fix: handle empty SPOKE API response in get_context_using_spoke_api

kg_rag/test/test_spoke_api_empty_response.py

@@ -0,0 +1,72 @@
+"""
+Unit test for issue #42: IndexError when SPOKE API returns empty list for a node.
+https://github.com/BaranziniLab/KG_RAG/issues/42
+"""
+import unittest
+from unittest.mock import patch, MagicMock
+
+
+class TestGetContextUsingSpokeApiEmptyResponse(unittest.TestCase):
+
+    @patch("kg_rag.utility.get_spoke_api_resp")
+    def test_empty_node_context_does_not_raise(self, mock_api):
+        """When SPOKE API returns an empty list, function should not raise IndexError."""
+        from kg_rag.utility import get_context_using_spoke_api
+
+        # First call returns types endpoint data
+        types_response = MagicMock()
+        types_response.json.return_value = {
+            "nodes": {"Disease": {}, "Protein": {}, "Compound": {}},
+            "edges": {"DaG": {}, "CtD": {}},
+        }
+
+        # Second call (neighborhood endpoint) returns empty list — the bug scenario
+        empty_response = MagicMock()
+        empty_response.json.return_value = []
+
+        mock_api.side_effect = [types_response, empty_response]
+
+        context, df = get_context_using_spoke_api("neurofibromatosis 2")
+
+        self.assertEqual(context, "")
+        self.assertTrue(df.empty)
+
+    @patch("kg_rag.utility.get_spoke_api_resp")
+    def test_non_empty_node_context_includes_identifier(self, mock_api):
+        """When SPOKE API returns data, identifier info is appended to context."""
+        from kg_rag.utility import get_context_using_spoke_api
+
+        types_response = MagicMock()
+        types_response.json.return_value = {
+            "nodes": {"Disease": {}, "Gene": {}},
+            "edges": {"DaG_association": {}},
+        }
+
+        # Edge types in SPOKE use underscores (e.g. "DaG_association")
+        # Nodes have no underscore in neo4j_type
+        node_response = MagicMock()
+        node_response.json.return_value = [
+            {
+                "data": {
+                    "neo4j_type": "Disease",
+                    "id": "D001",
+                    "properties": {
+                        "name": "psoriasis",
+                        "identifier": "DOID:8893",
+                        "source": "DOID",
+                    },
+                }
+            },
+        ]
+
+        mock_api.side_effect = [types_response, node_response]
+
+        context, df = get_context_using_spoke_api("psoriasis")
+
+        # Identifier info should be appended when node_context is non-empty
+        self.assertIn("DOID", context)
+        self.assertIn("DOID:8893", context)
+
+
+if __name__ == "__main__":
+    unittest.main()

kg_rag/utility.py

@@ -125,7 +125,8 @@ def get_context_using_spoke_api(node_value):
     merge_2.loc[:, "predicate"] = merge_2.edge_type.apply(lambda x:x.split("_")[0])
     merge_2.loc[:, "context"] =  merge_2.source + " " + merge_2.predicate.str.lower() + " " + merge_2.target + " and Provenance of this association is " + merge_2.provenance + "."
     context = merge_2.context.str.cat(sep=' ')
-    context += node_value + " has a " + node_context[0]["data"]["properties"]["source"] + " identifier of " + node_context[0]["data"]["properties"]["identifier"] + " and Provenance of this is from " + node_context[0]["data"]["properties"]["source"] + "."
+    if node_context:
+        context += node_value + " has a " + node_context[0]["data"]["properties"]["source"] + " identifier of " + node_context[0]["data"]["properties"]["identifier"] + " and Provenance of this is from " + node_context[0]["data"]["properties"]["source"] + "."
     return context, merge_2
 
 #         if edge_evidence: