bpo-46541: Replace core use of _Py_IDENTIFIER() with statically initialized global objects. (gh-30928)

We're no longer using _Py_IDENTIFIER() (or _Py_static_string()) in any core CPython code.  It is still used in a number of non-builtin stdlib modules.

The replacement is: PyUnicodeObject (not pointer) fields under _PyRuntimeState, statically initialized as part of _PyRuntime.  A new _Py_GET_GLOBAL_IDENTIFIER() macro facilitates lookup of the fields (along with _Py_GET_GLOBAL_STRING() for non-identifier strings).

https://bugs.python.org/issue46541#msg411799 explains the rationale for this change.

The core of the change is in:

* (new) Include/internal/pycore_global_strings.h - the declarations for the global strings, along with the macros
* Include/internal/pycore_runtime_init.h - added the static initializers for the global strings
* Include/internal/pycore_global_objects.h - where the struct in pycore_global_strings.h is hooked into _PyRuntimeState
* Tools/scripts/generate_global_objects.py - added generation of the global string declarations and static initializers

I've also added a --check flag to generate_global_objects.py (along with make check-global-objects) to check for unused global strings.  That check is added to the PR CI config.

The remainder of this change updates the core code to use _Py_GET_GLOBAL_IDENTIFIER() instead of _Py_IDENTIFIER() and the related _Py*Id functions (likewise for _Py_GET_GLOBAL_STRING() instead of _Py_static_string()).  This includes adding a few functions where there wasn't already an alternative to _Py*Id(), replacing the _Py_Identifier * parameter with PyObject *.

The following are not changed (yet):

* stop using _Py_IDENTIFIER() in the stdlib modules
* (maybe) get rid of _Py_IDENTIFIER(), etc. entirely -- this may not be doable as at least one package on PyPI using this (private) API
* (maybe) intern the strings during runtime init

https://bugs.python.org/issue46541
This commit is contained in:
Eric Snow 2022-02-08 13:39:07 -07:00 committed by GitHub
parent c018d3037b
commit 81c72044a1
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
108 changed files with 2282 additions and 1573 deletions

View file

@ -31,11 +31,6 @@ extern "C" {
/* Defined in tracemalloc.c */
extern void _PyMem_DumpTraceback(int fd, const void *ptr);
_Py_IDENTIFIER(Py_Repr);
_Py_IDENTIFIER(__bytes__);
_Py_IDENTIFIER(__dir__);
_Py_IDENTIFIER(__isabstractmethod__);
int
_PyObject_CheckConsistency(PyObject *op, int check_content)
@ -562,7 +557,7 @@ PyObject_Bytes(PyObject *v)
return v;
}
func = _PyObject_LookupSpecial(v, &PyId___bytes__);
func = _PyObject_LookupSpecial(v, &_Py_ID(__bytes__));
if (func != NULL) {
result = _PyObject_CallNoArgs(func);
Py_DECREF(func);
@ -600,12 +595,9 @@ def _PyObject_FunctionStr(x):
PyObject *
_PyObject_FunctionStr(PyObject *x)
{
_Py_IDENTIFIER(__module__);
_Py_IDENTIFIER(__qualname__);
_Py_IDENTIFIER(builtins);
assert(!PyErr_Occurred());
PyObject *qualname;
int ret = _PyObject_LookupAttrId(x, &PyId___qualname__, &qualname);
int ret = _PyObject_LookupAttr(x, &_Py_ID(__qualname__), &qualname);
if (qualname == NULL) {
if (ret < 0) {
return NULL;
@ -614,13 +606,9 @@ _PyObject_FunctionStr(PyObject *x)
}
PyObject *module;
PyObject *result = NULL;
ret = _PyObject_LookupAttrId(x, &PyId___module__, &module);
ret = _PyObject_LookupAttr(x, &_Py_ID(__module__), &module);
if (module != NULL && module != Py_None) {
PyObject *builtinsname = _PyUnicode_FromId(&PyId_builtins);
if (builtinsname == NULL) {
goto done;
}
ret = PyObject_RichCompareBool(module, builtinsname, Py_NE);
ret = PyObject_RichCompareBool(module, &_Py_ID(builtins), Py_NE);
if (ret < 0) {
// error
goto done;
@ -858,7 +846,7 @@ _PyObject_IsAbstract(PyObject *obj)
if (obj == NULL)
return 0;
res = _PyObject_LookupAttrId(obj, &PyId___isabstractmethod__, &isabstract);
res = _PyObject_LookupAttr(obj, &_Py_ID(__isabstractmethod__), &isabstract);
if (res > 0) {
res = PyObject_IsTrue(isabstract);
Py_DECREF(isabstract);
@ -892,8 +880,6 @@ static inline int
set_attribute_error_context(PyObject* v, PyObject* name)
{
assert(PyErr_Occurred());
_Py_IDENTIFIER(name);
_Py_IDENTIFIER(obj);
// Intercept AttributeError exceptions and augment them to offer
// suggestions later.
if (PyErr_ExceptionMatches(PyExc_AttributeError)){
@ -901,8 +887,8 @@ set_attribute_error_context(PyObject* v, PyObject* name)
PyErr_Fetch(&type, &value, &traceback);
PyErr_NormalizeException(&type, &value, &traceback);
if (PyErr_GivenExceptionMatches(value, PyExc_AttributeError) &&
(_PyObject_SetAttrId(value, &PyId_name, name) ||
_PyObject_SetAttrId(value, &PyId_obj, v))) {
(PyObject_SetAttr(value, &_Py_ID(name), name) ||
PyObject_SetAttr(value, &_Py_ID(obj), v))) {
return 1;
}
PyErr_Restore(type, value, traceback);
@ -1569,7 +1555,7 @@ static PyObject *
_dir_object(PyObject *obj)
{
PyObject *result, *sorted;
PyObject *dirfunc = _PyObject_LookupSpecial(obj, &PyId___dir__);
PyObject *dirfunc = _PyObject_LookupSpecial(obj, &_Py_ID(__dir__));
assert(obj != NULL);
if (dirfunc == NULL) {
@ -2148,7 +2134,7 @@ Py_ReprEnter(PyObject *obj)
early on startup. */
if (dict == NULL)
return 0;
list = _PyDict_GetItemIdWithError(dict, &PyId_Py_Repr);
list = PyDict_GetItemWithError(dict, &_Py_ID(Py_Repr));
if (list == NULL) {
if (PyErr_Occurred()) {
return -1;
@ -2156,7 +2142,7 @@ Py_ReprEnter(PyObject *obj)
list = PyList_New(0);
if (list == NULL)
return -1;
if (_PyDict_SetItemId(dict, &PyId_Py_Repr, list) < 0)
if (PyDict_SetItem(dict, &_Py_ID(Py_Repr), list) < 0)
return -1;
Py_DECREF(list);
}
@ -2184,7 +2170,7 @@ Py_ReprLeave(PyObject *obj)
if (dict == NULL)
goto finally;
list = _PyDict_GetItemIdWithError(dict, &PyId_Py_Repr);
list = PyDict_GetItemWithError(dict, &_Py_ID(Py_Repr));
if (list == NULL || !PyList_Check(list))
goto finally;