Add range validation

nomenclature/processor/data_validator.py

@@ -7,7 +7,7 @@
 from pandas import concat
 from pyam import IamDataFrame
 from pyam.logging import adjust_log_level
-from pydantic import computed_field, field_validator, model_validator
+from pydantic import computed_field, field_validator, model_validator, Field
 
 from nomenclature.definition import DataStructureDefinition
 from nomenclature.error import ErrorCollector
@@ -86,9 +86,49 @@ def criteria(self):
         )
 
 
+class DataValidationCriteriaRange(DataValidationCriteria):
+    range: list[float] = Field(..., min_length=2, max_length=2)
+
+    @model_validator(mode="after")
+    def check_range_is_valid(self):
+        if self.range[0] > self.range[1]:
+            raise ValueError("Validation range is invalid: " + str(self.criteria))
+        return self
+
+    @computed_field
+    def upper_bound(self) -> float:
+        return self.range[1]
+
+    @computed_field
+    def lower_bound(self) -> float:
+        return self.range[0]
+
+    @property
+    def validation_args(self):
+        """Attributes used for validation (as bounds)."""
+        return self.model_dump(
+            exclude_none=True,
+            exclude_unset=True,
+            exclude=["warning_level", "range"],
+        )
+
+    @property
+    def criteria(self):
+        return self.model_dump(
+            exclude_none=True,
+            exclude_unset=True,
+            exclude=["warning_level", "lower_bound", "upper_bound"],
+        )
+
+
 class DataValidationCriteriaMultiple(IamcDataFilter):
     validation: (
-        list[DataValidationCriteriaValue | DataValidationCriteriaBounds] | None
+        list[
+            DataValidationCriteriaValue
+            | DataValidationCriteriaBounds
+            | DataValidationCriteriaRange
+        ]
+        | None
     ) = None
 
     @model_validator(mode="after")

tests/data/validation/validate_data/validate_warning_range.yaml

@@ -0,0 +1,11 @@
+ - variable: Primary Energy
+   year: 2010
+   validation:
+    - range: [ 1, 5 ]
+    - warning_level: low
+      upper_bound: 2.5
+      lower_bound: 1
+ - variable: Primary Energy|Coal
+   year: 2010
+   upper_bound: 5
+   lower_bound: 1

tests/test_validate_data.py

@@ -137,7 +137,7 @@ def test_DataValidator_apply_fails(simple_df, file, item_1, item_2, item_3, capl
 
 @pytest.mark.parametrize(
     "file, value",
-    [("joined", 6.0), ("joined", 3.0), ("legacy", 6.0)],
+    [("joined", 6.0), ("joined", 3.0), ("legacy", 6.0), ("range", 6.0)],
 )
 def test_DataValidator_validate_with_warning(file, value, simple_df, caplog):
     """Checks that failed validation rows are printed in log."""
@@ -154,6 +154,7 @@ def test_DataValidator_validate_with_warning(file, value, simple_df, caplog):
     0  model_a   scen_a  World  Primary Energy  EJ/yr  2010    6.0         error
     1  model_a   scen_b  World  Primary Energy  EJ/yr  2010    7.0         error"""
     )
+
     if file == "legacy":
         # prints both error and low warning levels for legacy format
         # because these are treated as independent validation-criteria
@@ -164,6 +165,11 @@ def test_DataValidator_validate_with_warning(file, value, simple_df, caplog):
     0  model_a   scen_a  World  Primary Energy  EJ/yr  2010    6.0           low
     1  model_a   scen_b  World  Primary Energy  EJ/yr  2010    7.0           low"""
 
+    if file == "range":
+        failed_validation_message = failed_validation_message.replace(
+            "upper_bound: 5.0, lower_bound: 1.0", "range: [1.0, 5.0]"
+        )
+
     if value == 3.0:
         # prints each warning level when each is triggered by different rows
         failed_validation_message = """