work in progress
For more information about the arena and the LASSO platform, visit our website: https://softwareobservatorium.github.io/
Assumes Python3 >= 3.12 (tested with Ubuntu 24.04 LTS)
Example using Python's virtual environment (you need to have virtualenv installed; e.g. using pip)
pip3 install virtualenv # depending on your OS/Python distributionClone and install all required dependencies
git clone https://github.com/SoftwareObservatorium/arena-python.git python3 -m venv arena source arena/bin/activate cd arena-python/
Install packages
pip3 install -r requirements.txt
project is managed with hatch (https://hatch.pypa.io/latest/)
git clone https://github.com/SoftwareObservatorium/arena-python.git python3 -m venv arena source arena/bin/activate cd arena-python/
Install hatch
pip3 install hatch
Build project
python3 -m hatch build
Test project
python3 -m hatch testTo run the example notebooks, the following dependencies need to be installed as well
pip3 install jupyter
pip3 install pandas
# start jupyter locally
jupyter labLQL parser in python using antlr4 grammar
- antlr4-python3-runtime
- lqlparser.py
Sequence sheet notation and engine realized in python.
Adaptation strategies are realized in
SMs and SRMs are based on pandas DataFrame
see examples in notebooks
- Load units under tests from local machine
- Use builtin types (collections example)
- Use LLMs (via Ollama/OpenAI) to generate units under test (i.e., code candidates)
- Adapt to function signature mismatches - Use LLMs (via Ollama/OpenAI) to generate units under test (i.e., code candidates)
Tests that demonstrate the arena test driver can be found in tests/.
- arena_test.py contains integration tests (system tests)
docker build -t swtrepo.informatik.uni-mannheim.de:5050/docker/lasso/arena-python:latest -f Dockerfile .docker run --network host -v "$PWD":/project swtrepo.informatik.uni-mannheim.de:5050/docker/lasso/arena-python:latest \
--someparam=XXX