Linting and Formatting: adding Pylint checks to Ruff

.pre-commit-config.yaml

@@ -9,7 +9,7 @@ repos:
       - id: reuse
   - repo: https://github.com/astral-sh/ruff-pre-commit
     # Ruff version.
-    rev: v0.11.12
+    rev: v0.12.0
     hooks:
       # Run the linter.
       - id: ruff-check

pyproject.toml

@@ -113,6 +113,7 @@ select = [
     "FURB",
     "FLY",
     "SLOT",
+    "PL",
     "NPY",
 ]
 
@@ -123,6 +124,9 @@ combine-as-imports = true
 [tool.ruff.lint.per-file-ignores]
 # Ignore `F811` (redefinition violations) in all examples notebooks since we use redefinition.
 "docs/examples/*.ipynb" = ["F811", "E402"]
+# Pylint was only run in src directory before moving to Ruff
+"tests/*" = ["PLR0915", "PLR0912", "PLR0913"]
+"setup.py" = ["PL"]
 
 [tool.mypy]
 follow_imports = "silent"

src/power_grid_model/_core/buffer_handling.py

@@ -117,7 +117,8 @@ def _get_raw_attribute_data_view(data: np.ndarray, schema: ComponentMetaData, at
     Returns:
         a raw view on the data set.
     """
-    if schema.dtype[attribute].shape == (3,) and data.shape[-1] != 3:
+    supported_dim = 3
+    if schema.dtype[attribute].shape == (supported_dim,) and data.shape[-1] != supported_dim:
         raise ValueError("Given data has a different schema than supported.")
     return _get_raw_data_view(data, dtype=schema.dtype[attribute].base)
 
@@ -185,8 +186,11 @@ def _get_dense_buffer_properties(
     if actual_ndim not in (1, 2):
         raise ValueError(f"Array can only be 1D or 2D. {VALIDATOR_MSG}")
 
-    actual_is_batch = actual_ndim == 2
-    actual_batch_size = shape[0] if actual_is_batch else 1
+    singular_ndim = 1
+    batch_ndim = 2
+
+    actual_is_batch = actual_ndim == batch_ndim
+    actual_batch_size = shape[0] if actual_is_batch else singular_ndim
     n_elements_per_scenario = shape[-1]
     n_total_elements = actual_batch_size * n_elements_per_scenario
 

src/power_grid_model/_core/power_grid_core.py

@@ -460,7 +460,7 @@ def destroy_dataset_mutable(self, dataset: MutableDatasetPtr) -> None:  # type:
         pass  # pragma: no cover
 
     @make_c_binding
-    def dataset_mutable_add_buffer(  # type: ignore[empty-body]
+    def dataset_mutable_add_buffer(  # type: ignore[empty-body]  # noqa: PLR0913
         self,
         dataset: MutableDatasetPtr,
         component: str,

src/power_grid_model/_core/power_grid_model.py

@@ -235,7 +235,7 @@ def _handle_errors(self, continue_on_batch_error: bool, batch_size: int, decode_
             decode_error=decode_error,
         )
 
-    def _calculate_impl(
+    def _calculate_impl(  # noqa: PLR0913
         self,
         calculation_type: CalculationType,
         symmetric: bool,
@@ -300,7 +300,7 @@ def _calculate_impl(
 
         return output_data
 
-    def _calculate_power_flow(
+    def _calculate_power_flow(  # noqa: PLR0913
         self,
         *,
         symmetric: bool = True,
@@ -337,7 +337,7 @@ def _calculate_power_flow(
             experimental_features=experimental_features,
         )
 
-    def _calculate_state_estimation(
+    def _calculate_state_estimation(  # noqa: PLR0913
         self,
         *,
         symmetric: bool = True,
@@ -372,7 +372,7 @@ def _calculate_state_estimation(
             experimental_features=experimental_features,
         )
 
-    def _calculate_short_circuit(
+    def _calculate_short_circuit(  # noqa: PLR0913
         self,
         *,
         calculation_method: CalculationMethod | str = CalculationMethod.iec60909,
@@ -406,7 +406,7 @@ def _calculate_short_circuit(
             experimental_features=experimental_features,
         )
 
-    def calculate_power_flow(
+    def calculate_power_flow(  # noqa: PLR0913
         self,
         *,
         symmetric: bool = True,
@@ -505,7 +505,7 @@ def calculate_power_flow(
             tap_changing_strategy=tap_changing_strategy,
         )
 
-    def calculate_state_estimation(
+    def calculate_state_estimation(  # noqa: PLR0913
         self,
         *,
         symmetric: bool = True,
@@ -599,7 +599,7 @@ def calculate_state_estimation(
             decode_error=decode_error,
         )
 
-    def calculate_short_circuit(
+    def calculate_short_circuit(  # noqa: PLR0913
         self,
         *,
         calculation_method: CalculationMethod | str = CalculationMethod.iec60909,

src/power_grid_model/_core/utils.py

@@ -42,6 +42,10 @@
 from power_grid_model._core.power_grid_meta import initialize_array, power_grid_meta_data
 from power_grid_model._core.typing import ComponentAttributeMapping, _ComponentAttributeMappingDict
 
+SINGULAR_NDIM = 1
+BATCH_NDIM = 2
+UNSUPPORTED_NDIM = 3
+
 
 def is_nan(data) -> bool:
     """
@@ -166,17 +170,17 @@ def get_batch_size(
         for attribute, array in batch_data.items():
             if attribute in sym_attributes:
                 break
-            if array.ndim == 1:
+            if array.ndim == SINGULAR_NDIM:
                 raise TypeError("Incorrect dimension present in batch data.")
-            if array.ndim == 2:
+            if array.ndim == BATCH_NDIM:
                 return 1
             return array.shape[0]
         sym_array = next(iter(batch_data.values()))
 
     sym_array = cast(DenseBatchArray | BatchColumn, sym_array)
-    if sym_array.ndim == 3:
+    if sym_array.ndim == UNSUPPORTED_NDIM:
         raise TypeError("Incorrect dimension present in batch data.")
-    if sym_array.ndim == 1:
+    if sym_array.ndim == SINGULAR_NDIM:
         return 1
     return sym_array.shape[0]
 
@@ -222,7 +226,7 @@ def _split_numpy_array_in_batches(
     Returns:
         A list with a single numpy structured array per batch
     """
-    if data.ndim == 1:
+    if data.ndim == SINGULAR_NDIM:
         return [data]
     if data.ndim in [2, 3]:
         return [data[i, ...] for i in range(data.shape[0])]
@@ -325,7 +329,7 @@ def convert_dataset_to_python_dataset(data: Dataset) -> PythonDataset:
     # It is batch dataset if it is 2D array or a indptr/data structure
     is_batch: bool | None = None
     for component, array in data.items():
-        is_dense_batch = isinstance(array, np.ndarray) and array.ndim == 2
+        is_dense_batch = isinstance(array, np.ndarray) and array.ndim == BATCH_NDIM
         is_sparse_batch = isinstance(array, dict) and "indptr" in array and "data" in array
         if is_batch is not None and is_batch != (is_dense_batch or is_sparse_batch):
             raise ValueError(

src/power_grid_model/utils.py

@@ -406,7 +406,7 @@ def self_test():
             raise PowerGridError from e
 
 
-def _make_test_case(
+def _make_test_case(  # noqa: PLR0913
     *,
     output_path: Path,
     input_data: SingleDataset,

src/power_grid_model/validation/_rules.py

@@ -241,7 +241,7 @@ def not_less_or_equal(val: np.ndarray, *ref: np.ndarray):
     return none_match_comparison(data, component, field, not_less_or_equal, ref_value, NotLessOrEqualError)
 
 
-def all_between(
+def all_between(  # noqa: PLR0913
     data: SingleDataset,
     component: ComponentType,
     field: str,
@@ -281,7 +281,7 @@ def outside(val: np.ndarray, *ref: np.ndarray) -> np.ndarray:
     )
 
 
-def all_between_or_at(
+def all_between_or_at(  # noqa: PLR0913
     data: SingleDataset,
     component: ComponentType,
     field: str,
@@ -331,7 +331,7 @@ def outside(val: np.ndarray, *ref: np.ndarray) -> np.ndarray:
     )
 
 
-def none_match_comparison(
+def none_match_comparison(  # noqa: PLR0913
     data: SingleDataset,
     component: ComponentType,
     field: str,
@@ -554,7 +554,7 @@ def all_valid_enum_values(
     return []
 
 
-def all_valid_associated_enum_values(
+def all_valid_associated_enum_values(  # noqa: PLR0913
     data: SingleDataset,
     component: ComponentType,
     field: str,
@@ -774,7 +774,7 @@ def all_finite(data: SingleDataset, exceptions: dict[ComponentType, list[str]] |
 
             invalid = np.isinf(array[field])
             if invalid.any():
-                ids = data[component]["id"][invalid].flatten().tolist()
+                ids = array["id"][invalid].flatten().tolist()
                 errors.append(InfinityError(component, field, ids))
     return errors
 
@@ -841,7 +841,8 @@ def not_all_missing(data: SingleDataset, fields: list[str], component_type: Comp
         fields: List of fields
         component_type: component type to check
     """
-    if len(fields) < 2:
+    min_fields = 2
+    if len(fields) < min_fields:
         raise ValueError(
             "The fields parameter must contain at least 2 fields. Otherwise use the none_missing function."
         )

src/power_grid_model/validation/_validation.py

@@ -279,7 +279,7 @@ def _process_power_sigma_and_p_q_sigma(
         power_sigma[mask] = np.nansum(q_sigma[mask], axis=asym_axes)
 
 
-def validate_required_values(
+def validate_required_values(  # noqa: PLR0915
     data: SingleDataset, calculation_type: CalculationType | None = None, symmetric: bool = True
 ) -> list[MissingValueError]:
     """
@@ -631,7 +631,7 @@ def validate_branch3(data: SingleDataset, component: ComponentType) -> list[Vali
     return errors
 
 
-def validate_three_winding_transformer(data: SingleDataset) -> list[ValidationError]:
+def validate_three_winding_transformer(data: SingleDataset) -> list[ValidationError]:  # noqa: PLR0915
     errors = validate_branch3(data, ComponentType.three_winding_transformer)
     errors += _all_greater_than_zero(data, ComponentType.three_winding_transformer, "u1")
     errors += _all_greater_than_zero(data, ComponentType.three_winding_transformer, "u2")

src/power_grid_model/validation/errors.py

@@ -13,6 +13,9 @@
 
 from power_grid_model import ComponentType
 
+MIN_FIELDS = 2
+MIN_COMPONENTS = 2
+
 
 class ValidationError(ABC):
     """
@@ -54,6 +57,8 @@ class ValidationError(ABC):
 
     _delimiter: str = " and "
 
+    __hash__ = None  # type: ignore[assignment]
+
     @property
     def component_str(self) -> str:
         """
@@ -165,7 +170,7 @@ def __init__(self, component: ComponentType, fields: list[str], ids: list[int]):
         self.field = sorted(fields)
         self.ids = sorted(ids)
 
-        if len(self.field) < 2:
+        if len(self.field) < MIN_FIELDS:
             raise ValueError(f"{type(self).__name__} expects at least two fields: {self.field}")
 
 
@@ -191,9 +196,9 @@ def __init__(self, fields: list[tuple[ComponentType, str]], ids: list[tuple[Comp
         self.field = sorted(fields)
         self.ids = sorted(ids)
 
-        if len(self.field) < 2:
+        if len(self.field) < MIN_FIELDS:
             raise ValueError(f"{type(self).__name__} expects at least two fields: {self.field}")
-        if len(self.component) < 2:
+        if len(self.component) < MIN_COMPONENTS:
             raise ValueError(f"{type(self).__name__} expects at least two components: {self.component}")
 
 
@@ -241,6 +246,7 @@ class InvalidValueError(SingleFieldValidationError):
 
     _message = "Field {field} contains invalid values for {n} {objects}."
     values: list
+    __hash__ = None  # type: ignore[assignment]
 
     def __init__(self, component: ComponentType, field: str, ids: list[int], values: list):
         super().__init__(component, field, ids)
@@ -265,6 +271,7 @@ class InvalidEnumValueError(SingleFieldValidationError):
 
     _message = "Field {field} contains invalid {enum} values for {n} {objects}."
     enum: Type[Enum] | list[Type[Enum]]
+    __hash__ = None  # type: ignore[assignment]
 
     def __init__(self, component: ComponentType, field: str, ids: list[int], enum: Type[Enum] | list[Type[Enum]]):
         super().__init__(component, field, ids)
@@ -318,6 +325,7 @@ class IdNotInDatasetError(SingleFieldValidationError):
 
     _message = "ID does not exist in {ref_dataset} for {n} {objects}."
     ref_dataset: str
+    __hash__ = None  # type: ignore[assignment]
 
     def __init__(self, component: ComponentType, ids: list[int], ref_dataset: str):
         super().__init__(component=component, field="id", ids=ids)
@@ -345,6 +353,7 @@ class InvalidIdError(SingleFieldValidationError):
 
     _message = "Field {field} does not contain a valid {ref_components} id for {n} {objects}. {filters}"
     ref_components: list[ComponentType]
+    __hash__ = None  # type: ignore[assignment]
 
     def __init__(
         self,
@@ -399,6 +408,8 @@ class ComparisonError(SingleFieldValidationError):
 
     RefType = int | float | str | tuple[int | float | str, ...]
 
+    __hash__ = None  # type: ignore[assignment]
+
     def __init__(self, component: ComponentType, field: str, ids: list[int], ref_value: "ComparisonError.RefType"):
         super().__init__(component, field, ids)
         self.ref_value = ref_value
@@ -512,6 +523,7 @@ class InvalidAssociatedEnumValueError(MultiFieldValidationError):
 
     _message = "The combination of fields {field} results in invalid {enum} values for {n} {objects}."
     enum: Type[Enum] | list[Type[Enum]]
+    __hash__ = None  # type: ignore[assignment]
 
     def __init__(
         self,

src/power_grid_model/validation/utils.py

@@ -86,7 +86,8 @@ def _eval_field_expression(data: np.ndarray, expression: str) -> np.ndarray:
     if len(fields) == 1:
         return data[fields[0]]
 
-    assert len(fields) == 2
+    max_num_fields = 2
+    assert len(fields) == max_num_fields
     zero_div = np.logical_or(np.equal(data[fields[1]], 0.0), np.logical_not(np.isfinite(data[fields[1]])))
     if np.any(zero_div):
         result = np.full_like(data[fields[0]], np.nan)
@@ -126,6 +127,7 @@ def _update_component_array_data(
     Update the data in a numpy array, with another numpy array,
     indexed on the "id" field and only non-NaN values are overwritten.
     """
+    batch_ndim = 2
     if update_data.dtype.names is None:
         raise ValueError("Invalid data format")
 
@@ -142,7 +144,7 @@ def _update_component_array_data(
         nan = _nan_type(component, field, DatasetType.update)
         mask = ~np.isnan(update_data[field]) if np.isnan(nan) else np.not_equal(update_data[field], nan)
 
-        if mask.ndim == 2:
+        if mask.ndim == batch_ndim:
             for phase in range(mask.shape[1]):
                 # find indexers of to-be-updated object
                 sub_mask = mask[:, phase]