gh-133036: deprecate codecs.open

python · methane · Apr 30, 2025 · Apr 27, 2025 · Apr 28, 2025 · Apr 28, 2025
commit 655759b8b6ad2db338fdd53b140b2cc987329a14
diff --git a/Doc/library/codecs.rst b/Doc/library/codecs.rst
@@ -208,6 +208,10 @@ wider range of codecs when working with binary files:
    .. versionchanged:: 3.11
       The ``'U'`` mode has been removed.
 
+   .. deprecated:: next
+
+      :func:`codecs.open` has been superseded by :func:`open`.
+
 
 .. function:: EncodedFile(file, data_encoding, file_encoding=None, errors='strict')
 

@@ -1575,6 +1575,10 @@ Deprecated
   as a single positional argument.
   (Contributed by Serhiy Storchaka in :gh:`109218`.)
 
+* :mod:`codecs`:
+  :func:`codecs.open` is now deprecated. Use :func:`open` instead.
+  (Contributed by Inada Naoki in :gh:`133036`.)
+
 * :mod:`functools`:
   Calling the Python implementation of :func:`functools.reduce` with *function*
   or *sequence* as keyword arguments is now deprecated.

diff --git a/Lib/codecs.py b/Lib/codecs.py
@@ -884,7 +884,6 @@ def __reduce_ex__(self, proto):
 ### Shortcuts
 
 def open(filename, mode='r', encoding=None, errors='strict', buffering=-1):
-
     """ Open an encoded file using the given mode and return
         a wrapped version providing transparent encoding/decoding.
 
@@ -912,8 +911,11 @@ def open(filename, mode='r', encoding=None, errors='strict', buffering=-1):
         .encoding which allows querying the used encoding. This
         attribute is only available if an encoding was specified as
         parameter.
-
     """
+    import warnings
+    warnings.warn("codecs.open() is deprecated. Use open() instead.",
+                  DeprecationWarning, stacklevel=2)
+
     if encoding is not None and \
        'b' not in mode:
         # Force opening of the file in binary mode

diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
@@ -7,6 +7,7 @@
 import unittest
 import encodings
 from unittest import mock
+import warnings
 
 from test import support
 from test.support import os_helper
@@ -28,6 +29,13 @@
 else:
     SIZEOF_WCHAR_T = ctypes.sizeof(ctypes.c_wchar)
 
+def codecs_open_nowarn(filename, mode='r', encoding=None, errors='strict', buffering=-1):
+    with warnings.catch_warnings():
+        warnings.simplefilter("ignore")
+        return codecs.open(
+            filename, mode, encoding=encoding, errors=errors,
+            buffering=buffering)
+
 def coding_checker(self, coder):
     def check(input, expect):
         self.assertEqual(coder(input), (expect, len(input)))
@@ -719,19 +727,19 @@ def test_bug691291(self):
         self.addCleanup(os_helper.unlink, os_helper.TESTFN)
         with open(os_helper.TESTFN, 'wb') as fp:
             fp.write(s)
-        with codecs.open(os_helper.TESTFN, 'r',
+        with codecs_open_nowarn(os_helper.TESTFN, 'r',
                          encoding=self.encoding) as reader:
             self.assertEqual(reader.read(), s1)
 
     def test_invalid_modes(self):
         for mode in ('U', 'rU', 'r+U'):
             with self.assertRaises(ValueError) as cm:
-                codecs.open(os_helper.TESTFN, mode, encoding=self.encoding)
+                codecs_open_nowarn(os_helper.TESTFN, mode, encoding=self.encoding)
             self.assertIn('invalid mode', str(cm.exception))
 
         for mode in ('rt', 'wt', 'at', 'r+t'):
             with self.assertRaises(ValueError) as cm:
-                codecs.open(os_helper.TESTFN, mode, encoding=self.encoding)
+                codecs_open_nowarn(os_helper.TESTFN, mode, encoding=self.encoding)
             self.assertIn("can't have text and binary mode at once",
                           str(cm.exception))
 
@@ -1844,9 +1852,9 @@ def test_all(self):
     def test_open(self):
         self.addCleanup(os_helper.unlink, os_helper.TESTFN)
         for mode in ('w', 'r', 'r+', 'w+', 'a', 'a+'):
-            with self.subTest(mode), \
-                    codecs.open(os_helper.TESTFN, mode, 'ascii') as file:
-                self.assertIsInstance(file, codecs.StreamReaderWriter)
+            with self.subTest(mode), self.assertWarns(DeprecationWarning):
+                with codecs.open(os_helper.TESTFN, mode, 'ascii') as file:
+                    self.assertIsInstance(file, codecs.StreamReaderWriter)
 
     def test_undefined(self):
         self.assertRaises(UnicodeError, codecs.encode, 'abc', 'undefined')
@@ -1863,7 +1871,7 @@ def test_file_closes_if_lookup_error_raised(self):
         mock_open = mock.mock_open()
         with mock.patch('builtins.open', mock_open) as file:
             with self.assertRaises(LookupError):
-                codecs.open(os_helper.TESTFN, 'wt', 'invalid-encoding')
+                codecs_open_nowarn(os_helper.TESTFN, 'wt', 'invalid-encoding')
 
             file().close.assert_called()
 
@@ -2883,7 +2891,7 @@ def test_seek0(self):
         self.addCleanup(os_helper.unlink, os_helper.TESTFN)
         for encoding in tests:
             # Check if the BOM is written only once
-            with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
+            with codecs_open_nowarn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
                 f.write(data)
                 f.write(data)
                 f.seek(0)
@@ -2892,7 +2900,7 @@ def test_seek0(self):
                 self.assertEqual(f.read(), data * 2)
 
             # Check that the BOM is written after a seek(0)
-            with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
+            with codecs_open_nowarn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
                 f.write(data[0])
                 self.assertNotEqual(f.tell(), 0)
                 f.seek(0)
@@ -2901,7 +2909,7 @@ def test_seek0(self):
                 self.assertEqual(f.read(), data)
 
             # (StreamWriter) Check that the BOM is written after a seek(0)
-            with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
+            with codecs_open_nowarn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
                 f.writer.write(data[0])
                 self.assertNotEqual(f.writer.tell(), 0)
                 f.writer.seek(0)
@@ -2911,7 +2919,7 @@ def test_seek0(self):
 
             # Check that the BOM is not written after a seek() at a position
             # different than the start
-            with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
+            with codecs_open_nowarn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
                 f.write(data)
                 f.seek(f.tell())
                 f.write(data)
@@ -2920,7 +2928,7 @@ def test_seek0(self):
 
             # (StreamWriter) Check that the BOM is not written after a seek()
             # at a position different than the start
-            with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
+            with codecs_open_nowarn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
                 f.writer.write(data)
                 f.writer.seek(f.writer.tell())
                 f.writer.write(data)

diff --git a/Misc/NEWS.d/next/Library/2025-04-27-15-21-05.gh-issue-133036.HCNYA7.rst b/Misc/NEWS.d/next/Library/2025-04-27-15-21-05.gh-issue-133036.HCNYA7.rst
@@ -0,0 +1,2 @@
+:func:`codecs.open` is now deprecated. Use :func:`open` instead. Contributed
+by Inada Naoki.
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		:func:`codecs.open` is now deprecated. Use :func:`open` instead. Contributed
		by Inada Naoki.
Comment thread AA-Turner marked this conversation as resolved.