This repository is an example of a simple Python program that uses a C++ function for performance
In this directory, run pip install ., optionally with -e.
This will not only package and install the Python package: it will also build with your C++ compiler the
cpp files present in pythoncpptemplate/cpp.
This works on Linux provided you have a C++ compiler like GCC installed, it should work on Windows as well but I haven't tried.
pythoncpptemplate.my_library.my_function imports and calls a special function pythoncpptemplate.cpp.computeComplicatedNumber.
This is a costly function, for which performance could matter. It is implemented in C++.
It is defined from C++ in pythoncpptemplate/cpp which also contains some CPython code to make Python aware of it.
Look at pythoncpptemplate/cpp/src/computeComplicatedNumber.cpp, which contains the C++ code that implements the complicated function.
Look at pythoncpptemplate/cpp/src/pythonBindings.cpp which contains the administrative CPython code to make the connection between Python and C++ work.
This API is well documented here: https://docs.python.org/3/c-api/.