8000 gh-95781: More strict format string checking in PyUnicode_FromFormatV… · python/cpython@62f0650 · GitHub
[go: up one dir, main page]
Skip to content

Commit 62f0650

Browse files
serhiy-storchakaserhiy-storchaka
authored
gh-95781: More strict format string checking in PyUnicode_FromFormatV() (GH-95784)
An unrecognized format character in PyUnicode_FromFormat() and PyUnicode_FromFormatV() now sets a SystemError. In previous versions it caused all the rest of the format string to be copied as-is to the result string, and any extra arguments discarded.
1 parent 63140b4 commit 62f0650

File tree

5 files changed

+35
-39
lines changed

5 files changed

+35
-39
lines changed

‎Doc/c-api/unicode.rst

Lines changed: 5 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -477,9 +477,6 @@ APIs:
477477
| | | :c:func:`PyObject_Repr`. |
478478
+-------------------+---------------------+----------------------------------+
479479
480-
An unrecognized format character causes all the rest of the format string to be
481-
copied as-is to the result string, and any extra arguments discarded.
482-
483480
.. note::
484481
The width formatter unit is number of characters rather than bytes.
485482
The precision formatter unit is number of bytes for ``"%s"`` and
@@ -500,6 +497,11 @@ APIs:
500497
Support width and precision formatter for ``"%s"``, ``"%A"``, ``"%U"``,
501498
``"%V"``, ``"%S"``, ``"%R"`` added.
502499
500+
.. versionchanged:: 3.12
501+
An unrecognized format character now sets a :exc:`SystemError`.
502+
In previous versions it caused all the rest of the format string to be
503+
copied as-is to the result string, and any extra arguments discarded.
504+
503505
504506
.. c:function:: PyObject* PyUnicode_FromFormatV(const char *format, va_list vargs)
505507

‎Doc/whatsnew/3.12.rst

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -469,6 +469,12 @@ Porting to Python 3.12
469469
:py:meth:`~class.__subclasses__` (using :c:func:`PyObject_CallMethod`,
470470
for example).
471471

472+
* An unrecognized format character in :c:func:`PyUnicode_FromFormat` and
473+
:c:func:`PyUnicode_FromFormatV` now sets a :exc:`SystemError`.
474+
In previous versions it caused all the rest of the format string to be
475+
copied as-is to the result string, and any extra arguments discarded.
476+
(Contributed by Serhiy Storchaka in :gh:`95781`.)
477+
472478

473479
Deprecated
474480
----------

‎Lib/test/test_unicode.py

Lines changed: 10 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -2641,8 +2641,6 @@ def check_format(expected, format, *args):
26412641
b'%c%c', c_int(0x10000), c_int(0x100000))
26422642

26432643
# test "%"
2644-
check_format('%',
2645-
b'%')
26462644
check_format('%',
26472645
b'%%')
26482646
check_format('%s',
@@ -2819,23 +2817,22 @@ def check_format(expected, format, *args):
28192817
check_format('repr=abc\ufffd',
28202818
b'repr=%V', None, b'abc\xff')
28212819

2822-
# not supported: copy the raw format string. these tests are just here
2823-
# to check for crashes and should not be considered as specifications
2824-
check_format('%s',
2825-
b'%1%s', b'abc')
2826-
check_format('%1abc',
2827-
b'%1abc')
2828-
check_format('%+i',
2829-
b'%+i', c_int(10))
2830-
check_format('%.%s',
2831-
b'%.%s', b'abc')
2832-
28332820
# Issue #33817: empty strings
28342821
check_format('',
28352822
b'')
28362823
check_format('',
28372824
b'%s', b'')
28382825

2826+
# check for crashes
2827+
for fmt in (b'%', b'%0', b'%01', b'%.', b'%.1',
2828+
b'%0%s', b'%1%s', b'%.%s', b'%.1%s', b'%1abc',
2829+
b'%l', b'%ll', b'%z', b'%ls', b'%lls', b'%zs'):
2830+
with self.subTest(fmt=fmt):
2831+
self.assertRaisesRegex(SystemError, 'invalid format string',
2832+
PyUnicode_FromFormat, fmt, b'abc')
2833+
self.assertRaisesRegex(SystemError, 'invalid format string',
2834+
PyUnicode_FromFormat, b'%+i', c_int(10))
2835+
28392836
# Test PyUnicode_AsWideChar()
28402837
@support.cpython_only
28412838
@unittest.skipIf(_testcapi is None, 'need _testcapi module')

‎Misc/NEWS.d/next/C API/2022-08-08-14-36-31.gh-issue-95781.W_G8YW.rst

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,4 @@
1+
An unrecognized format character in :c:func:`PyUnicode_FromFormat` and
2+
:c:func:`PyUnicode_FromFormatV` now sets a :exc:`SystemError`.
3+
In previous versions it caused all the rest of the format string to be
4+
copied as-is to the result string, and any extra arguments discarded.

‎Objects/unicodeobject.c

Lines changed: 10 additions & 23 deletions
Original file line numberDiff line numberDiff line change
@@ -2355,6 +2355,13 @@ unicode_fromformat_arg(_PyUnicodeWriter *writer,
23552355

23562356
p = f;
23572357
f++;
2358+
if (*f == '%') {
2359+
if (_PyUnicodeWriter_WriteCharInline(writer, '%') < 0)
2360+
return NULL;
2361+
f++;
2362+
return f;
2363+
}
2364+
23582365
zeropad = 0;
23592366
if (*f == '0') {
23602367
zeropad = 1;
@@ -2392,14 +2399,6 @@ unicode_fromformat_arg(_PyUnicodeWriter *writer,
23922399
f++;
23932400
}
23942401
}
2395-
if (*f == '%') {
2396-
/* "%.3%s" => f points to "3" */
2397-
f--;
2398-
}
2399-
}
2400-
if (*f == '\0') {
2401-
/* bogus format "%.123" => go backward, f points to "3" */
2402-
f--;
24032402
}
24042403

24052404
/* Handle %ld, %lu, %lld and %llu. */
@@ -2423,7 +2422,7 @@ unicode_fromformat_arg(_PyUnicodeWriter *writer,
24232422
++f;
24242423
}
24252424

2426-
if (f[1] == '\0')
2425+
if (f[0] != '\0' && f[1] == '\0')
24272426
writer->overallocate = 0;
24282427

24292428
switch (*f) {
@@ -2616,21 +2615,9 @@ unicode_fromformat_arg(_PyUnicodeWriter *writer,
26162615
break;
26172616
}
26182617

2619-
case '%':
2620-
if (_PyUnicodeWriter_WriteCharInline(writer, '%') < 0)
2621-
return NULL;
2622-
break;
2623-
26242618
default:
2625-
/* if we stumble upon an unknown formatting code, copy the rest
2626-
of the format string to the output string. (we cannot just
2627-
skip the code, since there's no way to know what's in the
2628-
argument list) */
2629-
len = strlen(p);
2630-
if (_PyUnicodeWriter_WriteLatin1String(writer, p, len) == -1)
2631-
return NULL;
2632-
f = p+len;
2633-
return f;
2619+
PyErr_Format(PyExc_SystemError, "invalid format string: %s", p);
2620+
return NULL;
26342621
}
26352622

26362623
f++;

0 commit comments

Comments
 (0)
0