[3.12] gh-50644: Forbid pickling of codecs streams (GH-109180) (#109231)

gh-50644: Forbid pickling of codecs streams (GH-109180)

Attempts to pickle or create a shallow or deep copy of codecs streams
now raise a TypeError.

Previously, copying failed with a RecursionError, while pickling
produced wrong results that eventually caused unpickling to fail with
a RecursionError.
(cherry picked from commit d6892c2)

Co-authored-by: Serhiy Storchaka <[email protected]>

Lib/codecs.py

@@ -414,6 +414,9 @@ def __enter__(self):
     def __exit__(self, type, value, tb):
         self.stream.close()
 
+    def __reduce_ex__(self, proto):
+        raise TypeError("can't serialize %s" % self.__class__.__name__)
+
 ###
 
 class StreamReader(Codec):
@@ -663,6 +666,9 @@ def __enter__(self):
     def __exit__(self, type, value, tb):
         self.stream.close()
 
+    def __reduce_ex__(self, proto):
+        raise TypeError("can't serialize %s" % self.__class__.__name__)
+
 ###
 
 class StreamReaderWriter:
@@ -750,6 +756,9 @@ def __enter__(self):
     def __exit__(self, type, value, tb):
         self.stream.close()
 
+    def __reduce_ex__(self, proto):
+        raise TypeError("can't serialize %s" % self.__class__.__name__)
+
 ###
 
 class StreamRecoder:
@@ -866,6 +875,9 @@ def __enter__(self):
     def __exit__(self, type, value, tb):
         self.stream.close()
 
+    def __reduce_ex__(self, proto):
+        raise TypeError("can't serialize %s" % self.__class__.__name__)
+
 ### Shortcuts
 
 def open(filename, mode='r', encoding=None, errors='strict', buffering=-1):

Lib/test/test_codecs.py

@@ -1,7 +1,9 @@
 import codecs
 import contextlib
+import copy
 import io
 import locale
+import pickle
 import sys
 import unittest
 import encodings
@@ -1771,6 +1773,61 @@ def test_readlines(self):
         f = self.reader(self.stream)
         self.assertEqual(f.readlines(), ['\ud55c\n', '\uae00'])
 
+    def test_copy(self):
+        f = self.reader(Queue(b'\xed\x95\x9c\n\xea\xb8\x80'))
+        with self.assertRaisesRegex(TypeError, 'StreamReader'):
+            copy.copy(f)
+        with self.assertRaisesRegex(TypeError, 'StreamReader'):
+            copy.deepcopy(f)
+
+    def test_pickle(self):
+        for proto in range(pickle.HIGHEST_PROTOCOL + 1):
+            with self.subTest(protocol=proto):
+                f = self.reader(Queue(b'\xed\x95\x9c\n\xea\xb8\x80'))
+                with self.assertRaisesRegex(TypeError, 'StreamReader'):
+                    pickle.dumps(f, proto)
+
+
+class StreamWriterTest(unittest.TestCase):
+
+    def setUp(self):
+        self.writer = codecs.getwriter('utf-8')
+
+    def test_copy(self):
+        f = self.writer(Queue(b''))
+        with self.assertRaisesRegex(TypeError, 'StreamWriter'):
+            copy.copy(f)
+        with self.assertRaisesRegex(TypeError, 'StreamWriter'):
+            copy.deepcopy(f)
+
+    def test_pickle(self):
+        for proto in range(pickle.HIGHEST_PROTOCOL + 1):
+            with self.subTest(protocol=proto):
+                f = self.writer(Queue(b''))
+                with self.assertRaisesRegex(TypeError, 'StreamWriter'):
+                    pickle.dumps(f, proto)
+
+
+class StreamReaderWriterTest(unittest.TestCase):
+
+    def setUp(self):
+        self.reader = codecs.getreader('latin1')
+        self.writer = codecs.getwriter('utf-8')
+
+    def test_copy(self):
+        f = codecs.StreamReaderWriter(Queue(b''), self.reader, self.writer)
+        with self.assertRaisesRegex(TypeError, 'StreamReaderWriter'):
+            copy.copy(f)
+        with self.assertRaisesRegex(TypeError, 'StreamReaderWriter'):
+            copy.deepcopy(f)
+
+    def test_pickle(self):
+        for proto in range(pickle.HIGHEST_PROTOCOL + 1):
+            with self.subTest(protocol=proto):
+                f = codecs.StreamReaderWriter(Queue(b''), self.reader, self.writer)
+                with self.assertRaisesRegex(TypeError, 'StreamReaderWriter'):
+                    pickle.dumps(f, proto)
+
 
 class EncodedFileTest(unittest.TestCase):
 
@@ -3346,6 +3403,28 @@ def test_seeking_write(self):
         self.assertEqual(sr.readline(), b'abc\n')
         self.assertEqual(sr.readline(), b'789\n')
 
+    def test_copy(self):
+        bio = io.BytesIO()
+        codec = codecs.lookup('ascii')
+        sr = codecs.StreamRecoder(bio, codec.encode, codec.decode,
+                                  encodings.ascii.StreamReader, encodings.ascii.StreamWriter)
+
+        with self.assertRaisesRegex(TypeError, 'StreamRecoder'):
+            copy.copy(sr)
+        with self.assertRaisesRegex(TypeError, 'StreamRecoder'):
+            copy.deepcopy(sr)
+
+    def test_pickle(self):
+        q = Queue(b'')
+        codec = codecs.lookup('ascii')
+        sr = codecs.StreamRecoder(q, codec.encode, codec.decode,
+                                  encodings.ascii.StreamReader, encodings.ascii.StreamWriter)
+
+        for proto in range(pickle.HIGHEST_PROTOCOL + 1):
+            with self.subTest(protocol=proto):
+                with self.assertRaisesRegex(TypeError, 'StreamRecoder'):
+                    pickle.dumps(sr, proto)
+
 
 @unittest.skipIf(_testinternalcapi is None, 'need _testinternalcapi module')
 class LocaleCodecTest(unittest.TestCase):

Misc/NEWS.d/next/Library/2023-09-09-15-08-37.gh-issue-50644.JUAZOh.rst

@@ -0,0 +1,4 @@
+Attempts to pickle or create a shallow or deep copy of :mod:`codecs` streams
+now raise a TypeError. Previously, copying failed with a RecursionError,
+while pickling produced wrong results that eventually caused unpickling
+to fail with a RecursionError.