fix(cutter): preserve mixed Unicode output within the frame limit

Rebase the change from PR #3776 (21ad0bc2f1) while preserving current document-identity and host-version tests.

Review-support branch for the original contributor: the author-disabled maintainer-edit setting prevents updating the original PR directly.

bridge/cutter/rea_bridge.py

@@ -783,14 +783,18 @@ def _within_response_limit(value: Any) -> bool:
         current = pending.pop()
         if isinstance(current, str):
             size += 2
+            if len(current) > MAX_FRAME_BYTES - size:
+                return False
             if current.isascii():
                 size += len(current)
-                size += current.count('"') + current.count("\\")
-                size += sum(5 for character in current if ord(character) < 0x20)
             else:
-                size += 4 * len(current)
-                size += current.count('"') + current.count("\\")
-                size += sum(5 for character in current if ord(character) < 0x20)
+                # Count UTF-8 bytes without allocating a copy of the whole output.
+                size += sum(
+                    len(current[start:start + 65536].encode("utf-8"))
+                    for start in range(0, len(current), 65536)
+                )
+            size += current.count('"') + current.count("\\")
+            size += sum(5 for character in current if ord(character) < 0x20)
         elif isinstance(current, dict):
             identity = id(current)
             if identity in seen:

tests/fixtures/cutter_bridge_protocol_test.py

@@ -320,6 +320,34 @@ class HostVersionTest(unittest.TestCase):
                 self.assertIsNone(bridge._RequestDispatcher().cutter_version)
 
 
+class ResponseEncodingTest(unittest.TestCase):
+    def setUp(self):
+        self.server = bridge._BridgeServer.__new__(bridge._BridgeServer)
+        self.server.token = "test-token"
+        self.server.session_id = str(uuid.uuid4())
+
+    def test_large_mixed_unicode_output_fits_the_authenticated_frame(self):
+        for suffix in ("a", "é", "雪", "😀"):
+            with self.subTest(suffix=suffix):
+                output = "a" * (5 * 1024 * 1024 - 1) + suffix
+                value = {"ok": True, "output": output, "execution_state": "complete"}
+                encoded = self.server._encode_authenticated_response(value, "a" * 64, "b" * 64)
+                self.assertIsNotNone(encoded)
+                self.assertLessEqual(len(encoded), bridge.MAX_FRAME_BYTES)
+                frame = json.loads(encoded)
+                self.assertEqual(json.loads(frame["response"]), value)
+                expected = hmac.new(
+                    self.server.token.encode(),
+                    f"rea-cutter-response-v2:{self.server.session_id}:{'a' * 64}:{'b' * 64}:{frame['response']}".encode(),
+                    hashlib.sha256,
+                ).hexdigest()
+                self.assertTrue(hmac.compare_digest(frame["proof"], expected))
+        self.assertIsNone(self.server._encode_authenticated_response({
+            "ok": True, "output": "😀" * (5 * 1024 * 1024),
+            "execution_state": "complete",
+        }, "a" * 64, "b" * 64))
+
+
 class ProtocolProducerTest(unittest.TestCase):
     def test_rejects_legacy_peers_before_ui_dispatch(self):
         with socket.create_connection(("127.0.0.1", self.server.port), timeout=2) as connection: