Discrete Geographic Region Projection Engine
This is a C++/Python library based on GEOS. It relies on data from the 2016 census to project GPS coordinates across Canada and returns the corresponding dissemination areas. These areas typically contain 400-700 people, but may stretch over large regions in sparsely populated areas.
The C++ component of the library is used for on-device geo discretization. The Python package is used for geospatial data preprocessing and exposes utility functions to the simulator.
The C++ build system should be fully self-contained. In other words, CMake will automatically download and build all dependencies of the library. A typical build procedure is:
<PROJECT_ROOT> $ mkdir build
<PROJECT_ROOT> $ cd build
.../build $ cmake ../
.../build $ cmake --build .
To use mutiple workers in the build process, you can use:
.../build $ cmake --build . -j X
where X is roughly the number of workers.
Once built, you should have bin and lib subdirectories in the output directory. These
should contain CLI apps and API libraries, respectively.
To download and preprocess the census data, instructions are provided in the python
directory.