Chore: consolidate pickle hmac signing (#13899)

src/documents/tests/test_caching.py

@@ -1,6 +1,7 @@
-import pickle
-
 from documents.caching import StoredLRUCache
+from paperless.signed_pickle import HMAC_SIZE
+from paperless.signed_pickle import signed_pickle_dumps
+from paperless.signed_pickle import signed_pickle_loads
 
 
 def test_lru_cache_entries() -> None:
@@ -42,4 +43,16 @@ def test_stored_lru_cache_key_ttl(mocker) -> None:
     key, data, timeout = mock_backend.set.call_args[0]
     assert key == "test_key"
     assert timeout == 321
-    assert pickle.loads(data) == {"x": "X", "y": "Y"}
+    assert signed_pickle_loads(data) == {"x": "X", "y": "Y"}
+
+
+def test_stored_lru_cache_rejects_tampered_data(mocker) -> None:
+    serialized_data = bytearray(signed_pickle_dumps({"x": "X"}))
+    serialized_data[HMAC_SIZE] ^= 0xFF
+    mock_backend = mocker.Mock()
+    mock_backend.get.return_value = bytes(serialized_data)
+    cache = StoredLRUCache("test_key", backend=mock_backend)
+
+    cache.load()
+
+    assert cache.get("x") is None

src/documents/tests/test_classifier.py

@@ -19,6 +19,8 @@ from documents.models import MatchingModel
 from documents.models import StoragePath
 from documents.models import Tag
 from documents.tests.utils import DirectoriesMixin
+from paperless.signed_pickle import HMAC_SIZE
+from paperless.signed_pickle import signed_pickle_dumps
 
 
 def dummy_preprocess(content: str, **kwargs):
@@ -265,6 +267,27 @@ class TestClassifier(DirectoriesMixin, TestCase):
             self.assertEqual(mock_preprocess_content.call_count, 2)
             self.assertEqual(mock_transform.call_count, 2)
 
+    def test_vectorize_recomputes_tampered_cache_entry(self) -> None:
+        cached = bytearray(signed_pickle_dumps(["cached vector"]))
+        cached[HMAC_SIZE] ^= 0xFF
+        self.classifier.data_vectorizer = mock.Mock()
+        self.classifier.data_vectorizer.transform.return_value = ["fresh vector"]
+
+        with (
+            mock.patch(
+                "documents.classifier.read_cache.get",
+                return_value=bytes(cached),
+            ),
+            mock.patch("documents.classifier.read_cache.set") as cache_set,
+            mock.patch("documents.classifier.read_cache.touch") as cache_touch,
+        ):
+            result = self.classifier._vectorize("content")
+
+        self.assertEqual(result, ["fresh vector"])
+        self.classifier.data_vectorizer.transform.assert_called_once()
+        cache_set.assert_called_once()
+        cache_touch.assert_not_called()
+
     def test_no_retrain_if_no_change(self) -> None:
         """
         GIVEN: