NetworKit Development Guide

This guide is designed to give potential contributors a quick understanding of the structure of NetworKit as well as what they need to look out for when adding new functionality to NetworKit.

The following text assumes some basic familiarity with the Git version control software. It is not a Git tutorial, because you will find a good one here. Rather, it explains concepts and workflows for the development of this project.

If you want to contribute, you should consider reading the technical report on NetworKit to get familiar with the architecture.

If you use NetworKit in your research publications, please cite the mentioned technical report or the specific algorithm. A list of publications is available on the website.

Contribution Workflow

The development of the NetworKit project takes place at our official repository on GitHub.

There are two important branches you need to be aware of:

  • master: Release branch
  • Dev: Development (stage) branch used for active development

The master branch contains all the stable code that is ready for use. There is no active development on this branch, it is just being updated when there is a new release. Dev is the actual development branch and thus also contains more recent code that might not be completely stable yet. New features that are under development will be merged into this branch.

In order to successfully contribute to the project and add your feature to the development branch, there are a couple of steps you need to follow:

  1. Fork the NetworKit repository
  2. Create a feature branch
  3. Add your code
  4. Send a pull request
  5. Participate in the pull request discussion

In the following part we will take a closer look at each step of the workflow outlined above. To make it more tangible, let’s assume you want to contribute to NetworKit by adding a new algorithm to the existing suite.

1. Fork the NetworKit repository

Visit our official repository on GitHub and click the ‘Fork’ button in the top right to create a personal fork in your account.

Next up, you want to clone your fork to your local machine:

git clone git@github.com:[YOUR-USERNAME]/[FORKED-NETWORKIT].git

Since the algorithm you’re working on will take some time to implement, you want to make sure to stay up to date with the original NetworKit repository. This can be accomplished by tracking the original upstream branch:

git remote add upstream https://github.com/kit-parco/networkit.git

Now you can update your fork on a regular basis with changes from the original repository:

# Fetch changes from original repository
git fetch upstream

# Merge changes into your local branch
git checkout --track origin/Dev
git merge upstream/Dev

This will for example allow you to keep the development branch Dev up to date. Doing this will prevent an unpleasent surprise once you’re ready to submit your pull request since there will be less or no merge conflicts.

2. Create a Feature Branch

To make sure that the work on your algorithm does not disrupt the development process, be sure to create your own branch where you can add new code. In order to have a coherent naming scheme that allows for easier communication, please name your branch in the following way: feature/descriptive-name-of-feature. If we’re assuming you’re creating an implementation of the shortest path algorithm from Dijkstra, this could be feature/dijkstra-shortest-path for example. Make sure to pick a meaningful and short description that is easy to understand.

New feature branches should be based off the development branch Dev where they will be merged back into at a later stage.

# Make sure your branch is based off the development branch Dev
git checkout Dev

# Create your new branch
git checkout -b feature/[my-awesome-feature-name]

# You're now on your new feature branch

3. Add Your Code

In this step you’re going to make and commit the changes needed for your new feature. Please make sure to write clean code that adheres to the style guide outlined further below. It is also important that each feature has appropriate unit tests that cover all of the expected behaviour of the code. Please see the Test-driven development section below for details.

git checkout feature/[my-awesome-feature-name]
git add [files]
git commit -m "[descriptive message about the changes you made]"
git push

Also, from time to time, you should make sure to keep your feature branch up to date with the changes on the development branch Dev in the main repository. If you followed step 1 from above, this can be easily accomplished:

git checkout feature/[my-awesome-feature-name]
git fetch upstream
git merge upstream/Dev

4. Send a Pull Request

Once you finished the development and testing of your new feature, it is time to create a pull request to get your changes merged into the development branch of the NetworKit repository.

This can be done by visiting the Pull requests page (https://github.com/[YOUR-USERNAME]/[FORKED-NETWORKIT]/pulls) of your NetworKit fork on GitHub and clicking on the green New pull request button at the top right side of the page.

Here the base fork at the top should point to kit-parco/networkit and the base should be Dev. The head fork should point to your fork of networkit and the compare branch to the right should point to the feature branch (feature/[my-awesome-feature-name]) you would like to create the pull request for.

Once you’ve reviewed all changes, click the green Create pull request button and your pull request will be created.

5. Participate in the Pull Request Discussion

Once your pull request has been submitted, other developers of NetworKit, including the core development team, will take a look at your code and explanation. In this process there are oftentimes questions that arise or small adjustments that need to be incorporated into the pull request. For this reason, it is important that you actively participate in the discussions around your pull request. This ensures your new feature will eventually make it to the next release.

In case a developer points out a potential issue that needs to be resolved, please make the appropriate changes to your code and push these changes to your feature branch:

git checkout feature/[my-awesome-feature-name]
# Make appropriate changes to files
git add [files]
git commit -m "[message-about-the-resolved-issue]"
git push

The pull request will automatically show your newest changes and developers will know that you resolved the issue. Once all issues have been resolved and your code is accepted, the pull request will be closed and your feature will be merged into the development branch. In the next release, all users of NetworKit will have access to your awesome feature. Hooray!

Style guide

We want to ensure that code across NetworKit is easy to understand for existing as well as new developers. This is why new code added to the project should adhere to the existing code style. At this point in time, there is no comprehensive documentation about the code style being used in NetworKit but there are a few things to look out for:

  • Compiler warnings are likely to turn into future errors. Try to fix them as soon as they appear. Use the -Wall flag when compiling C++ code.
  • Read some code to get used to the code style and try to adopt it.
  • Document classes, methods and attributes in Doxygen style.
  • Use the count and index integer types for non-negative integer quantities and indices.
  • In most cases, objects are passed by reference. New objects are stack-allocated and returned by value. Avoid pointers and new where possible.
  • Use the override keyword to indicate that a method overrides a virtual method in the superclass.
  • In Python, indent using tabs, not spaces.

In a nutshell, new developers should familiarise themselves with the existing code base and adapt the existing style in the C++ as well as Python code base when contributing to NetworKit. Always ensure your code is easy to understand and properly documented.

Report bugs

Please report any bugs on the issues page of the official NetworKit repository on GitHub. In very urgent cases it might also make sense to write on the mailing list. Please provide a minimal example so that others can reproduce that bug.

Tags

A tag is nothing more than a “symbolic name” for a revision. In NetworKit tags are used to mark release versions in the master branch, with a MAJOR.MINOR version name scheme.

Using NetworKit

If you want to build and use NetworKit and do not plan to contribute changes, simply clone the repository. By default, you will be on the master branch, which represents the current release. Follow the setup instructions in the README file.

Student Projects

Students with long-term projects like Bachelor’s or Master’s theses should familiarize themselves with the guidelines and select a forking/branching model with their advisor.

Branching Cheat Sheet

  • list all available branches with highlight for the current branch: git branch
  • switch to a specific branch: git checkout <branchname>
  • start a new branch: git checkout -b <branchname>
  • merge branch-y into branch-x: git checkout branch-x, then git merge branch-y
  • see heads (most recent commits) of all branches: git show-ref --heads

Conventions

The following general conventions apply to all NetworKit developers.

Versioning

  • Before you commit, make sure your code compiles and run the unit tests. Never push code which breaks the build for others.
  • Commit regularly and often to your local repository.
  • Use meaningful commit messages.
  • Get the newest changes from the repository regularly and merge them into your local repository.
  • Make sure that you merged correctly and did not break other people’s work.
  • Push correct code early if possible. Merging is easier if all developers are up to date.
  • Never push --force to the main repository.

Unit Tests and Testing

Every new feature must be covered by a unit test. Omitting unit tests makes it very likely that your feature will break silently as the project develops, leading to unneccessary work in tracing back the source of the error. Also your pull request for this feature will most likely not be accepted.

Unit tests for the C++ part of NetworKit are based on the googletest library. For more information read the googletest primer. The Python test framework currently relies on nose to collect the tests.

  • Each source folder contains a test folder with googletest classes. Create the unit tests for each feature in the appropriate test/*GTest class by adding a TEST_F function.
  • Prefix standard unit tests with test and experimental feature tests with try. A test* must pass when pushed to the main repository, a try* is allowed to fail.
  • Keep the running time of test functions to the minimum needed for testing functionality. Testing should be fast, long-running unit tests look like infinite loops.
  • If the unit test requires a data set, add the file to the input/ folder. Only small data sets (a few kilobytes maximum) are acceptable in the repository.
  • Any output files produced by unit tests must be written to the output/ folder.

To build and run the tests you need the gtest library. Assuming, gtest is successfully installed and you add the paths to your build.conf, the unit tests should be compiled with:

scons --optimize=Dbg --target=Tests

To verify that the code was built correctly: Run all unit tests with

./NetworKit-Tests-Dbg --tests/-t

Performance tests will be selected with

./NetworKit-Tests-Dbg --benchmarks/-b

while experimental tests are called with

./NetworKit-Tests-Dbg --trials/-e

To run only specific unit tests, you can also add a filter expression, e. g.:

./NetworKit-Tests-Dbg --gtest_filter=*PartitionGTest*/-f*PartitionGTest*

initiates unit tests only for the Partition data structure.

For the Python unit tests, run:

python3 setup.py test [--cpp-tests/-c]

This command will compile the _NetworKit extension and then run all test cases on the Python layer. If you append --cpp-tests/-c, the unit tests of the c++ side will be compiled and run before the Python test cases.

Test-driven development

If you implement a new feature for NetworKit, we encourage you to adapt your development process to test driven development. This means that you start with a one or ideally several test-cases for your feature and then write the feature for the test case(s). If your feature is mostly implemented in C++, you should write your test cases there. If you expose your feature to Python, you should also write a test case for the extension module on the Python layer. The same applies for features in Pyton.

Algorithm interface and class hierarchy

We use the possibilities provided through inheritance to generalize the common behaviour of algorithm implementations:

  • Data and paramters should be passed in the constructor.
  • A void run()-method that takes no parameter triggers the execution.
  • To retrieve the result(s), getter-functions() may be defined.

The Algorithm base class also defines a few other other functions to query whether the algorithm can be run in parallel or to retrieve a string representation.

There may be more levels in the class hierarchy between an algorithm implementation and the base class, e.g. a single-source shortest-path class SSSP that generalizes the behaviour of BFS and Dijkstra implementations or the Centrality base class. When implementing new features or algorithms, make sure to adapt to the existing class hierarchies. The least thing to do is to inherit from the Algorithm base class. Changes to existing interfaces or suggestions for new interfaces should be discussed through the mailing list.

Exposing C++ Code to Python

Assuming the unit tests for the new feature you implemented are correct and successful, you need to make your features available to Python in order to use it. NetworKit uses Cython to bridge C++ and Python. All of this bridge code is contained in the Cython code file src/python/_Networkit.pyx. The content is automatically translated into C++ and then compiled to a Python extension module.

Cython syntax is a superset of Python that knows about static type declarations and other things from the C/C++ world. The best way to getting used to it is working on examples. Take the most common case of exposing a C++ class as a Python class. Care for the following example that exposes the class NetworKit::Dijkstra:

cdef extern from "cpp/graph/Dijkstra.h":
    cdef cppclass _Dijkstra "NetworKit::Dijkstra"(_SSSP):
        _Dijkstra(_Graph G, node source, bool storePaths, bool storeStack, node target) except +

The code above exposes the C++ class definition to Cython - but not yet to Python. First of all, Cython needs to know which C++ declarations to use so the the first line directs Cython to place an #include statement. The second line defines a class that is only accessible in the Cython world. Our convention is that the name of the new class is the name of the referenced C++ class with a prepended underscore to avoid namespace conflicts. What follows is the “real” C++ name of the class. After that, the declarations of the methods you want to make available for Python are needed. The except + statement is necessary for exceptions thrown by the C++ code to be rethrown as Python exceptions rather than causing a crash. Also, take care that the Cython declarations match the declarations from the referenced header file.

cdef extern from "cpp/graph/SSSP.h":
    cdef cppclass _SSSP "NetworKit::SSSP"(_Algorithm):
        _SSSP(_Graph G, node source, bool storePaths, bool storeStack, node target) except +
        vector[edgeweight] getDistances(bool moveOut) except +
        [...]

cdef class SSSP(Algorithm):
    """ Base class for single source shortest path algorithms. """

    cdef Graph _G

    def __init__(self, *args, **namedargs):
        if type(self) == SSSP:
            raise RuntimeError("Error, you may not use SSSP directly, use a sub-class instead")

    def __dealloc__(self):
        self._G = None # just to be sure the graph is deleted

    def getDistances(self, moveOut=True):
        """
        Returns a vector of weighted distances from the source node, i.e. the
        length of the shortest path from the source node to any other node.

        Returns
        -------
        vector
            The weighted distances from the source node to any other node in the graph.
        """
        return (<_SSSP*>(self._this)).getDistances(moveOut)
    [...]

We mirror the class hierarchy of the C++ world also in Cython and Python. This also saves some boiler plate wrapping code as the functions shared by Dijkstra and BFS only need to be wrapped through SSSP.

cdef class Dijkstra(SSSP):
    """ Dijkstra's SSSP algorithm.

    Returns list of weighted distances from node source, i.e. the length of the shortest path from source to
    any other node.

    Dijkstra(G, source, [storePaths], [storeStack], target)

    Creates Dijkstra for `G` and source node `source`.

    Parameters
    ----------
    G : Graph
        The graph.
    source : node
        The source node.
    storePaths : bool
        store paths and number of paths?
    storeStack : bool
        maintain a stack of nodes in order of decreasing distance?
    target : node
        target node. Search ends when target node is reached. t is set to None by default.
    """
    def __cinit__(self, Graph G, source, storePaths=True, storeStack=False, node target=none):
        self._G = G
        self._this = new _Dijkstra(G._this, source, storePaths, storeStack, target)

For the class to be accessible from the Python world, you need to define a Python wrapper class which delegates method calls to the native class. The Python class variable _this holds a pointer to an instance of the native class. Please note that the parameters are now Python objects. Method wrappers take these Python objects as parameters and pass the internal native objects to the actuall C++ method call. The constructor of such a wrapper class is called __cinit__, and it creates an instance of the native object.

The docstring between the triple quotation marks can be accessed through Python’s help(...) function and are the main documentation of NetworKit. Always provide at least a short and precise docstring so the user can get in idea of the functionality of the class. For C++ types available to Python and further examples, see through the _NetworKit.pyx-file. The whole process has certainly some intricacies, e.g. some tricks are needed to avoid memory waste when passing around large objects such as graphs. When in doubt, look at examples of similar classes already exposed. Listen to the Cython compiler - coming from C++, its error messages are in general pleasantly human-readable.

Make algorithms interruptable with CTRL+C/SIGINT

When an algorithms takes too long to produce a result, it can be interrupted with a SIGINT signal triggered by CTRL+C. When triggering from the Python shell while the runtime is in the C++ domain, execution is aborted and even terminates the Python shell. Therefore, we implemented a signal handler infrastructure in C++ that raises a special exception instead of aborting. When implementing an algorithm, it is strongly encouraged to integrate the signal handler into the implementation. There are many examples of how to use it, e.g. networkit/cpp/centrality/Betweenness.cpp or networkit/cpp/community/PartitionFragmentation.cpp

Contact

To discuss important changes to NetworKit, use the e-mail list (networkit@ira.uka.de).

We also appreciate new issues or pull requests on the GitHub repository.

Building the documentation

The documentation can be automatically generated with sphinx. You will need the following software to generate the documentation:

After you installed the above mentioned software, you can build the class documentation by calling ./make_doc.sh in the folder Doc/doc. This will generate the class documentation for C++ and Python in Doc/Documentation.