Contribute to This Plugin

Running the Tests

Install dev dependencies

cd packages/nomad-inl-base
pip install -e ".[dev]"

Or, if the project uses uv:

cd packages/nomad-inl-base
uv sync --extra dev

Run all tests

pytest tests/ -v

Run specific subsets

Goal	Command
Pure utility tests (no NOMAD infra)	pytest tests/test_utils.py -v
Schema normalization tests	pytest tests/schema_packages/ -v
Parser integration tests	pytest tests/parsers/ -v
Discover all tests without running	pytest --co -q

---

Test Layout

tests/
├── conftest.py                    # Shared fixtures (caplog, parsed_archive, sem_zip)
├── test_utils.py                  # Pure unit tests – no NOMAD infrastructure needed
├── parsers/
│   ├── __init__.py
│   └── test_parser.py             # Integration tests for all parsers
├── schema_packages/
│   ├── __init__.py
│   └── test_schema_package.py     # Schema normalization tests
└── data/
    ├── test.archive.yaml          # Demo schema fixture
    ├── schemas/                   # Synthetic YAML archives for schema tests
    │   ├── substrate.archive.yaml
    │   ├── thinfilm.archive.yaml
    │   ├── cleaning.archive.yaml
    │   └── spin_coating.archive.yaml
    └── <parser data files>        # Real instrument output files

---

Adding a New Parser Test

1. Add a data file – place the instrument output file in tests/data/.

2. Write the test using the parsed_archive and caplog fixtures:

   @pytest.mark.parametrize(
       'parsed_archive, caplog',
       [('tests/data/my_file.txt', None)],
       indirect=True,
   )
   def test_my_parser(parsed_archive, caplog):
       entry = parsed_archive.data
       assert entry is not None
       assert entry.some_quantity == expected_value
   

   The parsed_archive fixture calls nomad.client.parse() and nomad.client.normalize_all() on the file, and yields the resulting archive. It cleans up the generated .archive.json after the test.

   The caplog fixture uses structlog LogCapture and fails the test if any error or critical level log entry is emitted.

3. Remove the skip mark (if the test was previously skipped due to missing data).

---

Adding a New Schema Test

1. Create a YAML archive in tests/data/schemas/ that instantiates your schema:

   data:
     m_def: nomad_inl_base.schema_packages.my_module.MySchema
     my_quantity: some_value
   

2. Write the test:

   @pytest.mark.parametrize(
       'parsed_archive, caplog',
       [('tests/data/schemas/my_schema.archive.yaml', None)],
       indirect=True,
   )
   def test_my_schema(parsed_archive, caplog):
       entry = parsed_archive.data
       assert entry.my_quantity == 'some_value'
   

---

Currently Skipped Tests

Four parsers have no test data yet and are marked with @pytest.mark.skip:

Test	Reason
test_cv_parser	No CV data file available
test_ed_parser	No ED data file available
test_emsa_edx_parser	No EMSA/EDX data file available
test_bruker_afm_parser	No Bruker AFM binary file available

To enable these tests, add the corresponding data file to tests/data/ and remove the @pytest.mark.skip decorator.