Tests and Checks

This page lists the BobLib make targets and the test files behind them. Run every command from the BobLib repository root. make help prints the current list.

Release gate

Run the full local gate before you trust a branch or cut a release:

make ci PYTHON=.venv/bin/python

make ci runs make lint and then make test. make test runs the Python checks and every Modelica check. The Modelica checks load Modelica Standard Library 4.1.0 and VehicleInterfaces 2.0.2.

GitHub CI runs a lighter gate on pull requests: make lint, make test-python, make modelica-lint, make modelica-smoke, and make modelica-initialization. The separate Release Gate workflow runs the full Modelica set on demand.

Make targets

Target	Action
make lint	Run Ruff over Tests/
make modelica-lint	Check Modelica formatting without rewriting files
make modelica-format	Rewrite Modelica formatting in BobLib and Tests/BobLibTest
make test-python	Run the pure Python tests only
make modelica-deps	Install Modelica 4.1.0 and VehicleInterfaces 2.0.2 through omc msl_setup.mos
make modelica-smoke	Translate the default standard and regression models only
make modelica-translation	Translate the standard entry points and every BobLibTest fixture
make modelica-initialization	Initialize every BobLibTest fixture and compare the baseline CSV
make modelica-physics	Simulate physical validation baselines with scaled runtime budgets
make modelica-regression	Simulate signal-level regressions and smoke-check BobLib and BobLibTest
make test-modelica	Run translation, initialization, physics, and regression checks
make test	Run test-python and test-modelica
make test-pytest	Run every pytest-collected check in one invocation
make ci	Run lint and test

Every Modelica target runs make modelica-deps first. Pass PYTHON=.venv/bin/python to use your virtual environment. The Makefile default is python3.

What each check runs

Target	Test files	What it checks
test-python	Tests/test_modelica_linter.py	The Modelica formatter and linter
modelica-translation	Tests/test_modelica_translation.py, Tests/modelica_translation_checks.py	checkModel on VehicleFMI, VehicleSim, FourPostSim, two regression models, and every BobLibTest fixture
modelica-initialization	Tests/test_modelica_initialization.py, Tests/modelica_initialization_checks.py	Zero-time simulation of every BobLibTest fixture, compared against Tests/modelica_initialization_baseline.csv
modelica-physics	Tests/test_modelica_physics_validation.py	Short simulations compared against Tests/modelica_physics_baseline.csv, with runtime budgets from Tests/modelica_runtime_baseline.csv
modelica-regression	Tests/test_modelica_regression.py, Tests/test_boblib_modelica.py	Signal-level regressions and checkModel smoke checks

Each model is its own pytest case, so a failure names the model in the node id. For example:

Tests/test_modelica_translation.py::test_modelica_model_translates[BobLib.Experiments.Standards.VehicleSim]

Signal regression coverage

Tests/test_modelica_regression.py covers:

• MF52 pure-slip force and moment sanity (BobLibTest.Regression.MF52PureSlipSmoke)
• bilinear aero interpolation (TestAero.TestBilinear2D)
• CFD aero map drag and downforce (TestAero.TestCFDAeroMap)
• VCU power request and torque limit (TestPowertrain.TestVCU)

Tests/test_boblib_modelica.py runs checkModel on the standard entry points, selected subsystem models, and representative chassis, tire, aero, and powertrain fixtures. It also checks the VCU motoring power limit, right-wheel angle mirroring, and source-level structure rules.

Faster loops

Run one test file directly when you work on one area:

python -m pytest Tests/test_boblib_modelica.py -q
python -m pytest Tests/test_modelica_regression.py -q

make modelica-smoke translates only the default standard and regression models. It skips the fixtures and equation counts.

Baselines

Initialization and physics baselines are review signals. A change to a baseline means a model or fixture behaves differently.

To regenerate the initialization baseline after you confirm the new state is expected:

python Tests/modelica_initialization_checks.py --update-baseline

Physics runtime budgets use a hardware-tolerant scale factor. Set BOBLIB_RUNTIME_SCALE or pass --boblib-runtime-scale for strict benchmarks.

When a check fails, see Troubleshooting.