Awesome
VSS Tools
The overall goal of VSS Tools is to provide a set of tools that can be used to convert or verify Vehicle Signal Specifications defined by the format specified by the COVESA VSS project. VSS Tools is developed in parallel with VSS, please visit the COVESA VSS project for information on how to contribute. If any questions arise, please check out the FAQ for more information.
Installation
There are several ways of installing vss-tools
.
If you would like to contribute then please follow the contribute section instead.
All of them are recommended to be done in an activated python virtual environment:
python -m venv .venv
source .venv/bin/activate
pypi
pip install vss-tools
from source:
# default branch
pip install git+https://github.com/COVESA/vss-tools.git
# explicit branch
pip install git+https://github.com/COVESA/vss-tools.git@master
# commit
pip install git+https://github.com/COVESA/vss-tools.git@1234567
See usage for how to start using it.
Usage
General CLI help should be used for up-to-date info of how to use the tools.
# Help for toplevel options and lists sub commands
vspec --help
# Help for export sub command options and lists sub commands
vspec export --help
# Help for json exporter options
vspec export json --help
Please check here for generic info about exporters and their arguments as well as here for design decision, architecture and limitations.
Compatibility with VSS
The COVESA VSS project repository includes vss-tools as a submodule.
The vss-tools version linked by the VSS repository is the preferred vss-tools version to use for that particular version of the VSS repository.
It is not guaranteed that newer or older versions of vss-tools can successfully handle that particular version of the VSS repository.
The table below gives an overview of basic version support for e.g. vspec export json
.
Other exporters may have stricter requirements.
VSS-tools version | Supported VSS catalog versions | Comments |
---|---|---|
v3.0 | v3.0 - v3.1.1 | |
v3.1 | v3.0 -v4.0 | |
v4.0 | v4.0 | |
v4.1 | v4.0 - | |
4.2 | v4.0 - | |
5.0 | v5.0 - | |
<latest source> | v5.0 - |
Changes affecting compatibility
Examples on changes affecting compatibility
- Somewhat stricter control in VSS-tools 5.0 onwards, minor changes required in VSS 4.X catalogs to make it pass the control.
- VSS version
v4.1
introduced a new syntax for the unit files that cannot be handled byvss-tools < v4.1
- From
v4.0
vss-tools expects unit file to be explicitly specified or provided in the same directory as the VSS input. VSSv3.1
is the first VSS version including a unit file in the VSS repository. This means vss-tools fromv4.0
onwards cannot handle VSS-versions prior to VSSv3.1
- VSS-tools
v3.1
only supporteddefault
for attributes, resulting in that newer VSS-versions is not supported. - VSS-tools
v4.0
requires case-sensitive for type, resulting in that VSS versionsv3.1
and earlier is not supported.
Contributing
UV
We are using uv as a python package and project manager.
Therefore, a requirement to develop is to install uv
on your host machine.
Check here for official installation instructions. The recommended one is the following, however installing it through pip/pipx works as well:
# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
# Windows
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"
You can then use the following command in the root directory of vss-tools
to install all dependencies:
uv sync
It will create a .venv
in the root of the project.
You can use the venv
like you would without the usage of uv
:
source .venv/bin/activate
vspec --help
Alternatively you can use uv
to run the vspec
tool:
uv run vspec --help
If you have a working setup of direnv, navigating into the directory
activates the venv
(after an initial uv sync
and direnv allow
)
vspec --help
uv
can manage independent python versions.
If you would like to use a specific python version (e.g. python3.12
):
# Python 3.12
uv sync -p python3.12
# From the activated env:
python --version # --> Python 3.12.7
# Python 3.13
uv sync -p python3.13
python --version # --> Python 3.13.0
Pre-commit
This project uses pre-commit which helps to format the source code in a streamlined way.
It is very recommended to use it.
To install hooks you can do pre-commit install
from an activated environment.
Hooks will then run every time you do a git commit
on changed files.
If you would like to run all hooks on all files you can do pre-commit run --all
.
Since pre-commit
dependencies are installed in the virtual environment, it needs
to be activated to be able to run them on a commit.
Installing additional tools
If you intend to run test cases related to vspec
proto exporter you need to install the protobuf compiler.
Please follow official instructions for your OS. For Debian systems it would be:
sudo apt update
sudo apt install -y golang-go protobuf-compiler
protoc --version # Ensure compiler version is 3+