Skip to content

Commit ba14dfa

Browse files
picnixzpicnixz
authored
gh-123378: fix a crash in UnicodeError.__str__ (#124935)
1 parent 19984fe commit ba14dfa

File tree

3 files changed

+93
-45
lines changed

3 files changed

+93
-45
lines changed

‎Lib/test/test_exceptions.py‎

Lines changed: 24 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -8,6 +8,7 @@
88
import weakref
99
import errno
1010
from codecs import BOM_UTF8
11+
from itertools import product
1112
from textwrap import dedent
1213

1314
from test.support import (captured_stderr, check_impl_detail,
@@ -1336,6 +1337,29 @@ def test_unicode_errors_no_object(self):
13361337
for klass in klasses:
13371338
self.assertEqual(str(klass.__new__(klass)), "")
13381339

1340+
def test_unicode_error_str_does_not_crash(self):
1341+
# Test that str(UnicodeError(...)) does not crash.
1342+
# See https://github.com/python/cpython/issues/123378.
1343+
1344+
for start, end, objlen in product(
1345+
range(-5, 5),
1346+
range(-5, 5),
1347+
range(7),
1348+
):
1349+
obj = 'a' * objlen
1350+
with self.subTest('encode', objlen=objlen, start=start, end=end):
1351+
exc = UnicodeEncodeError('utf-8', obj, start, end, '')
1352+
self.assertIsInstance(str(exc), str)
1353+
1354+
with self.subTest('translate', objlen=objlen, start=start, end=end):
1355+
exc = UnicodeTranslateError(obj, start, end, '')
1356+
self.assertIsInstance(str(exc), str)
1357+
1358+
encoded = obj.encode()
1359+
with self.subTest('decode', objlen=objlen, start=start, end=end):
1360+
exc = UnicodeDecodeError('utf-8', encoded, start, end, '')
1361+
self.assertIsInstance(str(exc), str)
1362+
13391363
@no_tracing
13401364
def test_badisinstance(self):
13411365
# Bug #2542: if issubclass(e, MyException) raises an exception,

‎Misc/NEWS.d/next/Core_and_Builtins/2024-10-03-14-39-41.gh-issue-123378.dCxANf.rst‎

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,3 @@
1+
Fix a crash in the :meth:`~object.__str__` method of :exc:`UnicodeError`
2+
objects when the :attr:`UnicodeError.start` and :attr:`UnicodeError.end`
3+
values are invalid or out-of-range. Patch by Bénédikt Tran.

‎Objects/exceptions.c‎

Lines changed: 66 additions & 45 deletions
Original file line numberDiff line numberDiff line change
@@ -2994,46 +2994,55 @@ UnicodeEncodeError_init(PyObject *self, PyObject *args, PyObject *kwds)
29942994
static PyObject *
29952995
UnicodeEncodeError_str(PyObject *self)
29962996
{
2997-
PyUnicodeErrorObject *uself = (PyUnicodeErrorObject *)self;
2997+
PyUnicodeErrorObject *exc = (PyUnicodeErrorObject *)self;
29982998
PyObject *result = NULL;
29992999
PyObject *reason_str = NULL;
30003000
PyObject *encoding_str = NULL;
30013001

3002-
if (!uself->object)
3002+
if (exc->object == NULL) {
30033003
/* Not properly initialized. */
30043004
return PyUnicode_FromString("");
3005+
}
30053006

30063007
/* Get reason and encoding as strings, which they might not be if
30073008
they've been modified after we were constructed. */
3008-
reason_str = PyObject_Str(uself->reason);
3009-
if (reason_str == NULL)
3009+
reason_str = PyObject_Str(exc->reason);
3010+
if (reason_str == NULL) {
30103011
goto done;
3011-
encoding_str = PyObject_Str(uself->encoding);
3012-
if (encoding_str == NULL)
3012+
}
3013+
encoding_str = PyObject_Str(exc->encoding);
3014+
if (encoding_str == NULL) {
30133015
goto done;
3016+
}
3017+
3018+
Py_ssize_t len = PyUnicode_GET_LENGTH(exc->object);
3019+
Py_ssize_t start = exc->start, end = exc->end;
30143020

3015-
if (uself->start < PyUnicode_GET_LENGTH(uself->object) && uself->end == uself->start+1) {
3016-
Py_UCS4 badchar = PyUnicode_ReadChar(uself->object, uself->start);
3021+
if ((start >= 0 && start < len) && (end >= 0 && end <= len) && end == start + 1) {
3022+
Py_UCS4 badchar = PyUnicode_ReadChar(exc->object, start);
30173023
const char *fmt;
3018-
if (badchar <= 0xff)
3024+
if (badchar <= 0xff) {
30193025
fmt = "'%U' codec can't encode character '\\x%02x' in position %zd: %U";
3020-
else if (badchar <= 0xffff)
3026+
}
3027+
else if (badchar <= 0xffff) {
30213028
fmt = "'%U' codec can't encode character '\\u%04x' in position %zd: %U";
3022-
else
3029+
}
3030+
else {
30233031
fmt = "'%U' codec can't encode character '\\U%08x' in position %zd: %U";
3032+
}
30243033
result = PyUnicode_FromFormat(
30253034
fmt,
30263035
encoding_str,
30273036
(int)badchar,
3028-
uself->start,
3037+
start,
30293038
reason_str);
30303039
}
30313040
else {
30323041
result = PyUnicode_FromFormat(
30333042
"'%U' codec can't encode characters in position %zd-%zd: %U",
30343043
encoding_str,
3035-
uself->start,
3036-
uself->end-1,
3044+
start,
3045+
end - 1,
30373046
reason_str);
30383047
}
30393048
done:
@@ -3107,41 +3116,46 @@ UnicodeDecodeError_init(PyObject *self, PyObject *args, PyObject *kwds)
31073116
static PyObject *
31083117
UnicodeDecodeError_str(PyObject *self)
31093118
{
3110-
PyUnicodeErrorObject *uself = (PyUnicodeErrorObject *)self;
3119+
PyUnicodeErrorObject *exc = (PyUnicodeErrorObject *)self;
31113120
PyObject *result = NULL;
31123121
PyObject *reason_str = NULL;
31133122
PyObject *encoding_str = NULL;
31143123

3115-
if (!uself->object)
3124+
if (exc->object == NULL) {
31163125
/* Not properly initialized. */
31173126
return PyUnicode_FromString("");
3127+
}
31183128

31193129
/* Get reason and encoding as strings, which they might not be if
31203130
they've been modified after we were constructed. */
3121-
reason_str = PyObject_Str(uself->reason);
3122-
if (reason_str == NULL)
3131+
reason_str = PyObject_Str(exc->reason);
3132+
if (reason_str == NULL) {
31233133
goto done;
3124-
encoding_str = PyObject_Str(uself->encoding);
3125-
if (encoding_str == NULL)
3134+
}
3135+
encoding_str = PyObject_Str(exc->encoding);
3136+
if (encoding_str == NULL) {
31263137
goto done;
3138+
}
3139+
3140+
Py_ssize_t len = PyBytes_GET_SIZE(exc->object);
3141+
Py_ssize_t start = exc->start, end = exc->end;
31273142

3128-
if (uself->start < PyBytes_GET_SIZE(uself->object) && uself->end == uself->start+1) {
3129-
int byte = (int)(PyBytes_AS_STRING(((PyUnicodeErrorObject *)self)->object)[uself->start]&0xff);
3143+
if ((start >= 0 && start < len) && (end >= 0 && end <= len) && end == start + 1) {
3144+
int badbyte = (int)(PyBytes_AS_STRING(exc->object)[start] & 0xff);
31303145
result = PyUnicode_FromFormat(
31313146
"'%U' codec can't decode byte 0x%02x in position %zd: %U",
31323147
encoding_str,
3133-
byte,
3134-
uself->start,
3148+
badbyte,
3149+
start,
31353150
reason_str);
31363151
}
31373152
else {
31383153
result = PyUnicode_FromFormat(
31393154
"'%U' codec can't decode bytes in position %zd-%zd: %U",
31403155
encoding_str,
3141-
uself->start,
3142-
uself->end-1,
3143-
reason_str
3144-
);
3156+
start,
3157+
end - 1,
3158+
reason_str);
31453159
}
31463160
done:
31473161
Py_XDECREF(reason_str);
@@ -3204,42 +3218,49 @@ UnicodeTranslateError_init(PyUnicodeErrorObject *self, PyObject *args,
32043218
static PyObject *
32053219
UnicodeTranslateError_str(PyObject *self)
32063220
{
3207-
PyUnicodeErrorObject *uself = (PyUnicodeErrorObject *)self;
3221+
PyUnicodeErrorObject *exc = (PyUnicodeErrorObject *)self;
32083222
PyObject *result = NULL;
32093223
PyObject *reason_str = NULL;
32103224

3211-
if (!uself->object)
3225+
if (exc->object == NULL) {
32123226
/* Not properly initialized. */
32133227
return PyUnicode_FromString("");
3228+
}
32143229

32153230
/* Get reason as a string, which it might not be if it's been
32163231
modified after we were constructed. */
3217-
reason_str = PyObject_Str(uself->reason);
3218-
if (reason_str == NULL)
3232+
reason_str = PyObject_Str(exc->reason);
3233+
if (reason_str == NULL) {
32193234
goto done;
3235+
}
3236+
3237+
Py_ssize_t len = PyUnicode_GET_LENGTH(exc->object);
3238+
Py_ssize_t start = exc->start, end = exc->end;
32203239

3221-
if (uself->start < PyUnicode_GET_LENGTH(uself->object) && uself->end == uself->start+1) {
3222-
Py_UCS4 badchar = PyUnicode_ReadChar(uself->object, uself->start);
3240+
if ((start >= 0 && start < len) && (end >= 0 && end <= len) && end == start + 1) {
3241+
Py_UCS4 badchar = PyUnicode_ReadChar(exc->object, start);
32233242
const char *fmt;
3224-
if (badchar <= 0xff)
3243+
if (badchar <= 0xff) {
32253244
fmt = "can't translate character '\\x%02x' in position %zd: %U";
3226-
else if (badchar <= 0xffff)
3245+
}
3246+
else if (badchar <= 0xffff) {
32273247
fmt = "can't translate character '\\u%04x' in position %zd: %U";
3228-
else
3248+
}
3249+
else {
32293250
fmt = "can't translate character '\\U%08x' in position %zd: %U";
3251+
}
32303252
result = PyUnicode_FromFormat(
32313253
fmt,
32323254
(int)badchar,
3233-
uself->start,
3234-
reason_str
3235-
);
3236-
} else {
3255+
start,
3256+
reason_str);
3257+
}
3258+
else {
32373259
result = PyUnicode_FromFormat(
32383260
"can't translate characters in position %zd-%zd: %U",
3239-
uself->start,
3240-
uself->end-1,
3241-
reason_str
3242-
);
3261+
start,
3262+
end - 1,
3263+
reason_str);
32433264
}
32443265
done:
32453266
Py_XDECREF(reason_str);

0 commit comments

Comments
 (0)