dynamic-graph

This software provides an efficient way to modelize a C++ data-flow.

A dynamic graph data-flow is composed of:

• entities (graph nodes)
• signals (input/output of a graph action)

Output signals can then be plugged into input signals to data transmission.

An efficient caching mechanism avoid useless data recomputation and a simple built-in language can be used to control the graph actions.

It is released under the BSD-clause 2 license.

Warning: this repository contains Git submodules. Please clone this repository using the git clone --recursive command. If you already have cloned the repository, you can run git submodule init && git submodule update to retrieve the submodules.

Documentation

To get started with this library, please read the online Doxygen documentation.

It can also be generated locally by running the make doc command. After the package is installed, the documentation will be located in the $prefix/share/doc/dynamic-graph directoy where $prefix is your installation prefix (/usr/local by default).

Getting Help

Support is provided through issues on the github interface.

How can I install dynamic-graph?

Installing dependencies

The matrix abstract layer depends on several packages which have to be available on your machine.

• Libraries:
  • Boost (>= 1.40)
  • Doxygen
  • Eigen3
  • pthread
• System tools:
  • CMake (>=2.6)
  • pkg-config
  • usual compilation tools (GCC/G++, make, etc.) If you are using Ubuntu, these tools are gathered in the build-essential package.

Compiling and installing the package

The manual compilation requires two steps:

1. configuration of the build and generation of the build files
2. compilation of the sources and installation of the package

dynamic-graph uses CMake to generate build files. It is recommended to create a separate build directory:

mkdir _build         # (1) Create a build directory
cd _build            # (2) Go to the newly created build directory
cmake [options] ..   # (3) Generate the build files

Options which can be passed to CMake are detailed in the next section.

make                 # (4) Compile the package
make test            # (5) Execute the package tests
make install         # (6) Install the package into the prefix (see step 3)

Options

Additional options can be set on the command line through the following command: -D<option>=<value>.

For instance: cmake -DCMAKE_BUILD_TYPE=RelWithDebInfo .. will set the CMAKE_BUILD_TYPE option to the value RelWithDebInfo.

Available options are:

• CMAKE_BUILD_TYPE set the build profile that should be used (debug, release, etc.). We recommend RelWithDebInfo as it will provide performances while keeping debugging symbols enabled.
• CMAKE_INSTALL_PREFIX set the installation prefix (the directory where the software will be copied to after it has been compiled).

Running the test suite

The test suite can be run from your build directory by running:

   make test

Please open a ticket if some tests are failing on your computer, it should not be the case.

Contributing

If you want to contribute, please refer to the CONTRIBUTING.md file

Credits

This package authors are credited in the AUTHORS file.

Available Packages

• RobotPkg : http://robotpkg.openrobots.org/robotpkg/motion/dynamic-graph/index.html