Skip to content

Commit f49a07b

Browse files
vstinnerZeroIntensitypicnixz
authored
gh-133968: Add PyUnicodeWriter_WriteASCII() function (#133973)
Replace most PyUnicodeWriter_WriteUTF8() calls with PyUnicodeWriter_WriteASCII(). Unrelated change to please the linter: remove an unused import in test_ctypes. Co-authored-by: Peter Bierma <zintensitydev@gmail.com> Co-authored-by: Bénédikt Tran <10796600+picnixz@users.noreply.github.com>
1 parent 4109a9c commit f49a07b

17 files changed

Lines changed: 103 additions & 31 deletions

File tree

‎Doc/c-api/unicode.rst‎

Lines changed: 16 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1802,9 +1802,24 @@ object.
18021802
18031803
See also :c:func:`PyUnicodeWriter_DecodeUTF8Stateful`.
18041804
1805+
.. c:function:: int PyUnicodeWriter_WriteASCII(PyUnicodeWriter *writer, const char *str, Py_ssize_t size)
1806+
1807+
Write the ASCII string *str* into *writer*.
1808+
1809+
*size* is the string length in bytes. If *size* is equal to ``-1``, call
1810+
``strlen(str)`` to get the string length.
1811+
1812+
*str* must only contain ASCII characters. The behavior is undefined if
1813+
*str* contains non-ASCII characters.
1814+
1815+
On success, return ``0``.
1816+
On error, set an exception, leave the writer unchanged, and return ``-1``.
1817+
1818+
.. versionadded:: next
1819+
18051820
.. c:function:: int PyUnicodeWriter_WriteWideChar(PyUnicodeWriter *writer, const wchar_t *str, Py_ssize_t size)
18061821
1807-
Writer the wide string *str* into *writer*.
1822+
Write the wide string *str* into *writer*.
18081823
18091824
*size* is a number of wide characters. If *size* is equal to ``-1``, call
18101825
``wcslen(str)`` to get the string length.

‎Doc/whatsnew/3.15.rst‎

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -226,6 +226,13 @@ New features
226226
functions as replacements for :c:func:`PySys_GetObject`.
227227
(Contributed by Serhiy Storchaka in :gh:`108512`.)
228228

229+
* Add :c:func:`PyUnicodeWriter_WriteASCII` function to write an ASCII string
230+
into a :c:type:`PyUnicodeWriter`. The function is faster than
231+
:c:func:`PyUnicodeWriter_WriteUTF8`, but has an undefined behavior if the
232+
input string contains non-ASCII characters.
233+
(Contributed by Victor Stinner in :gh:`133968`.)
234+
235+
229236
Porting to Python 3.15
230237
----------------------
231238

‎Include/cpython/unicodeobject.h‎

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -478,6 +478,10 @@ PyAPI_FUNC(int) PyUnicodeWriter_WriteUTF8(
478478
PyUnicodeWriter *writer,
479479
const char *str,
480480
Py_ssize_t size);
481+
PyAPI_FUNC(int) PyUnicodeWriter_WriteASCII(
482+
PyUnicodeWriter *writer,
483+
const char *str,
484+
Py_ssize_t size);
481485
PyAPI_FUNC(int) PyUnicodeWriter_WriteWideChar(
482486
PyUnicodeWriter *writer,
483487
const wchar_t *str,

‎Lib/test/test_capi/test_unicode.py‎

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1776,6 +1776,13 @@ def test_utf8(self):
17761776
self.assertEqual(writer.finish(),
17771777
"ascii-latin1=\xE9-euro=\u20AC.")
17781778

1779+
def test_ascii(self):
1780+
writer = self.create_writer(0)
1781+
writer.write_ascii(b"Hello ", -1)
1782+
writer.write_ascii(b"", 0)
1783+
writer.write_ascii(b"Python! <truncated>", 6)
1784+
self.assertEqual(writer.finish(), "Hello Python")
1785+
17791786
def test_invalid_utf8(self):
17801787
writer = self.create_writer(0)
17811788
with self.assertRaises(UnicodeDecodeError):

‎Lib/test/test_ctypes/test_incomplete.py‎

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,5 @@
11
import ctypes
22
import unittest
3-
import warnings
43
from ctypes import Structure, POINTER, pointer, c_char_p
54

65
# String-based "incomplete pointers" were implemented in ctypes 0.6.3 (2003, when

‎Misc/NEWS.d/next/C_API/2025-05-13-16-06-46.gh-issue-133968.6alWst.rst‎

Lines changed: 4 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,4 @@
1+
Add :c:func:`PyUnicodeWriter_WriteASCII` function to write an ASCII string
2+
into a :c:type:`PyUnicodeWriter`. The function is faster than
3+
:c:func:`PyUnicodeWriter_WriteUTF8`, but has an undefined behavior if the
4+
input string contains non-ASCII characters. Patch by Victor Stinner.

‎Modules/_json.c‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -1476,13 +1476,13 @@ encoder_listencode_obj(PyEncoderObject *s, PyUnicodeWriter *writer,
14761476
int rv;
14771477

14781478
if (obj == Py_None) {
1479-
return PyUnicodeWriter_WriteUTF8(writer, "null", 4);
1479+
return PyUnicodeWriter_WriteASCII(writer, "null", 4);
14801480
}
14811481
else if (obj == Py_True) {
1482-
return PyUnicodeWriter_WriteUTF8(writer, "true", 4);
1482+
return PyUnicodeWriter_WriteASCII(writer, "true", 4);
14831483
}
14841484
else if (obj == Py_False) {
1485-
return PyUnicodeWriter_WriteUTF8(writer, "false", 5);
1485+
return PyUnicodeWriter_WriteASCII(writer, "false", 5);
14861486
}
14871487
else if (PyUnicode_Check(obj)) {
14881488
PyObject *encoded = encoder_encode_string(s, obj);
@@ -1649,7 +1649,7 @@ encoder_listencode_dict(PyEncoderObject *s, PyUnicodeWriter *writer,
16491649

16501650
if (PyDict_GET_SIZE(dct) == 0) {
16511651
/* Fast path */
1652-
return PyUnicodeWriter_WriteUTF8(writer, "{}", 2);
1652+
return PyUnicodeWriter_WriteASCII(writer, "{}", 2);
16531653
}
16541654

16551655
if (s->markers != Py_None) {
@@ -1753,7 +1753,7 @@ encoder_listencode_list(PyEncoderObject *s, PyUnicodeWriter *writer,
17531753
return -1;
17541754
if (PySequence_Fast_GET_SIZE(s_fast) == 0) {
17551755
Py_DECREF(s_fast);
1756-
return PyUnicodeWriter_WriteUTF8(writer, "[]", 2);
1756+
return PyUnicodeWriter_WriteASCII(writer, "[]", 2);
17571757
}
17581758

17591759
if (s->markers != Py_None) {

‎Modules/_ssl.c‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -563,7 +563,7 @@ fill_and_set_sslerror(_sslmodulestate *state,
563563
goto fail;
564564
}
565565
}
566-
if (PyUnicodeWriter_WriteUTF8(writer, "] ", 2) < 0) {
566+
if (PyUnicodeWriter_WriteASCII(writer, "] ", 2) < 0) {
567567
goto fail;
568568
}
569569
}

‎Modules/_testcapi/unicode.c‎

Lines changed: 22 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -332,6 +332,27 @@ writer_write_utf8(PyObject *self_raw, PyObject *args)
332332
}
333333

334334

335+
static PyObject*
336+
writer_write_ascii(PyObject *self_raw, PyObject *args)
337+
{
338+
WriterObject *self = (WriterObject *)self_raw;
339+
if (writer_check(self) < 0) {
340+
return NULL;
341+
}
342+
343+
char *str;
344+
Py_ssize_t size;
345+
if (!PyArg_ParseTuple(args, "yn", &str, &size)) {
346+
return NULL;
347+
}
348+
349+
if (PyUnicodeWriter_WriteASCII(self->writer, str, size) < 0) {
350+
return NULL;
351+
}
352+
Py_RETURN_NONE;
353+
}
354+
355+
335356
static PyObject*
336357
writer_write_widechar(PyObject *self_raw, PyObject *args)
337358
{
@@ -513,6 +534,7 @@ writer_finish(PyObject *self_raw, PyObject *Py_UNUSED(args))
513534
static PyMethodDef writer_methods[] = {
514535
{"write_char", _PyCFunction_CAST(writer_write_char), METH_VARARGS},
515536
{"write_utf8", _PyCFunction_CAST(writer_write_utf8), METH_VARARGS},
537+
{"write_ascii", _PyCFunction_CAST(writer_write_ascii), METH_VARARGS},
516538
{"write_widechar", _PyCFunction_CAST(writer_write_widechar), METH_VARARGS},
517539
{"write_ucs4", _PyCFunction_CAST(writer_write_ucs4), METH_VARARGS},
518540
{"write_str", _PyCFunction_CAST(writer_write_str), METH_VARARGS},

‎Objects/genericaliasobject.c‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -65,7 +65,7 @@ ga_repr_items_list(PyUnicodeWriter *writer, PyObject *p)
6565

6666
for (Py_ssize_t i = 0; i < len; i++) {
6767
if (i > 0) {
68-
if (PyUnicodeWriter_WriteUTF8(writer, ", ", 2) < 0) {
68+
if (PyUnicodeWriter_WriteASCII(writer, ", ", 2) < 0) {
6969
return -1;
7070
}
7171
}
@@ -109,7 +109,7 @@ ga_repr(PyObject *self)
109109
}
110110
for (Py_ssize_t i = 0; i < len; i++) {
111111
if (i > 0) {
112-
if (PyUnicodeWriter_WriteUTF8(writer, ", ", 2) < 0) {
112+
if (PyUnicodeWriter_WriteASCII(writer, ", ", 2) < 0) {
113113
goto error;
114114
}
115115
}
@@ -126,7 +126,7 @@ ga_repr(PyObject *self)
126126
}
127127
if (len == 0) {
128128
// for something like tuple[()] we should print a "()"
129-
if (PyUnicodeWriter_WriteUTF8(writer, "()", 2) < 0) {
129+
if (PyUnicodeWriter_WriteASCII(writer, "()", 2) < 0) {
130130
goto error;
131131
}
132132
}

0 commit comments

Comments
 (0)