This is the Kinematics/Geometric primitives Python package.
This project provides some types and functions to represent entities like rigid bodies, frames, points, rigid motions.
Most of the types are purely symbolic placeholders, and are meant to be used for building models, with Python.
The ct
package contains facilities to model coordinate-transforms, and also
to get concrete matrix representations.
More information about the packages and the modules of this project is available in the docstrings.
You can generate e.g. html documentation using pdoc. For example:
cd src/
pdoc --no-show-source -o /tmp/docs kgprim/ motiondsl/
git clone <repo> kgprim # replace <repo> with the right URL
cd kgprim/
pip install . # should also fetch and install the dependencies
Before doing the above, you might want to set up a Python3 virtual environment:
mkdir myvenv && python3 -m venv myvenv
source myvenv/bin/activate # may need to pick another script depending on your shell
Python > 3.3
Python packages:
-
SymPy, for the representation of variables, parameters and constants in the
values
module. Also for the symbolic matrix representation of coordinate transforms. -
NumPy, for the numerical matrix representation of coordinate transforms.
-
textX, for the grammar of the MotionDSL language (required by the
motiondsl
package only).
Testing code is based on the unittest
framework from the standard library.
After installation (see above), one may run all the tests with something like:
python -m unittest discover --start-directory test/ --pattern '*.py'
The test modules can also be executed individually, e.g.:
python test/ct/testcore.py
# or
python -m test.ct.testcore
# or
python -m unittest test.ct.testcore
The module test/ct/testcore.py
is a test suite for the kgprim.ct
package
which indirectly also covers the kgprim.motions
module.
Similarly, the test/values.py
performs some simple tests on the types defined
in the kgprim.values
module.
The package test/motiondsl
has tests for the motiondsl
package, though it
may also involve kgprim.ct
.
Copyright 2020-2024, Marco Frigerio
Distributed under the BSD 3-clause license. See the LICENSE
file for more
details.