fix(validation): handle exceptions in sampling and processing

src/macro_lactone_toolkit/validation/sampling.py

@@ -23,9 +23,14 @@ def stratified_sample_by_ring_size(
     ring_sizes = []
 
     for smiles in df[smiles_col]:
-        result = analyzer.classify_macrocycle(smiles)
+        try:
-        classifications.append(result.classification)
+            result = analyzer.classify_macrocycle(smiles)
-        ring_sizes.append(result.ring_size)
+            classifications.append(result.classification)
+            ring_sizes.append(result.ring_size)
+        except Exception:
+            # Handle ambiguous or invalid molecules
+            classifications.append("not_macrolactone")
+            ring_sizes.append(None)
 
     df = df.copy()
     df["_classification"] = classifications

src/macro_lactone_toolkit/validation/validator.py

@@ -88,9 +88,20 @@ class MacrolactoneValidator:
         name = row.get("molecule_pref_name", None)
 
         # Classify
-        classification_result = self.analyzer.classify_macrocycle(smiles)
+        try:
-        classification = classification_result.classification
+            classification_result = self.analyzer.classify_macrocycle(smiles)
-        ring_size = classification_result.ring_size
+            classification = classification_result.classification
+            ring_size = classification_result.ring_size
+        except Exception as e:
+            # Handle classification errors - treat as not_macrolactone
+            classification = ClassificationType.NOT_MACROLACTONE
+            ring_size = None
+            classification_result = type('obj', (object,), {
+                'classification': classification,
+                'ring_size': ring_size,
+                'primary_reason_code': 'classification_error',
+                'primary_reason_message': str(e)
+            })()
 
         # Create parent record
         parent = ParentMolecule(