Skip to content

gh-119182: Decode PyUnicode_FromFormat() format string from UTF-8 #120248

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
vstinner wants to merge 7 commits into from
Closed

gh-119182: Decode PyUnicode_FromFormat() format string from UTF-8 #120248

Show file tree
Hide file tree
Changes from 6 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
3d5bca4
gh-119182: Decode PyUnicode_FromFormat() format from UTF-8
vstinner Jun 7, 2024
6a87915
Update test_exceptions
vstinner Jun 7, 2024
e830944
Use strict error handler
vstinner Jun 11, 2024
242e6cb
Fix error handling
vstinner Jun 11, 2024
d04269f
Add tests on truncated UTF-8 format strings
vstinner Jun 11, 2024
94da5e7
Don't mention the strict error handler
vstinner Jun 11, 2024
89fd69a
Revert consumed parameter
vstinner Jun 11, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions Doc/c-api/unicode.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -387,7 +387,7 @@ APIs:
arguments, calculate the size of the resulting Python Unicode string and return
a string with the values formatted into it. The variable arguments must be C
types and must correspond exactly to the format characters in the *format*
ASCII-encoded string.
string. The *format* string is decoded from UTF-8.

A conversion specifier contains two or more characters and has the following
components, which must occur in this order:
Expand Down Expand Up @@ -487,7 +487,8 @@ APIs:

* - ``s``
- :c:expr:`const char*` or :c:expr:`const wchar_t*`
- A null-terminated C character array.
- A null-terminated C character array. :c:expr:`const char*` is decoded
from UTF-8 with the "replace" error handler.

* - ``p``
- :c:expr:`const void*`
Expand Down Expand Up @@ -576,6 +577,9 @@ APIs:
.. versionchanged:: 3.13
Support for ``%T``, ``%#T``, ``%N`` and ``%#N`` formats added.

.. versionchanged:: 3.14
The format string is now decoded from UTF-8 instead of ASCII.


.. c:function:: PyObject* PyUnicode_FromFormatV(const char *format, va_list vargs)

Expand Down
4 changes: 4 additions & 0 deletions Doc/whatsnew/3.14.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -261,6 +261,10 @@ New Features
Porting to Python 3.14
----------------------

* :c:func:`PyUnicode_FromFormat` now decodes the format string from UTF-8,
instead of ASCII.
(Contributed by Victor Stinner in :gh:`119182`.)

Deprecated
----------

Expand Down
4 changes: 3 additions & 1 deletion Lib/test/test_capi/test_exceptions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -273,8 +273,10 @@ def test_format(self):

with self.assertRaisesRegex(OverflowError, 'not in range'):
PyErr_Format(ZeroDivisionError, b'%c', c_int(-1))
with self.assertRaisesRegex(ValueError, 'format string'):
with self.assertRaisesRegex(ValueError, 'format string') as cm:
PyErr_Format(ZeroDivisionError, b'\xff')
self.assertIsInstance(cm.exception.__context__, UnicodeDecodeError)

self.assertRaises(SystemError, PyErr_Format, list, b'error')
# CRASHES PyErr_Format(ZeroDivisionError, NULL)
# CRASHES PyErr_Format(py_object(), b'error')
Expand Down
21 changes: 14 additions & 7 deletions Lib/test/test_capi/test_unicode.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -380,16 +380,23 @@ def check_format(expected, format, *args):
text = PyUnicode_FromFormat(format, *args)
self.assertEqual(expected, text)

# ascii format, non-ascii argument
# ASCII format, non-ASCII %U argument
check_format('ascii\x7f=unicode\xe9',
b'ascii\x7f=%U', 'unicode\xe9')

# non-ascii format, ascii argument: ensure that PyUnicode_FromFormatV()
# raises an error
self.assertRaisesRegex(ValueError,
r'^PyUnicode_FromFormatV\(\) expects an ASCII-encoded format '
'string, got a non-ASCII byte: 0xe9$',
PyUnicode_FromFormat, b'unicode\xe9=%s', 'ascii')
# The %s arguments are decoded from UTF-8/replace.
# The format string is decoded from UTF-8/strict.
check_format('value=utf8 \u20ac',
'value=%s'.encode(), 'utf8 \u20ac'.encode())
with self.assertRaisesRegex(ValueError, 'format string') as cm:
PyUnicode_FromFormat(b'invalid format string\xff: %s', b'abc')
Comment thread
vstinner marked this conversation as resolved.
self.assertIsInstance(cm.exception.__context__, UnicodeDecodeError)

# Truncated UTF-8 format strings
with self.assertRaisesRegex(ValueError, 'format string'):
PyUnicode_FromFormat(b'truncated utf8: \xc3')
with self.assertRaisesRegex(ValueError, 'format string'):
PyUnicode_FromFormat(b'truncated utf8: \xe2\x82')

# test "%c"
check_format('\uabcd',
Expand Down
2 changes: 2 additions & 0 deletions Misc/NEWS.d/next/C API/2024-06-07-22-38-08.gh-issue-119182.P3nXBm.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,2 @@
:c:func:`PyUnicode_FromFormat` now decodes the format string from UTF-8,
instead of ASCII. Patch by Victor Stinner.
48 changes: 19 additions & 29 deletions Objects/unicodeobject.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -205,8 +205,7 @@ unicode_decode_utf8(const char *s, Py_ssize_t size,
static int
unicode_decode_utf8_writer(_PyUnicodeWriter *writer,
const char *s, Py_ssize_t size,
_Py_error_handler error_handler, const char *errors,
Py_ssize_t *consumed);
Comment thread
vstinner marked this conversation as resolved.
_Py_error_handler error_handler, const char *errors);
#ifdef Py_DEBUG
static inline int unicode_is_finalizing(void);
static int unicode_is_singleton(PyObject *unicode);
Expand Down Expand Up @@ -2402,7 +2401,7 @@ unicode_fromformat_write_utf8(_PyUnicodeWriter *writer, const char *str,

if (width < 0) {
return unicode_decode_utf8_writer(writer, str, length,
_Py_ERROR_REPLACE, "replace", NULL);
_Py_ERROR_REPLACE, "replace");
}

PyObject *unicode = PyUnicode_DecodeUTF8Stateful(str, length,
Expand Down Expand Up @@ -2896,28 +2895,26 @@ PyUnicode_FromFormatV(const char *format, va_list vargs)
const char *p;
Py_ssize_t len;

p = f;
do
{
if ((unsigned char)*p > 127) {
PyErr_Format(PyExc_ValueError,
"PyUnicode_FromFormatV() expects an ASCII-encoded format "
"string, got a non-ASCII byte: 0x%02x",
(unsigned char)*p);
goto fail;
}
p++;
p = strchr(f, '%');
if (p != NULL) {
len = p - f;
}
while (*p != '\0' && *p != '%');
len = p - f;

if (*p == '\0')
else {
len = strlen(f);
writer.overallocate = 0;
}

if (_PyUnicodeWriter_WriteASCIIString(&writer, f, len) < 0)
if (unicode_decode_utf8_writer(&writer, f, len,
_Py_ERROR_STRICT, "strict") < 0) {
PyObject *exc = PyErr_GetRaisedException();
PyErr_SetString(PyExc_ValueError,
Copy link
Copy Markdown
Member

@serhiy-storchaka serhiy-storchaka Jun 11, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why raise ValueError explicitly? If you want a ValueError for compatibility, UnicodeDecode is a subclass of ValueError, so this is a backward compatible change. Other functions which take const char * do not raise ValueError explicitly.

Copy link
Copy Markdown
Member Author

@vstinner vstinner Jun 11, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The error message helps debugging such issue: it points directly to the format string.

"PyUnicode_FromFormatV() expects a valid UTF-8-encoded "
"format string, got an invalid UTF-8 string");
_PyErr_ChainExceptions1(exc);
goto fail;
}

f = p;
f += len;
}
}
va_end(vargs2);
Expand Down Expand Up @@ -4930,13 +4927,9 @@ unicode_decode_utf8(const char *s, Py_ssize_t size,
static int
unicode_decode_utf8_writer(_PyUnicodeWriter *writer,
const char *s, Py_ssize_t size,
_Py_error_handler error_handler, const char *errors,
Py_ssize_t *consumed)
_Py_error_handler error_handler, const char *errors)
{
if (size == 0) {
if (consumed) {
*consumed = 0;
}
return 0;
}

Expand All @@ -4954,17 +4947,14 @@ unicode_decode_utf8_writer(_PyUnicodeWriter *writer,
writer->pos += decoded;

if (decoded == size) {
if (consumed) {
*consumed = size;
}
return 0;
}
s += decoded;
size -= decoded;
}

return unicode_decode_utf8_impl(writer, starts, s, end,
error_handler, errors, consumed);
error_handler, errors, NULL);
}


Expand Down