Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

Commitb8b3e6a

Browse files
[3.11]gh-99612: Fix PyUnicode_DecodeUTF8Stateful() for ASCII-only data (GH-99613) (GH-107224)
Previously *consumed was not set in this case.(cherry picked from commitf08e52c)
1 parent058741c commitb8b3e6a

File tree

4 files changed

+95
-1
lines changed

4 files changed

+95
-1
lines changed

‎Lib/test/test_capi/test_codecs.py

Lines changed: 54 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,54 @@
1+
importunittest
2+
fromtest.supportimportimport_helper
3+
4+
_testcapi=import_helper.import_module('_testcapi')
5+
6+
7+
classCAPITest(unittest.TestCase):
8+
9+
deftest_decodeutf8(self):
10+
"""Test PyUnicode_DecodeUTF8()"""
11+
decodeutf8=_testcapi.unicode_decodeutf8
12+
13+
forsin ['abc','\xa1\xa2','\u4f60\u597d','a\U0001f600']:
14+
b=s.encode('utf-8')
15+
self.assertEqual(decodeutf8(b),s)
16+
self.assertEqual(decodeutf8(b,'strict'),s)
17+
18+
self.assertRaises(UnicodeDecodeError,decodeutf8,b'\x80')
19+
self.assertRaises(UnicodeDecodeError,decodeutf8,b'\xc0')
20+
self.assertRaises(UnicodeDecodeError,decodeutf8,b'\xff')
21+
self.assertRaises(UnicodeDecodeError,decodeutf8,b'a\xf0\x9f')
22+
self.assertEqual(decodeutf8(b'a\xf0\x9f','replace'),'a\ufffd')
23+
self.assertEqual(decodeutf8(b'a\xf0\x9fb','replace'),'a\ufffdb')
24+
25+
self.assertRaises(LookupError,decodeutf8,b'a\x80','foo')
26+
# TODO: Test PyUnicode_DecodeUTF8() with NULL as data and
27+
# negative size.
28+
29+
deftest_decodeutf8stateful(self):
30+
"""Test PyUnicode_DecodeUTF8Stateful()"""
31+
decodeutf8stateful=_testcapi.unicode_decodeutf8stateful
32+
33+
forsin ['abc','\xa1\xa2','\u4f60\u597d','a\U0001f600']:
34+
b=s.encode('utf-8')
35+
self.assertEqual(decodeutf8stateful(b), (s,len(b)))
36+
self.assertEqual(decodeutf8stateful(b,'strict'), (s,len(b)))
37+
38+
self.assertRaises(UnicodeDecodeError,decodeutf8stateful,b'\x80')
39+
self.assertRaises(UnicodeDecodeError,decodeutf8stateful,b'\xc0')
40+
self.assertRaises(UnicodeDecodeError,decodeutf8stateful,b'\xff')
41+
self.assertEqual(decodeutf8stateful(b'a\xf0\x9f'), ('a',1))
42+
self.assertEqual(decodeutf8stateful(b'a\xf0\x9f','replace'), ('a',1))
43+
self.assertRaises(UnicodeDecodeError,decodeutf8stateful,b'a\xf0\x9fb')
44+
self.assertEqual(decodeutf8stateful(b'a\xf0\x9fb','replace'), ('a\ufffdb',4))
45+
46+
self.assertRaises(LookupError,decodeutf8stateful,b'a\x80','foo')
47+
# TODO: Test PyUnicode_DecodeUTF8Stateful() with NULL as data and
48+
# negative size.
49+
# TODO: Test PyUnicode_DecodeUTF8Stateful() with NULL as the address of
50+
# "consumed".
51+
52+
53+
if__name__=="__main__":
54+
unittest.main()
Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,2 @@
1+
Fix:c:func:`PyUnicode_DecodeUTF8Stateful` for ASCII-only data:
2+
``*consumed`` was not set.

‎Modules/_testcapimodule.c

Lines changed: 36 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -2307,6 +2307,40 @@ unicode_asutf8andsize(PyObject *self, PyObject *args)
23072307
returnPy_BuildValue("(Nn)",result,utf8_len);
23082308
}
23092309

2310+
/* Test PyUnicode_DecodeUTF8() */
2311+
staticPyObject*
2312+
unicode_decodeutf8(PyObject*self,PyObject*args)
2313+
{
2314+
constchar*data;
2315+
Py_ssize_tsize;
2316+
constchar*errors=NULL;
2317+
2318+
if (!PyArg_ParseTuple(args,"y#|z",&data,&size,&errors))
2319+
returnNULL;
2320+
2321+
returnPyUnicode_DecodeUTF8(data,size,errors);
2322+
}
2323+
2324+
/* Test PyUnicode_DecodeUTF8Stateful() */
2325+
staticPyObject*
2326+
unicode_decodeutf8stateful(PyObject*self,PyObject*args)
2327+
{
2328+
constchar*data;
2329+
Py_ssize_tsize;
2330+
constchar*errors=NULL;
2331+
Py_ssize_tconsumed=123456789;
2332+
PyObject*result;
2333+
2334+
if (!PyArg_ParseTuple(args,"y#|z",&data,&size,&errors))
2335+
returnNULL;
2336+
2337+
result=PyUnicode_DecodeUTF8Stateful(data,size,errors,&consumed);
2338+
if (!result) {
2339+
returnNULL;
2340+
}
2341+
returnPy_BuildValue("(Nn)",result,consumed);
2342+
}
2343+
23102344
staticPyObject*
23112345
unicode_findchar(PyObject*self,PyObject*args)
23122346
{
@@ -6562,7 +6596,8 @@ static PyMethodDef TestMethods[] = {
65626596
{"unicode_asucs4",unicode_asucs4,METH_VARARGS},
65636597
{"unicode_asutf8",unicode_asutf8,METH_VARARGS},
65646598
{"unicode_asutf8andsize",unicode_asutf8andsize,METH_VARARGS},
6565-
{"unicode_findchar",unicode_findchar,METH_VARARGS},
6599+
{"unicode_decodeutf8",unicode_decodeutf8,METH_VARARGS},
6600+
{"unicode_decodeutf8stateful",unicode_decodeutf8stateful,METH_VARARGS}, {"unicode_findchar",unicode_findchar,METH_VARARGS},
65666601
{"unicode_copycharacters",unicode_copycharacters,METH_VARARGS},
65676602
#ifUSE_UNICODE_WCHAR_CACHE
65686603
{"unicode_legacy_string",unicode_legacy_string,METH_VARARGS},

‎Objects/unicodeobject.c

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -5120,6 +5120,9 @@ unicode_decode_utf8(const char *s, Py_ssize_t size,
51205120
}
51215121
s+=ascii_decode(s,end,PyUnicode_1BYTE_DATA(u));
51225122
if (s==end) {
5123+
if (consumed) {
5124+
*consumed=size;
5125+
}
51235126
returnu;
51245127
}
51255128

0 commit comments

Comments
 (0)
[8]ページ先頭
©2009-2025 Movatter.jp