Stowage/cpython
2
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-11-03 23:21:29 +00:00
Code Issues Projects Releases Packages Wiki Activity

gh-111972: Make Unicode name C APIcapsule initialization thread-safe (#112249)

Browse source
This commit is contained in:
Kirill Podoprigora 2023-11-30 13:12:49 +03:00 • committed by GitHub
parent 81261fa67f
commit 0785c68559
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 26 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

2
Include/internal/pycore_ucnhash.h
View file

@ -28,6 +28,8 @@ typedef struct {
} _PyUnicode_Name_CAPI; } _PyUnicode_Name_CAPI;
extern _PyUnicode_Name_CAPI* _PyUnicode_GetNameCAPI(void);
#ifdef __cplusplus #ifdef __cplusplus
} }
#endif #endif

26
Objects/unicodeobject.c
View file

@ -5869,6 +5869,23 @@ PyUnicode_AsUTF16String(PyObject *unicode)
return _PyUnicode_EncodeUTF16(unicode, NULL, 0); return _PyUnicode_EncodeUTF16(unicode, NULL, 0);
} }
_PyUnicode_Name_CAPI *
_PyUnicode_GetNameCAPI(void)
{
PyInterpreterState *interp = _PyInterpreterState_GET();
_PyUnicode_Name_CAPI *ucnhash_capi;
ucnhash_capi = _Py_atomic_load_ptr(&interp->unicode.ucnhash_capi);
if (ucnhash_capi == NULL) {
ucnhash_capi = (_PyUnicode_Name_CAPI *)PyCapsule_Import(
PyUnicodeData_CAPSULE_NAME, 1);
// It's fine if we overwite the value here. It's always the same value.
_Py_atomic_store_ptr(&interp->unicode.ucnhash_capi, ucnhash_capi);
}
return ucnhash_capi;
}
/* --- Unicode Escape Codec ----------------------------------------------- */ /* --- Unicode Escape Codec ----------------------------------------------- */
PyObject * PyObject *
@ -5884,7 +5901,6 @@ _PyUnicode_DecodeUnicodeEscapeInternal(const char *s,
PyObject *errorHandler = NULL; PyObject *errorHandler = NULL;
PyObject *exc = NULL; PyObject *exc = NULL;
_PyUnicode_Name_CAPI *ucnhash_capi; _PyUnicode_Name_CAPI *ucnhash_capi;
PyInterpreterState *interp = _PyInterpreterState_GET();
// so we can remember if we've seen an invalid escape char or not // so we can remember if we've seen an invalid escape char or not
*first_invalid_escape = NULL; *first_invalid_escape = NULL;
@ -6032,11 +6048,7 @@ _PyUnicode_DecodeUnicodeEscapeInternal(const char *s,
/* \N{name} */ /* \N{name} */
case 'N': case 'N':
ucnhash_capi = interp->unicode.ucnhash_capi; ucnhash_capi = _PyUnicode_GetNameCAPI();
if (ucnhash_capi == NULL) {
/* load the unicode data module */
ucnhash_capi = (_PyUnicode_Name_CAPI *)PyCapsule_Import(
PyUnicodeData_CAPSULE_NAME, 1);
if (ucnhash_capi == NULL) { if (ucnhash_capi == NULL) {
PyErr_SetString( PyErr_SetString(
PyExc_UnicodeError, PyExc_UnicodeError,
@ -6044,8 +6056,6 @@ _PyUnicode_DecodeUnicodeEscapeInternal(const char *s,
); );
goto onError; goto onError;
} }
interp->unicode.ucnhash_capi = ucnhash_capi;
}
message = "malformed \\N character escape"; message = "malformed \\N character escape";
if (s >= end) { if (s >= end) {

10
Python/codecs.c
View file

@ -931,8 +931,6 @@ PyObject *PyCodec_BackslashReplaceErrors(PyObject *exc)
return Py_BuildValue("(Nn)", res, end); return Py_BuildValue("(Nn)", res, end);
} }
static _PyUnicode_Name_CAPI *ucnhash_capi = NULL;
PyObject *PyCodec_NameReplaceErrors(PyObject *exc) PyObject *PyCodec_NameReplaceErrors(PyObject *exc)
{ {
if (PyObject_TypeCheck(exc, (PyTypeObject *)PyExc_UnicodeEncodeError)) { if (PyObject_TypeCheck(exc, (PyTypeObject *)PyExc_UnicodeEncodeError)) {
@ -953,14 +951,10 @@ PyObject *PyCodec_NameReplaceErrors(PyObject *exc)
return NULL; return NULL;
if (!(object = PyUnicodeEncodeError_GetObject(exc))) if (!(object = PyUnicodeEncodeError_GetObject(exc)))
return NULL; return NULL;
if (!ucnhash_capi) { _PyUnicode_Name_CAPI *ucnhash_capi = _PyUnicode_GetNameCAPI();
/* load the unicode data module */ if (ucnhash_capi == NULL) {
ucnhash_capi = (_PyUnicode_Name_CAPI *)PyCapsule_Import(
PyUnicodeData_CAPSULE_NAME, 1);
if (!ucnhash_capi) {
return NULL; return NULL;
} }
}
for (i = start, ressize = 0; i < end; ++i) { for (i = start, ressize = 0; i < end; ++i) {
/* object is guaranteed to be "ready" */ /* object is guaranteed to be "ready" */
c = PyUnicode_READ_CHAR(object, i); c = PyUnicode_READ_CHAR(object, i);