1212: Implement case-insensitive containment checks for column comparisons

cdisc_rules_engine/check_operators/dataframe_operators.py

@@ -624,9 +624,16 @@ def contains_case_insensitive(self, other_value):
                 comparison_data, self.value[target]
             )
         elif self.value.is_series(comparison_data):
-            results = self._series_is_in(
-                self.convert_string_data_to_lower(self.value[target]),
-                self.convert_string_data_to_lower(comparison_data),
+            # column vs column case: perform element-wise case-insensitive substring check
+            target_series = self.convert_string_data_to_lower(self.value[target])
+            comparison_series = self.convert_string_data_to_lower(comparison_data)
+            results = target_series.combine(
+                comparison_series,
+                lambda t, c: (
+                    vectorized_case_insensitive_is_in(c, [t])[0]
+                    if pd.notna(t) and pd.notna(c)
+                    else False
+                ),
             )
         else:
             results = vectorized_case_insensitive_is_in(

tests/unit/test_check_operators/test_containment_checks.py

@@ -34,14 +34,20 @@ def test_contains(data, comparator, dataset_type, expected_result):
     "data,comparator,dataset_type,expected_result",
     [
         (
-            {"target": ["Ctt", "Btt", "A"], "VAR2": ["a", "btt", "lll"]},
+            {"target": ["A", "Btt", "Ctt"], "VAR2": ["a", "btt", "lll"]},
             "VAR2",
             DaskDataset,
             [True, True, False],
         ),
         (
-            {"target": [["A", "B", "C"], ["A", "B", "L"], ["L", "Q", "R"]]},
-            "l",
+            {
+                "target": [
+                    ["A", "B", "C"],
+                    ["A", "left hind limb", "L"],
+                    ["L", "NON-ULCERATED left hind limb", "R"],
+                ]
+            },
+            "LEFT HIND LIMB",
             PandasDataset,
             [False, True, True],
         ),
@@ -86,7 +92,7 @@ def test_does_not_contain(data, comparator, dataset_type, expected_result):
     "data,comparator,dataset_type,expected_result",
     [
         (
-            {"target": ["Ctt", "Btt", "A"], "VAR2": ["a", "btt", "lll"]},
+            {"target": ["A", "Btt", "Ctt"], "VAR2": ["a", "btt", "lll"]},
             "VAR2",
             DaskDataset,
             [False, False, True],
@@ -600,3 +606,39 @@ def test_is_column_of_iterables(column_data, expected):
     dataframe_operator = DataframeType({"value": df})
     result = dataframe_operator.is_column_of_iterables(df["col"])
     assert result == expected
+
+
+@pytest.mark.parametrize(
+    "data,target_col,comparator_col,dataset_type,expected_result",
+    [
+        (
+            {
+                "PMSTRESC": [
+                    "2.0",
+                    "2.0",
+                    "NON-ULCERATED left hind limb",
+                    "LEFT LEG",
+                ],
+                "PMLOC": [
+                    "LEFT HIND LIMB ",
+                    "LEFT HIND LIMB ",
+                    "LEFT HIND LIMB",
+                    "left leg",
+                ],
+            },
+            "PMSTRESC",
+            "PMLOC",
+            PandasDataset,
+            [False, False, True, True],
+        ),
+    ],
+)
+def test_contains_case_insensitive_column_vs_column(
+    data, target_col, comparator_col, dataset_type, expected_result
+):
+    df = dataset_type.from_dict(data)
+    dataframe_operator = DataframeType({"value": df})
+    result = dataframe_operator.contains_case_insensitive(
+        {"target": target_col, "comparator": comparator_col}
+    )
+    assert result.equals(df.convert_to_series(expected_result))