as you work, make sure to
export ANYWIDGET_HMR=1
environment variable. This can be set directly in a notebook with%env ANYWIDGET_HMR=1
in a cell.
release.yml
workflow.v*
is pushed
to the repository.git tag -a vX.X.X -m "vX.X.X"
git push --follow-tags
antfu/changelogithub
By following this release process and utilizing conventional commits, you can ensure consistent and informative releases for your project.