Clone the repository and install with full extras using:
pip install -e .[dev,extras]
Use gb.py to plot the full data set in results/gb.png.
Zoom in on e.g. the NY square using gb.py NY, or zoom even further with e.g. gb.py NY3
Use interpolation.py to create a whole bunch of plots showing cross-sections of the landscape, all stored in results.
Every method and every class should have a docstring that describes in plain terms what it does, and what the expected input and output is. Each docstring should start with a single sentence explaining what it does. Optionally, this can be followed by a blank line and a more elaborate explanation.
Docstrings can make use of reStructuredText.
For example, you can link to other classes and methods by writing :class:`nevis.Coords` and :meth:`gb()` .
Sphinx Sphinx is used to convert the docstrings into a HTML format, hosted on http://nevis.readthedocs.io.
The order in which the classes and methods appear here is specified by the .rst file in the docs directory.
At the moment, no unit tests are implemented. (Continuous integration testing would require the data to be kept on github, which I haven't explored.)
Use flake8, or flake8 -j4 to use up to 4 cores.
Flake8 should be installed automatically if you added the dev switch when installing (see above).
Some style guidelines are ignored, as specified in the file .flake8.
Documentation can be built locally with
cd docs
make clean
make html
To test if all classes and methods are covered, use
python nevis/tests/docs.py
Workflows for github actions are stored in .github/workflows.
If you're me, you can upload to PyPI using:
python setup.py sdist bdist_wheel
twine upload dist/*
rm build/ dist/ -rf
Remember you can only do this once per version number.