ROS2 Development Container

This repository contains a development container workspace setup for ROS2 development. It provides a convenient environment for running and debugging ROS2 nodes written in both C++ and Python.

Current workspace works out of the box in Visual Studio Code running on Windows, macOS and Linux.

Features

• Pre-configured development environment for ROS2
• Support for both C++ and Python development
• Integrated debugging capabilities
• Containerized development environment for consistent builds

Getting Started

1. Make sure you have Docker and VS Code with the Remote - Containers extension installed
2. Clone this repository
3. Open the repository in VS Code
4. When prompted, click "Reopen in Container"

The development container will be built, and you'll have a fully configured ROS2 development environment ready to use.

Build System

The build process is managed by the ROS2 build system (colcon) for both C++ and Python nodes. Colcon uses CMake under the hood, with all compilation and installation steps described in corresponding CMakeLists.txt files throughout the package.

Building and Installing

The package can be built and installed by running the scripts/build.sh script or by running the Build Task in Visual Studio Code.
The build process will:

• Build C++ nodes and their dependencies
• Install C++ and Python nodes along with their dependencies to the install directory

Optimizations

The build process includes the following optimizations to minimize build times during development:

• Gold linker for faster linking
• ccache for caching and reusing compilation results
• Symlink installation for Python files to avoid copying

Development

Directory Structure

For convenience, C++ and Python nodes are split into distinct subdirectories, with the CMakeLists.txt in the package directory including both of them.

.
├── src/                   # Source code directory
│   └── my_cpp_py_pkg/     # ROS2 package containing both C++ and Python nodes
│       ├── cpp/           # C++ nodes and dependencies
│       ├── python/        # Python nodes and modules
│       ├── launch/        # Launch files for running nodes
│       ├── CMakeLists.txt # CMake configuration for the package
│       └── package.xml    # Package manifest file

Code Completion

Code completion and IntelliSense are already configured to include:

• C++ ROS2 and system include paths
• Python ROS2 packages
• Package's C++ and Python sources

C++

The current setup allows adding dependencies using CMake's find_module() and colcon's ament_target_dependencies() as well as add_subdirectory() and target_link_libraries() commands. See cpp_node for details. Dependencies can be built within the same package. See cpp_lib as an example.

Python

Python nodes and their packages reside in the python folder. Each node has a main executable file and related packages. For example, py_node1 has py_node1.py as its main executable and node1_package as its package. Both are installed by the corresponding CMakeLists.txt.

Packages that are shared between multiple nodes reside in the share folder and are managed by CMakeLists.txt.

Make sure that your Python package names are unique, as colcon's ament_python_install_package() command installs all packages into the same site-packages directory.

Running and Debugging

The workspace supports convenient node execution via launch files and debugging capabilities for both C++ and Python nodes.

Select and launch the ROS: Launch debug configuration to launch nodes defined in src/my_cpp_py_pkg/launch/default.xml. You can set breakpoints in your code and simultaneously debug ALL C++ and Python nodes using VS Code's debugging features.