Pydantic model for deployment config

environment.yml

@@ -3,6 +3,7 @@ channels:
   - conda-forge
 dependencies:
   - python>=3.10
+  - pydantic
   - numpy
   - pip
   - xarray

pyglider/_config_components.py

@@ -0,0 +1,142 @@
+from typing import Optional
+
+from pydantic import BaseModel, HttpUrl
+
+
+class Metadata(BaseModel):
+    acknowledgement: str
+    comment: str
+    contributor_name: str
+    contributor_role: str
+    creator_email: str
+    creator_name: str
+    creator_url: HttpUrl
+    deployment_id: str
+    deployment_name: str
+    deployment_start: str
+    deployment_end: str
+    format_version: str
+    glider_name: str
+    glider_serial: str
+    glider_model: str
+    glider_instrument_name: str
+    glider_wmo: str
+    institution: str
+    keywords: str
+    keywords_vocabulary: str
+    license: str
+    metadata_link: HttpUrl
+    Metadata_Conventions: str
+    naming_authority: str
+    platform_type: str
+    processing_level: str
+    project: str
+    project_url: HttpUrl
+    publisher_email: str
+    publisher_name: str
+    publisher_url: HttpUrl
+    references: str
+    sea_name: str
+    source: str
+    standard_name_vocabulary: str
+    summary: str
+    transmission_system: str
+    wmo_id: str
+
+
+class Device(BaseModel):
+    make: str
+    model: str
+    serial: str
+    long_name: Optional[str] = None
+    make_model: Optional[str] = None
+    factory_calibrated: Optional[str] = None
+    calibration_date: Optional[str] = None
+    calibration_report: Optional[str] = None
+    comment: Optional[str] = None
+
+
+class GliderDevices(BaseModel):
+    pressure: Device
+    ctd: Device
+    optics: Device
+    oxygen: Device
+
+
+class NetCDFVariable(BaseModel):
+    source: str
+    long_name: Optional[str] = None
+    standard_name: Optional[str] = None
+    units: Optional[str] = None
+    axis: Optional[str] = None
+    coordinates: Optional[str] = None
+    conversion: Optional[str] = None
+    comment: Optional[str] = None
+    observation_type: Optional[str] = None
+    platform: Optional[str] = None
+    reference: Optional[str] = None
+    valid_max: Optional[float] = None
+    valid_min: Optional[float] = None
+    coordinate_reference_frame: Optional[str] = None
+    instrument: Optional[str] = None
+    accuracy: Optional[float] = None
+    precision: Optional[float] = None
+    resolution: Optional[float] = None
+    positive: Optional[str] = None
+    reference_datum: Optional[str] = None
+    coarsen: Optional[int] = None
+
+
+class NetCDFVariables(BaseModel):
+    timebase: Optional[NetCDFVariable] = (
+        None  #! Is this required? `example-slocum`` doesn't have it
+    )
+    time: NetCDFVariable
+    latitude: NetCDFVariable
+    longitude: NetCDFVariable
+    heading: NetCDFVariable
+    pitch: NetCDFVariable
+    roll: NetCDFVariable
+    conductivity: NetCDFVariable
+    temperature: NetCDFVariable
+    pressure: NetCDFVariable
+    chlorophyll: NetCDFVariable
+    cdom: NetCDFVariable
+    backscatter_700: NetCDFVariable
+    oxygen_concentration: NetCDFVariable
+    temperature_oxygen: Optional[NetCDFVariable] = (
+        None  #! Is this required? `example-slocum`` doesn't have it
+    )
+
+
+class ProfileVariable(BaseModel):
+    comment: str
+    long_name: str
+    valid_max: Optional[float] = None
+    valid_min: Optional[float] = None
+    observation_type: Optional[str] = None
+    platform: Optional[str] = None
+    standard_name: Optional[str] = None
+    units: Optional[str] = None
+    calendar: Optional[str] = None
+    type: Optional[str] = None
+    calibration_date: Optional[str] = None
+    calibration_report: Optional[str] = None
+    factory_calibrated: Optional[str] = None
+    make_model: Optional[str] = None
+    serial_number: Optional[str] = None
+
+
+class ProfileVariables(BaseModel):
+    profile_id: ProfileVariable
+    profile_time: ProfileVariable
+    profile_time_start: ProfileVariable
+    profile_time_end: ProfileVariable
+    profile_lat: ProfileVariable
+    profile_lon: ProfileVariable
+    u: ProfileVariable
+    v: ProfileVariable
+    lon_uv: ProfileVariable
+    lat_uv: ProfileVariable
+    time_uv: ProfileVariable
+    instrument_ctd: ProfileVariable

pyglider/config.py

@@ -0,0 +1,33 @@
+import yaml
+from pydantic import BaseModel
+
+from pyglider._config_components import (
+    GliderDevices,
+    Metadata,
+    NetCDFVariables,
+    ProfileVariables,
+)
+
+__all__ = ['Deployment', 'dump_yaml']
+
+
+class Deployment(BaseModel):
+    metadata: Metadata
+    glider_devices: GliderDevices
+    netcdf_variables: NetCDFVariables
+    profile_variables: ProfileVariables
+
+    @classmethod
+    def load_yaml(cls, yaml_str: str) -> 'Deployment':
+        """Load a yaml string into a Deployment model."""
+        return _generic_load_yaml(yaml_str, cls)
+
+
+def dump_yaml(model: BaseModel) -> str:
+    """Dump a pydantic model to a yaml string."""
+    return yaml.safe_dump(model.model_dump(), default_flow_style=False)
+
+
+def _generic_load_yaml(data: str, model: BaseModel) -> BaseModel:
+    """Load a yaml string into a pydantic model."""
+    return model.model_validate(yaml.safe_load(data))

tests/test_config.py

@@ -0,0 +1,55 @@
+from pathlib import Path
+from typing import Callable
+
+import pytest
+import yaml
+
+from pyglider.config import Deployment
+
+library_dir = Path(__file__).parent.parent.absolute()
+example_dir = library_dir / 'tests/example-data/'
+
+VALID_CONFIG_PATHS = [
+    example_dir / 'example-slocum/deploymentRealtime.yml',
+    example_dir / 'example-seaexplorer/deploymentRealtime.yml',
+    # TODO: Add other valid example configs?
+]
+
+
+@pytest.fixture(params=VALID_CONFIG_PATHS)
+def valid_config(request):
+    return request.param.read_text()
+
+
+def test_valid_config(valid_config):
+    """Checks all valid configs can be loaded."""
+    Deployment.load_yaml(valid_config)
+
+
+def patch_config(yaml_str: str, f: Callable) -> str:
+    """Patch a yaml string with a function.
+    Function should do an in place operation on a dictionary/list.
+    """
+    d = yaml.safe_load(yaml_str)
+    f(d)
+    return yaml.safe_dump(d, default_flow_style=False)
+
+
+@pytest.mark.parametrize(
+    'input_, expected, f',
+    [
+        (
+            'a: 1\n' 'b: 2\n',
+            'a: 1\n' 'b: 3\n',
+            lambda d: d.update({'b': 3}),
+        ),
+    ],
+)
+def test_patch_config(input_, expected, f):
+    assert patch_config(input_, f) == expected
+
+
+# TODO: Stress test the model by taking existing configs and modifying them in breaking ways
+
+
+def test_incorrect_date_format(): ...