This repository includes all of the student material for the SciPy 2014 workshop "Integrating Python and C++ with Boost.Python". Participants should be sure to clone this repository to their working machine prior to attending the workshop. They should also be sure to verify that their system can compile and link Boost.Python-based programs and shared libraries.

Pre-Workshop Instructions

The primary activity of this workshop is writing C++ code and compiling it into Python modules. As such, in order to get the most out of this workshop, each participant should bring their own laptop, though sharing a system between two people should work reasonably well.

We will spend as little time as possible - ideally none - on installing Python, Boost.Python, compilers, etc. What follows are instructions for setting up your system and verifying that it's ready for the workshop.

The easy way: Ready-to-go virtual machine image with Ubuntu

The simplest way to make sure your system is ready for the workshop is to use the VM we've prepared for it. The VM comes with a full Ubuntu installation along with all of the libraries and build tools that you need to do the exercises.

The image is packaged as an OVF 1.0 archive (.ova) that should run on most of the popular virtualization systems.

You can download the image here.

Once you boot the image, you should be taken to an Ubuntu desktop. From there, start a terminal (you may need to use alt-F1 to see the launcher) and type:

#!bash

% cd workshop/exercises
% make plumbing_test
% plumbing_test
<information on Python version>
% make smoke_test
% python3 -c "import smoke_test; print(smoke_test.test())"
42

If all of this works, then your system is ready for the workshop.

Notes

• There's been a report that this image crashes on VMWare, but it's known to work properly on several VirtualBox installations.

Setting up a linux system

Using your distributions package management system, first install the following packages:

• Boost.Python This may just be part of a "boost" package, or it may be on its own. Be sure to install the "dev" package with headers.
• Python Use the version that matches the Boost.Python you installed. This will probably be Python 2.7, but there's no guarantee of that.
• A C++ compiler. g++ is probably the best option.
• An editor of some sort.
• git To clone the exercises
• make To build the exercises

Once you have all of this installed, you need to clone the git repository containing the exercises.

#!bash

% git clone git@bitbucket.org:sixty-north/scipy2014_boost_python_workshop_student_material.git workshop
% cd workshop

Once this is in place, you may need to modify make.common to match your system. It should be fairly self-explanatory.

Next, you should run a few test to make sure the system is ready:

#!bash
% make plumbing_test
% plumbing_test
<information on Python version>
% make smoke_test
% python -c "import smoke_test; print(smoke_test.test())"
42

Note that the Python executable may be python3 or perhaps even python2 depending on your distribution and configuration.

Notes

• On Ubuntu 12 (Precise) the boost.python package you should install is libboost-python1.46-dev.
• On Ubuntu 14 (Trusty) the boost.python package you should install is libboost-python1.55-dev. This can make libraries for Python 2.7, 3.3, and 3.4.
• On Ubuntu 14 you may need to update the PYTHON_EXTENSION_SUFFIX in make.common to .so rather than relying on python3-config. This is because python3-config --extension-suffix reports @SO@ rather than .so for some reason.

Setting up the system for Linux, MacOSX or Windows using CMake

Windows

• Download Git and CMake

• Start Git Bash

git clone git://github.com/jcfr/scipy2014_boost_python_workshop_student_material.git
mkdir scipy2014_boost_python_workshop_student_material-build && cd $_
cmake -G "Visual Studio 10 2010" ../scipy2014_boost_python_workshop_student_material
cmake --build ../ --config Release

Linux

sudo apt-get install cmake
sudo apt-get build-essentials
sudo apt-get install libpython-dev
sudo apt-get libboost1.54-dev
git clone git://github.com/jcfr/scipy2014_boost_python_workshop_student_material.git
mkdir scipy2014_boost_python_workshop_student_material-build && cd $_
cmake ../scipy2014_boost_python_workshop_student_material
make

MacOSX

• Download Git and CMake

• Install Boost and Python

git clone git://github.com/jcfr/scipy2014_boost_python_workshop_student_material.git
mkdir scipy2014_boost_python_workshop_student_material-build && cd $_
cmake ../scipy2014_boost_python_workshop_student_material
make

Running tests

In all three cases, convenience tests have been added and can easily be run:

$ cd scipy2014_boost_python_workshop_student_material-build
$ ctest -N
Test project [...]
  Test #1: plumbing_test
  Test #2: smoke_test

Total Tests: 2

$ ctest
Test project [...]
Start 1: plumbing_test
1/2 Test #1: plumbing_test ....................   Passed    0.02 sec
    Start 2: smoke_test
2/2 Test #2: smoke_test .......................   Passed    0.02 sec

100% tests passed, 0 tests failed out of 2

ProTips: Adding the -V option to ctest will disp

Setting up a Mac OS X system

There are a number of ways to set up a Mac OS X system, including compiling everything yourself, homebrew, and macports. The homebrew approach is known to work, but there's no reason that the others shouldn't. Ultimately, you need to make sure you install the following things:

• Boost Including Boost.Python and all development headers
• Python
• A c++ compiler
• make
• git
• an editor

Once these are installed, clone the workshop git repository:

#!bash
git clone git@bitbucket.org:sixty-north/scipy2014_boost_python_workshop_student_material.git workshop
cd workshop

You may now need to edit make.common to match your system. There is a make.common.OSX that should be closer to a standard OSX system than the default make.common, so you can copy it over or use it as a template if you want.

Next, build and run a few tests:

#!bash
% make plumbing_test
% plumbing_test
<information on Python version>
% make smoke_test
% python -c "import smoke_test; print(smoke_test.test())"
42

Note that the Python executable may be python3 or perhaps even python2 depending on your configuration.

Windows

For windows we have provided a VisualStudio 2013 solution as part of the course material. To use this, you'll need to install:

• git
• Python Version 2.7 or 3.x
• Boost Including Boost.Python and development headers
• VisualStudio 2013

Once this is all in place, clone the git repository:

> git clone git@bitbucket.org:sixty-north/scipy2014_boost_python_workshop_student_material.git workshop

(Note that you can also use GUI git tools as well.)

The VisualStudio project is in workshop/vs. Open the solution in VisualStudio and build the plumbing_test and smoke_test projects. You should build "release" versions.

Then in the directory workshop/vs/BoostPythonWorkshop/Release, check the two items you just built:

> plumbing_test.exe
<information about your Python version>
> python -c "import smoke_test; print(smoke_test.test())"
42

If this all works, then you're ready for the workshop.