Call C++ code from python using the Boost libraries

In this short tutorial, I will show you how to call C++ code from python using Boost. I present a very simple example; raising a base to a power (exponentiation). However, the methods I present can be generalized to myriad applications using more elaborate C++ interfaces/libraries/etc...

Dependencies (version)

• cmake (3.5.1)
• libboost-dev (or equivalent if not using debian) (1.58.0)
• python (2.7.12)
• gcc (or another C++ compiler) (5.4.0)
• make (or equivalent if not using Linux) (4.1)
• pytest* (3.6.0)

*pytest is a python module used for unit-testing and is strictly optional. The other dependencies are required.

Building the package

From path_to_repo, the top directory of the project, execute the following from the command line:

mkdir build ; cd build ; cmake .. ; make ; make install

make install is optional ( and may require superuser privileges, like sudo ); it installs the package alongside other site-packages (i.e. those from pip). If you choose not to run make install, you can add the build directory to $PYTHONPATH.

The end-products of the build stage are:

• a shared object library containing the wrapped C++ code, callable from python: pyRTP.so
• a test executable for verifying the C++ code is functional: test_cpp_rtp

Unit testing

After installing pyRTP using EITHER:

• make install from path_to_repo/build OR
• export PYTHONPATH=path_to_repo/build:$PYTHONPATH from the command line

you are ready-to-rock!

I use pytest to unit test the build. From the test directory, you can type

pytest test.py

and you should see the following output:

========================================================== test session starts ==========================================================
platform linux2 -- Python 2.7.12, pytest-3.6.0, py-1.5.3, pluggy-0.6.0
rootdir: /home/joe/github/call-cpp-from-python-with-boost/test, inifile:
collected 2 items                                                                                                                       

test.py ..                                                                                                                        [100%]

======================================================= 2 passed in 0.04 seconds ========================================================

which indicates that the two provided unit tests both pass. Feel free to play around with it!