This repository contains C++ code for the analysis and mining of networks produced by the Uppsala University Information Laboratory (InfoLab, https://uuinfolab.github.io). uunet provides most of the code behind the R and Python multinet libraries.

If you are interested in using the C++ library, please consider that this is research software: major changes can happen for example when the underlying theory is developed. While we try not to modify the interface of the library if not deemed necessary, our main effort to provide a stable interface goes into the R and Python versions.

To install, use and modify the library you need:

• A recent version of git.
• The cmake build system version 3.15 or higher.
• A C++14 compiler.
• Doxygen (optional, only to generate the API reference documentation).

To download the library:

git clone --recurse-submodules https://github.com/uuinfolab/uunet.git
cd uunet

Then create a build directory and cd into it (all the following instructions will be executed inside this directory):

mkdir build
cd build

To install the library:

cmake ..
make
make install

To compile the tests, run cmake with the option -DCMAKE_BUILD_TYPE=Debug. Make will then also generate an executable file running all the unit tests:

cmake .. -DCMAKE_BUILD_TYPE=Debug
make
./run_tests

An overview of the library is available under docs/ as a pdf file. Its latex source is also available.

All classes, functions, etc. are documented inside the .hpp files. This documentation can be compiled into an API reference listing all types and functions using Doxygen. To produce the API reference documentation, use the option -DDOXY=On:

cmake .. -DDOXY=On
make doc

The documentation is generated inside the build/doc directory.

Short examples are also provided to illustrate the functionality of the library. Individual examples correspond to the .cxx files inside the examples/ directory, and can be individually compiled and executed using the name of the example (from inside the build/ directory):

make creation
./creation

One can also compile all the examples in one line:

make examples
# now all the individual examples can be executed
./creation
./networks
# ...

Unit Tests also provide small code snippets used to test most of the features of the library, which can also be used as documentation.

The master branch contains the most stable code, mainly updated when new versions of the R and Python libraries are released and when new research articles using the library are submitted. The master branch is updated by the administrators from the development branch.

To contribute to the library, please branch from development and open a pull request when your update is ready to be merged.

For any inquiries regarding this repository you can contact matteo.magnani@it.uu.se.