8000 gh-133036: Deprecate codecs.open (#133038) · python/cpython@4e294f6 · GitHub
[go: up one dir, main page]
Skip to content

Commit 4e294f6

Browse files
methanehugovkvstinner
authored
gh-133036: Deprecate codecs.open (#133038)
Co-authored-by: Hugo van Kemenade <1324225+hugovk@users.noreply.github.com> Co-authored-by: Victor Stinner <vstinner@python.org>
1 parent 732d1b0 commit 4e294f6

File tree

11 files changed

+58
-37
lines changed

11 files changed

+58
-37
lines changed

‎Doc/deprecations/pending-removal-in-future.rst

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -47,6 +47,8 @@ although there is currently no date scheduled for their removal.
4747
:data:`calendar.FEBRUARY`.
4848
(Contributed by Prince Roshan in :gh:`103636`.)
4949

50+
* :mod:`codecs`: use :func:`open` instead of :func:`codecs.open`. (:gh:`133038`)
51+
5052
* :attr:`codeobject.co_lnotab`: use the :meth:`codeobject.co_lines` method
5153
instead.
5254

‎Doc/library/codecs.rst

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -208,6 +208,10 @@ wider range of codecs when working with binary files:
208208
.. versionchanged:: 3.11
209209
The ``'U'`` mode has been removed.
210210

211+
.. deprecated:: next
212+
213+
:func:`codecs.open` has been superseded by :func:`open`.
214+
211215

212216
.. function:: EncodedFile(file, data_encoding, file_encoding=None, errors='strict')
213217

‎Doc/whatsnew/3.14.rst

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1597,6 +1597,10 @@ Deprecated
15971597
as a single positional argument.
15981598
(Contributed by Serhiy Storchaka in :gh:`109218`.)
15991599

1600+
* :mod:`codecs`:
1601+
:func:`codecs.open` is now deprecated. Use :func:`open` instead.
1602+
(Contributed by Inada Naoki in :gh:`133036`.)
1603+
16001604
* :mod:`functools`:
16011605
Calling the Python implementation of :func:`functools.reduce` with *function*
16021606
or *sequence* as keyword arguments is now deprecated.

‎Lib/_pyio.py

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -2056,8 +2056,7 @@ def __init__(self, buffer, encoding=None, errors=None, newline=None,
20562056
raise ValueError("invalid encoding: %r" % encoding)
20572057

20582058
if not codecs.lookup(encoding)._is_text_encoding:
2059-
msg = ("%r is not a text encoding; "
2060-
"use codecs.open() to handle arbitrary codecs")
2059+
msg = "%r is not a text encoding"
20612060
raise LookupError(msg % encoding)
20622061

20632062
if errors is None:

‎Lib/codecs.py

Lines changed: 4 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -884,7 +884,6 @@ def __reduce_ex__(self, proto):
884884
### Shortcuts
885885

886886
def open(filename, mode='r', encoding=None, errors='strict', buffering=-1):
887-
888887
""" Open an encoded file using the given mode and return
889888
a wrapped version providing transparent encoding/decoding.
890889
@@ -912,8 +911,11 @@ def open(filename, mode='r', encoding=None, errors='strict', buffering=-1):
912911
.encoding which allows querying the used encoding. This
913912
attribute is only available if an encoding was specified as
914913
parameter.
915-
916914
"""
915+
import warnings
916+
warnings.warn("codecs.open() is deprecated. Use open() instead.",
917+
DeprecationWarning, stacklevel=2)
918+
917919
if encoding is not None and \
918920
'b' not in mode:
919921
# Force opening of the file in binary mode

‎Lib/test/test_codecs.py

Lines changed: 21 additions & 21 deletions
Original file line numberDiff line numberDiff line change
@@ -7,6 +7,7 @@
77
import unittest
88
import encodings
99
from unittest import mock
10+
import warnings
1011

1112
from test import support
1213
from test.support import os_helper
@@ -20,13 +21,12 @@
2021
except ImportError:
2122
_testinternalcapi = None
2223

23-
try:
24-
import ctypes
25-
except ImportError:
26-
ctypes = None
27-
SIZEOF_WCHAR_T = -1
28-
else:
29-
SIZEOF_WCHAR_T = ctypes.sizeof(ctypes.c_wchar)
24+
25+
def codecs_open_no_warn(*args, **kwargs):
26+
"""Call codecs.open(*args, **kwargs) ignoring DeprecationWarning."""
27+
with warnings.catch_warnings():
28+
warnings.simplefilter("ignore")
29+
return codecs.open(*args, **kwargs)
3030

3131
def coding_checker(self, coder):
3232
def check(input, expect):
@@ -35,13 +35,13 @@ def check(input, expect):
3535

3636
# On small versions of Windows like Windows IoT or Windows Nano Server not all codepages are present
3737
def is_code_page_present(cp):
38-
from ctypes import POINTER, WINFUNCTYPE, WinDLL
38+
from ctypes import POINTER, WINFUNCTYPE, WinDLL, Structure
3939
from ctypes.wintypes import BOOL, BYTE, WCHAR, UINT, DWORD
4040

4141
MAX_LEADBYTES = 12 # 5 ranges, 2 bytes ea., 0 term.
4242
MAX_DEFAULTCHAR = 2 # single or double byte
4343
MAX_PATH = 260
44-
class CPINFOEXW(ctypes.Structure):
44+
class CPINFOEXW(Structure):
4545
_fields_ = [("MaxCharSize", UINT),
4646
("DefaultChar", BYTE*MAX_DEFAULTCHAR),
4747
("LeadByte", BYTE*MAX_LEADBYTES),
@@ -719,19 +719,19 @@ def test_bug691291(self):
719719
self.addCleanup(os_helper.unlink, os_helper.TESTFN)
720720
with open(os_helper.TESTFN, 'wb') as fp:
721721
fp.write(s)
722-
with codecs.open(os_helper.TESTFN, 'r',
722+
with codecs_open_no_warn(os_helper.TESTFN, 'r',
723723
encoding=self.encoding) as reader:
724724
self.assertEqual(reader.read(), s1)
725725

726726
def test_invalid_modes(self):
727727
for mode in ('U', 'rU', 'r+U'):
728728
with self.assertRaises(ValueError) as cm:
729-
codecs.open(os_helper.TESTFN, mode, encoding=self.encoding)
729+
codecs_open_no_warn(os_helper.TESTFN, mode, encoding=self.encoding)
730730
self.assertIn('invalid mode', str(cm.exception))
731731

732732
for mode in ('rt', 'wt', 'at', 'r+t'):
733733
with self.assertRaises(ValueError) as cm:
734-
codecs.open(os_helper.TESTFN, mode, encoding=self.encoding)
734+
codecs_open_no_warn(os_helper.TESTFN, mode, encoding=self.encoding)
735735
self.assertIn("can't have text and binary mode at once",
736736
str(cm.exception))
737737

@@ -1844,9 +1844,9 @@ def test_all(self):
18441844
def test_open(self):
18451845
self.addCleanup(os_helper.unlink, os_helper.TESTFN)
18461846
for mode in ('w', 'r', 'r+', 'w+', 'a', 'a+'):
1847-
with self.subTest(mode), \
1848-
codecs.open(os_helper.TESTFN, mode, 'ascii') as file:
1849-
self.assertIsInstance(file, codecs.StreamReaderWriter)
1847+
with self.subTest(mode), self.assertWarns(DeprecationWarning):
1848+
with codecs.open(os_helper.TESTFN, mode, 'ascii') as file:
1849+
self.assertIsInstance(file, codecs.StreamReaderWriter)
18501850

18511851
def test_undefined(self):
18521852
self.assertRaises(UnicodeError, codecs.encode, 'abc', 'undefined')
@@ -1863,7 +1863,7 @@ def test_file_closes_if_lookup_error_raised(self):
18631863
mock_open = mock.mock_open()
18641864
with mock.patch('builtins.open', mock_open) as file:
18651865
with self.assertRaises(LookupError):
1866-
codecs.open(os_helper.TESTFN, 'wt', 'invalid-encoding')
1866+
codecs_open_no_warn(os_helper.TESTFN, 'wt', 'invalid-encoding')
18671867

18681868
file().close.assert_called()
18691869

@@ -2883,7 +2883,7 @@ def test_seek0(self):
28832883
self.addCleanup(os_helper.unlink, os_helper.TESTFN)
28842884
for encoding in tests:
28852885
# Check if the BOM is written only once
2886-
with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
2886+
with codecs_open_no_warn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
28872887
f.write(data)
28882888
f.write(data)
28892889
f.seek(0)
@@ -2892,7 +2892,7 @@ def test_seek0(self):
28922892
self.assertEqual(f.read(), data * 2)
28932893

28942894
# Check that the BOM is written after a seek(0)
2895-
with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
2895+
with codecs_open_no_warn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
28962896
f.write(data[0])
28972897
self.assertNotEqual(f.tell(), 0)
28982898
f.seek(0)
@@ -2901,7 +2901,7 @@ def test_seek0(self):
29012901
self.assertEqual(f.read(), data)
29022902

29032903
# (StreamWriter) Check that the BOM is written after a seek(0)
2904-
with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
2904+
with codecs_open_no_warn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
29052905
f.writer.write(data[0])
29062906
self.assertNotEqual(f.writer.tell(), 0)
29072907
f.writer.seek(0)
@@ -2911,7 +2911,7 @@ def test_seek0(self):
29112911

29122912
# Check that the BOM is not written after a seek() at a position
29132913
# different than the start
2914-
with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
2914+
with codecs_open_no_warn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
29152915
f.write(data)
29162916
f.seek(f.tell())
29172917
f.write(data)
@@ -2920,7 +2920,7 @@ def test_seek0(self):
29202920

29212921
# (StreamWriter) Check that the BOM is not written after a seek()
29222922
# at a position different than the start
2923-
with codecs.open(os_helper.TESTFN, 'w+', encoding=encoding) as f:
2923+
with codecs_open_no_warn(os_helper.TESTFN, 'w+', encoding=encoding) as f:
29242924
f.writer.write(data)
29252925
f.writer.seek(f.writer.tell())
29262926
f.writer.write(data)

‎Lib/test/test_multibytecodec.py

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -314,7 +314,8 @@ def test_bug1728403(self):
314314
f.write(b'\xa1')
315315
finally:
316316
f.close()
317-
f = codecs.open(TESTFN, encoding='cp949')
317+
with self.assertWarns(DeprecationWarning):
318+
f = codecs.open(TESTFN, encoding='cp949')
318319
try:
319320
self.assertRaises(UnicodeDecodeError, f.read, 2)
320321
finally:

‎Lib/test/test_sax.py

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,4 @@
11
# regression test for SAX 2.0
2-
# $Id$
32

43
from xml.sax import make_parser, ContentHandler, \
54
SAXException, SAXReaderNotAvailable, SAXParseException
@@ -832,8 +831,9 @@ class StreamReaderWriterXmlgenTest(XmlgenTest, unittest.TestCase):
832831
fname = os_helper.TESTFN + '-codecs'
833832

834833
def ioclass(self):
835-
writer = codecs.open(self.fname, 'w', encoding='ascii',
836-
errors='xmlcharrefreplace', buffering=0)
834+
with self.assertWarns(DeprecationWarning):
835+
writer = codecs.open(self.fname, 'w', encoding='ascii',
836+
errors='xmlcharrefreplace', buffering=0)
837837
def cleanup():
838838
writer.close()
839839
os_helper.unlink(self.fname)

‎Misc/NEWS.d/next/Library/2025-04-27-15-21-05.gh-issue-133036.HCNYA7.rst

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,2 @@
1+
:func:`codecs.open` is now deprecated. Use :func:`open` instead. Contributed
2+
by Inada Naoki.

‎Modules/_io/textio.c

Lines changed: 2 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1185,7 +1185,7 @@ _io_TextIOWrapper___init___impl(textio *self, PyObject *buffer,
11851185
}
11861186

11871187
/* Check we have been asked for a real text encoding */
1188-
codec_info = _PyCodec_LookupTextEncoding(encoding, "codecs.open()");
1188+
codec_info = _PyCodec_LookupTextEncoding(encoding, NULL);
11891189
if (codec_info == NULL) {
11901190
Py_CLEAR(self->encoding);
11911191
goto error;
@@ -1324,8 +1324,7 @@ textiowrapper_change_encoding(textio *self, PyObject *encoding,
13241324
}
13251325

13261326
// Create new encoder & decoder
1327-
PyObject *codec_info = _PyCodec_LookupTextEncoding(
1328-
c_encoding, "codecs.open()");
1327+
PyObject *codec_info = _PyCodec_LookupTextEncoding(c_encoding, NULL);
13291328
if (codec_info == NULL) {
13301329
Py_DECREF(encoding);
13311330
Py_DECREF(errors);

‎Python/codecs.c

Lines changed: 13 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -540,11 +540,19 @@ PyObject * _PyCodec_LookupTextEncoding(const char *encoding,
540540
Py_DECREF(attr);
541541
if (is_text_codec <= 0) {
542542
Py_DECREF(codec);
543-
if (!is_text_codec)
544-
PyErr_Format(PyExc_LookupError,
545-
"'%.400s' is not a text encoding; "
546-
"use %s to handle arbitrary codecs",
547-
encoding, alternate_command);
543+
if (!is_text_codec) {
544+
if (alternate_command != NULL) {
545+
PyErr_Format(PyExc_LookupError,
546+
"'%.400s' is not a text encoding; "
547+
"use %s to handle arbitrary codecs",
548+
encoding, alternate_command);
549+
}
550+
else {
551+
PyErr_Format(PyExc_LookupError,
552+
"'%.400s' is not a text encoding",
553+
encoding);
554+
}
555+
}
548556
return NULL;
549557
}
550558
}

0 commit comments

Comments
 (0)
0