fix(search): validate topic hierarchy input

backend/test_search_topic_taxonomy.py

@@ -66,6 +66,37 @@ class SearchTopicTaxonomyTest(unittest.TestCase):
             with self.assertRaisesRegex(RuntimeError, "normalized document 1"):
                 build(root)
 
+    def test_rebuild_rejects_normalized_document_without_classifier_list(self):
+        with tempfile.TemporaryDirectory() as temporary:
+            root = Path(temporary)
+            document = root / "documents/1"
+            document.mkdir(parents=True)
+            with closing(sqlite3.connect(root / "manifest.sqlite3")) as db:
+                db.execute("CREATE TABLE documents (code TEXT, state TEXT)")
+                db.execute("INSERT INTO documents VALUES ('1', 'success')")
+                db.commit()
+            (document / "document.json").write_text("{}", encoding="utf-8")
+
+            with self.assertRaisesRegex(RuntimeError, "general_classifiers list"):
+                build(root)
+
+    def test_rebuild_rejects_invalid_nested_classifier_branch(self):
+        with tempfile.TemporaryDirectory() as temporary:
+            root = Path(temporary)
+            document = root / "documents/1"
+            document.mkdir(parents=True)
+            with closing(sqlite3.connect(root / "manifest.sqlite3")) as db:
+                db.execute("CREATE TABLE documents (code TEXT, state TEXT)")
+                db.execute("INSERT INTO documents VALUES ('1', 'success')")
+                db.commit()
+            (document / "document.json").write_text(
+                json.dumps({"general_classifiers": [{"Name": {"Rus": "Корень"}, "GeneralClassifiers": "bad"}]}),
+                encoding="utf-8",
+            )
+
+            with self.assertRaisesRegex(RuntimeError, "non-list GeneralClassifiers branch"):
+                build(root)
+
 
 if __name__ == "__main__":
     unittest.main()

tools/build_search_topic_taxonomy.py

@@ -65,12 +65,14 @@ def node_id(path):
     return "source-" + hashlib.sha256(raw.encode("utf-8")).hexdigest()[:16]
 
 
-def add_nodes(nodes, classifiers, parent_path, seen):
-    if not isinstance(classifiers, list):
+def add_nodes(nodes, classifiers, parent_path, seen, document_code):
+    if classifiers is None:
         return
+    if not isinstance(classifiers, list):
+        raise RuntimeError(f"normalized document {document_code} contains a non-list GeneralClassifiers branch")
     for source in classifiers:
         if not isinstance(source, dict):
-            continue
+            raise RuntimeError(f"normalized document {document_code} contains a non-object GeneralClassifiers node")
         name = source.get("Name") if isinstance(source.get("Name"), dict) else {}
         labels = {"ru": clean(name.get("Rus")), "ky": clean(name.get("Kyr"))}
         code = source.get("Code")
@@ -92,7 +94,7 @@ def add_nodes(nodes, classifiers, parent_path, seen):
         if identifier not in seen:
             node["document_count"] += 1
             seen.add(identifier)
-        add_nodes(nodes, source.get("GeneralClassifiers"), path, seen)
+        add_nodes(nodes, source.get("GeneralClassifiers"), path, seen, document_code)
 
 
 def emit_node(node, children):
@@ -125,13 +127,16 @@ def build(normalized_root):
             raise RuntimeError(f"cannot read normalized document {code}: {error}") from error
         if not isinstance(data, dict):
             raise RuntimeError(f"normalized document {code} must contain a JSON object")
-        return code, data.get("general_classifiers", [])
+        classifiers = data.get("general_classifiers")
+        if not isinstance(classifiers, list):
+            raise RuntimeError(f"normalized document {code} must contain a general_classifiers list")
+        return code, classifiers
 
     with ThreadPoolExecutor(max_workers=24) as pool:
         for offset in range(0, len(codes), 512):
             for code, classifiers in pool.map(read, codes[offset:offset + 512]):
                 seen = set()
-                add_nodes(nodes, classifiers, [], seen)
+                add_nodes(nodes, classifiers, [], seen, code)
                 for source in classifiers if isinstance(classifiers, list) else []:
                     if not isinstance(source, dict):
                         continue