https://github.com/python/cpython/commit/e0637cebe5bf863897f2e89dfcb76be0015c1877
commit: e0637cebe5bf863897f2e89dfcb76be0015c1877
branch: main
author: Daniel Pope <lordma...@users.noreply.github.com>
committer: vstinner <vstin...@python.org>
date: 2025-03-12T11:40:11+01:00
summary:
gh-129349: Accept bytes in bytes.fromhex()/bytearray.fromhex() (#129844)
Co-authored-by: Bénédikt Tran <10796600+picn...@users.noreply.github.com>
Co-authored-by: Victor Stinner <vstin...@python.org>
files:
A
Misc/NEWS.d/next/Core_and_Builtins/2025-02-08-09-55-33.gh-issue-129349.PkcG-l.rst
M Doc/library/stdtypes.rst
M Doc/whatsnew/3.14.rst
M Lib/test/test_bytes.py
M Objects/bytearrayobject.c
M Objects/bytesobject.c
M Objects/clinic/bytearrayobject.c.h
M Objects/clinic/bytesobject.c.h
diff --git a/Doc/library/stdtypes.rst b/Doc/library/stdtypes.rst
index a6260ecd77f520..7b3fa218317fcc 100644
--- a/Doc/library/stdtypes.rst
+++ b/Doc/library/stdtypes.rst
@@ -2744,6 +2744,10 @@ data and are closely related to string objects in a
variety of other ways.
:meth:`bytes.fromhex` now skips all ASCII whitespace in the string,
not just spaces.
+ .. versionchanged:: next
+ :meth:`bytes.fromhex` now accepts ASCII :class:`bytes` and
+ :term:`bytes-like objects <bytes-like object>` as input.
+
A reverse conversion function exists to transform a bytes object into its
hexadecimal representation.
@@ -2829,6 +2833,10 @@ objects.
:meth:`bytearray.fromhex` now skips all ASCII whitespace in the
string,
not just spaces.
+ .. versionchanged:: next
+ :meth:`bytearray.fromhex` now accepts ASCII :class:`bytes` and
+ :term:`bytes-like objects <bytes-like object>` as input.
+
A reverse conversion function exists to transform a bytearray object into
its
hexadecimal representation.
diff --git a/Doc/whatsnew/3.14.rst b/Doc/whatsnew/3.14.rst
index 6539b23380b6e0..6898b50b2c932a 100644
--- a/Doc/whatsnew/3.14.rst
+++ b/Doc/whatsnew/3.14.rst
@@ -354,6 +354,10 @@ Other language changes
(with :func:`format` or :ref:`f-strings`).
(Contrubuted by Sergey B Kirpichev in :gh:`87790`.)
+* The :func:`bytes.fromhex` and :func:`bytearray.fromhex` methods now accept
+ ASCII :class:`bytes` and :term:`bytes-like objects <bytes-like object>`.
+ (Contributed by Daniel Pope in :gh:`129349`.)
+
* ``\B`` in :mod:`regular expression <re>` now matches empty input string.
Now it is always the opposite of ``\b``.
(Contributed by Serhiy Storchaka in :gh:`124130`.)
diff --git a/Lib/test/test_bytes.py b/Lib/test/test_bytes.py
index f6ffe83c5d69e8..d5490a20a0525d 100644
--- a/Lib/test/test_bytes.py
+++ b/Lib/test/test_bytes.py
@@ -450,13 +450,34 @@ def test_fromhex(self):
# check that ASCII whitespace is ignored
self.assertEqual(self.type2test.fromhex(' 1A\n2B\t30\v'), b)
+ self.assertEqual(self.type2test.fromhex(b' 1A\n2B\t30\v'), b)
for c in "\x09\x0A\x0B\x0C\x0D\x20":
self.assertEqual(self.type2test.fromhex(c), self.type2test())
for c in "\x1C\x1D\x1E\x1F\x85\xa0\u2000\u2002\u2028":
self.assertRaises(ValueError, self.type2test.fromhex, c)
+ # Check that we can parse bytes and bytearray
+ tests = [
+ ("bytes", bytes),
+ ("bytearray", bytearray),
+ ("memoryview", memoryview),
+ ("array.array", lambda bs: array.array('B', bs)),
+ ]
+ for name, factory in tests:
+ with self.subTest(name=name):
+ self.assertEqual(self.type2test.fromhex(factory(b' 1A 2B 30
')), b)
+
+ # Invalid bytes are rejected
+ for u8 in b"\0\x1C\x1D\x1E\x1F\x85\xa0":
+ b = bytes([30, 31, u8])
+ self.assertRaises(ValueError, self.type2test.fromhex, b)
+
self.assertEqual(self.type2test.fromhex('0000'), b'\0\0')
- self.assertRaises(TypeError, self.type2test.fromhex, b'1B')
+ with self.assertRaisesRegex(
+ TypeError,
+ r'fromhex\(\) argument must be str or bytes-like, not tuple',
+ ):
+ self.type2test.fromhex(())
self.assertRaises(ValueError, self.type2test.fromhex, 'a')
self.assertRaises(ValueError, self.type2test.fromhex, 'rt')
self.assertRaises(ValueError, self.type2test.fromhex, '1a b cd')
diff --git
a/Misc/NEWS.d/next/Core_and_Builtins/2025-02-08-09-55-33.gh-issue-129349.PkcG-l.rst
b/Misc/NEWS.d/next/Core_and_Builtins/2025-02-08-09-55-33.gh-issue-129349.PkcG-l.rst
new file mode 100644
index 00000000000000..db2af780bd5cbb
--- /dev/null
+++
b/Misc/NEWS.d/next/Core_and_Builtins/2025-02-08-09-55-33.gh-issue-129349.PkcG-l.rst
@@ -0,0 +1,2 @@
+:meth:`bytes.fromhex` and :meth:`bytearray.fromhex` now accepts ASCII
+:class:`bytes` and :term:`bytes-like objects <bytes-like object>`.
diff --git a/Objects/bytearrayobject.c b/Objects/bytearrayobject.c
index 34f43eb8c31315..f1c76664198abc 100644
--- a/Objects/bytearrayobject.c
+++ b/Objects/bytearrayobject.c
@@ -2533,7 +2533,7 @@ bytearray_splitlines_impl(PyByteArrayObject *self, int
keepends)
@classmethod
bytearray.fromhex
- string: unicode
+ string: object
/
Create a bytearray object from a string of hexadecimal numbers.
@@ -2543,8 +2543,8 @@ Example: bytearray.fromhex('B9 01EF') ->
bytearray(b'\\xb9\\x01\\xef')
[clinic start generated code]*/
static PyObject *
-bytearray_fromhex_impl(PyTypeObject *type, PyObject *string)
-/*[clinic end generated code: output=8f0f0b6d30fb3ba0 input=f033a16d1fb21f48]*/
+bytearray_fromhex(PyTypeObject *type, PyObject *string)
+/*[clinic end generated code: output=da84dc708e9c4b36 input=7e314e5b2d7ab484]*/
{
PyObject *result = _PyBytes_FromHex(string, type == &PyByteArray_Type);
if (type != &PyByteArray_Type && result != NULL) {
diff --git a/Objects/bytesobject.c b/Objects/bytesobject.c
index ba642d3788fc78..ada0d0024827dc 100644
--- a/Objects/bytesobject.c
+++ b/Objects/bytesobject.c
@@ -2484,7 +2484,7 @@ bytes_splitlines_impl(PyBytesObject *self, int keepends)
@classmethod
bytes.fromhex
- string: unicode
+ string: object
/
Create a bytes object from a string of hexadecimal numbers.
@@ -2494,8 +2494,8 @@ Example: bytes.fromhex('B9 01EF') -> b'\\xb9\\x01\\xef'.
[clinic start generated code]*/
static PyObject *
-bytes_fromhex_impl(PyTypeObject *type, PyObject *string)
-/*[clinic end generated code: output=0973acc63661bb2e input=bf4d1c361670acd3]*/
+bytes_fromhex(PyTypeObject *type, PyObject *string)
+/*[clinic end generated code: output=d458ec88195da6b3 input=f37d98ed51088a21]*/
{
PyObject *result = _PyBytes_FromHex(string, 0);
if (type != &PyBytes_Type && result != NULL) {
@@ -2510,37 +2510,55 @@ _PyBytes_FromHex(PyObject *string, int use_bytearray)
char *buf;
Py_ssize_t hexlen, invalid_char;
unsigned int top, bot;
- const Py_UCS1 *str, *end;
+ const Py_UCS1 *str, *start, *end;
_PyBytesWriter writer;
+ Py_buffer view;
+ view.obj = NULL;
_PyBytesWriter_Init(&writer);
writer.use_bytearray = use_bytearray;
- assert(PyUnicode_Check(string));
- hexlen = PyUnicode_GET_LENGTH(string);
+ if (PyUnicode_Check(string)) {
+ hexlen = PyUnicode_GET_LENGTH(string);
- if (!PyUnicode_IS_ASCII(string)) {
- const void *data = PyUnicode_DATA(string);
- int kind = PyUnicode_KIND(string);
- Py_ssize_t i;
+ if (!PyUnicode_IS_ASCII(string)) {
+ const void *data = PyUnicode_DATA(string);
+ int kind = PyUnicode_KIND(string);
+ Py_ssize_t i;
- /* search for the first non-ASCII character */
- for (i = 0; i < hexlen; i++) {
- if (PyUnicode_READ(kind, data, i) >= 128)
- break;
+ /* search for the first non-ASCII character */
+ for (i = 0; i < hexlen; i++) {
+ if (PyUnicode_READ(kind, data, i) >= 128)
+ break;
+ }
+ invalid_char = i;
+ goto error;
}
- invalid_char = i;
- goto error;
- }
- assert(PyUnicode_KIND(string) == PyUnicode_1BYTE_KIND);
- str = PyUnicode_1BYTE_DATA(string);
+ assert(PyUnicode_KIND(string) == PyUnicode_1BYTE_KIND);
+ str = PyUnicode_1BYTE_DATA(string);
+ }
+ else if (PyObject_CheckBuffer(string)) {
+ if (PyObject_GetBuffer(string, &view, PyBUF_SIMPLE) != 0) {
+ return NULL;
+ }
+ hexlen = view.len;
+ str = view.buf;
+ }
+ else {
+ PyErr_Format(PyExc_TypeError,
+ "fromhex() argument must be str or bytes-like, not %T",
+ string);
+ return NULL;
+ }
/* This overestimates if there are spaces */
buf = _PyBytesWriter_Alloc(&writer, hexlen / 2);
- if (buf == NULL)
- return NULL;
+ if (buf == NULL) {
+ goto release_buffer;
+ }
+ start = str;
end = str + hexlen;
while (str < end) {
/* skip over spaces in the input */
@@ -2554,7 +2572,7 @@ _PyBytes_FromHex(PyObject *string, int use_bytearray)
top = _PyLong_DigitValue[*str];
if (top >= 16) {
- invalid_char = str - PyUnicode_1BYTE_DATA(string);
+ invalid_char = str - start;
goto error;
}
str++;
@@ -2565,7 +2583,7 @@ _PyBytes_FromHex(PyObject *string, int use_bytearray)
if (str >= end){
invalid_char = -1;
} else {
- invalid_char = str - PyUnicode_1BYTE_DATA(string);
+ invalid_char = str - start;
}
goto error;
}
@@ -2574,6 +2592,9 @@ _PyBytes_FromHex(PyObject *string, int use_bytearray)
*buf++ = (unsigned char)((top << 4) + bot);
}
+ if (view.obj != NULL) {
+ PyBuffer_Release(&view);
+ }
return _PyBytesWriter_Finish(&writer, buf);
error:
@@ -2586,6 +2607,11 @@ _PyBytes_FromHex(PyObject *string, int use_bytearray)
"fromhex() arg at position %zd", invalid_char);
}
_PyBytesWriter_Dealloc(&writer);
+
+ release_buffer:
+ if (view.obj != NULL) {
+ PyBuffer_Release(&view);
+ }
return NULL;
}
diff --git a/Objects/clinic/bytearrayobject.c.h
b/Objects/clinic/bytearrayobject.c.h
index fa105f74c58512..8ed10d8ab5a8b2 100644
--- a/Objects/clinic/bytearrayobject.c.h
+++ b/Objects/clinic/bytearrayobject.c.h
@@ -1601,26 +1601,6 @@ PyDoc_STRVAR(bytearray_fromhex__doc__,
#define BYTEARRAY_FROMHEX_METHODDEF \
{"fromhex", (PyCFunction)bytearray_fromhex, METH_O|METH_CLASS,
bytearray_fromhex__doc__},
-static PyObject *
-bytearray_fromhex_impl(PyTypeObject *type, PyObject *string);
-
-static PyObject *
-bytearray_fromhex(PyTypeObject *type, PyObject *arg)
-{
- PyObject *return_value = NULL;
- PyObject *string;
-
- if (!PyUnicode_Check(arg)) {
- _PyArg_BadArgument("fromhex", "argument", "str", arg);
- goto exit;
- }
- string = arg;
- return_value = bytearray_fromhex_impl(type, string);
-
-exit:
- return return_value;
-}
-
PyDoc_STRVAR(bytearray_hex__doc__,
"hex($self, /, sep=<unrepresentable>, bytes_per_sep=1)\n"
"--\n"
@@ -1789,4 +1769,4 @@ bytearray_sizeof(PyObject *self, PyObject
*Py_UNUSED(ignored))
{
return bytearray_sizeof_impl((PyByteArrayObject *)self);
}
-/*[clinic end generated code: output=7c924a56e0a8bfe6 input=a9049054013a1b77]*/
+/*[clinic end generated code: output=13a4231325b7d3c1 input=a9049054013a1b77]*/
diff --git a/Objects/clinic/bytesobject.c.h b/Objects/clinic/bytesobject.c.h
index 11cb81a9c5c9d7..c0f61f1368ba2b 100644
--- a/Objects/clinic/bytesobject.c.h
+++ b/Objects/clinic/bytesobject.c.h
@@ -1204,26 +1204,6 @@ PyDoc_STRVAR(bytes_fromhex__doc__,
#define BYTES_FROMHEX_METHODDEF \
{"fromhex", (PyCFunction)bytes_fromhex, METH_O|METH_CLASS,
bytes_fromhex__doc__},
-static PyObject *
-bytes_fromhex_impl(PyTypeObject *type, PyObject *string);
-
-static PyObject *
-bytes_fromhex(PyTypeObject *type, PyObject *arg)
-{
- PyObject *return_value = NULL;
- PyObject *string;
-
- if (!PyUnicode_Check(arg)) {
- _PyArg_BadArgument("fromhex", "argument", "str", arg);
- goto exit;
- }
- string = arg;
- return_value = bytes_fromhex_impl(type, string);
-
-exit:
- return return_value;
-}
-
PyDoc_STRVAR(bytes_hex__doc__,
"hex($self, /, sep=<unrepresentable>, bytes_per_sep=1)\n"
"--\n"
@@ -1404,4 +1384,4 @@ bytes_new(PyTypeObject *type, PyObject *args, PyObject
*kwargs)
exit:
return return_value;
}
-/*[clinic end generated code: output=61cb2cf6506df4c6 input=a9049054013a1b77]*/
+/*[clinic end generated code: output=967aae4b46423586 input=a9049054013a1b77]*/
_______________________________________________
Python-checkins mailing list -- python-checkins@python.org
To unsubscribe send an email to python-checkins-le...@python.org
https://mail.python.org/mailman3/lists/python-checkins.python.org/
Member address: arch...@mail-archive.com
