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

1 files changed, 30 insertions, 42 deletions

diff --git a/Python/bltinmodule.c b/Python/bltinmodule.c
index ed612091d83..b253f88a04b 100644
--- a/Python/bltinmodule.c
+++ b/Python/bltinmodule.c
@@ -11,21 +11,6 @@
 #include "pycore_tuple.h"         // _PyTuple_FromArray()
 #include "pycore_ceval.h"         // _PyEval_Vector()
 
-_Py_IDENTIFIER(__builtins__);
-_Py_IDENTIFIER(__dict__);
-_Py_IDENTIFIER(__prepare__);
-_Py_IDENTIFIER(__round__);
-_Py_IDENTIFIER(__mro_entries__);
-_Py_IDENTIFIER(encoding);
-_Py_IDENTIFIER(errors);
-_Py_IDENTIFIER(fileno);
-_Py_IDENTIFIER(flush);
-_Py_IDENTIFIER(metaclass);
-_Py_IDENTIFIER(sort);
-_Py_IDENTIFIER(stdin);
-_Py_IDENTIFIER(stdout);
-_Py_IDENTIFIER(stderr);
-
 #include "clinic/bltinmodule.c.h"
 
 static PyObject*
@@ -47,7 +32,7 @@ update_bases(PyObject *bases, PyObject *const *args, Py_ssize_t nargs)
             }
             continue;
         }
-        if (_PyObject_LookupAttrId(base, &PyId___mro_entries__, &meth) < 0) {
+        if (_PyObject_LookupAttr(base, &_Py_ID(__mro_entries__), &meth) < 0) {
             goto error;
         }
         if (!meth) {
@@ -148,10 +133,10 @@ builtin___build_class__(PyObject *self, PyObject *const *args, Py_ssize_t nargs,
             goto error;
         }
 
-        meta = _PyDict_GetItemIdWithError(mkw, &PyId_metaclass);
+        meta = _PyDict_GetItemWithError(mkw, &_Py_ID(metaclass));
         if (meta != NULL) {
             Py_INCREF(meta);
-            if (_PyDict_DelItemId(mkw, &PyId_metaclass) < 0) {
+            if (PyDict_DelItem(mkw, &_Py_ID(metaclass)) < 0) {
                 goto error;
             }
             /* metaclass is explicitly given, check if it's indeed a class */
@@ -191,7 +176,7 @@ builtin___build_class__(PyObject *self, PyObject *const *args, Py_ssize_t nargs,
     }
     /* else: meta is not a class, so we cannot do the metaclass
        calculation, so we will use the explicitly given object as it is */
-    if (_PyObject_LookupAttrId(meta, &PyId___prepare__, &prep) < 0) {
+    if (_PyObject_LookupAttr(meta, &_Py_ID(__prepare__), &prep) < 0) {
         ns = NULL;
     }
     else if (prep == NULL) {
@@ -946,10 +931,9 @@ builtin_eval_impl(PyObject *module, PyObject *source, PyObject *globals,
         return NULL;
     }
 
-    int r = _PyDict_ContainsId(globals, &PyId___builtins__);
+    int r = PyDict_Contains(globals, &_Py_ID(__builtins__));
     if (r == 0) {
-        r = _PyDict_SetItemId(globals, &PyId___builtins__,
-                              PyEval_GetBuiltins());
+        r = PyDict_SetItem(globals, &_Py_ID(__builtins__), PyEval_GetBuiltins());
     }
     if (r < 0) {
         return NULL;
@@ -1034,10 +1018,9 @@ builtin_exec_impl(PyObject *module, PyObject *source, PyObject *globals,
             Py_TYPE(locals)->tp_name);
         return NULL;
     }
-    int r = _PyDict_ContainsId(globals, &PyId___builtins__);
+    int r = PyDict_Contains(globals, &_Py_ID(__builtins__));
     if (r == 0) {
-        r = _PyDict_SetItemId(globals, &PyId___builtins__,
-                              PyEval_GetBuiltins());
+        r = PyDict_SetItem(globals, &_Py_ID(__builtins__), PyEval_GetBuiltins());
     }
     if (r < 0) {
         return NULL;
@@ -1960,7 +1943,8 @@ builtin_print_impl(PyObject *module, PyObject *args, PyObject *sep,
     int i, err;
 
     if (file == Py_None) {
-        file = _PySys_GetObjectId(&PyId_stdout);
+        PyThreadState *tstate = _PyThreadState_GET();
+        file = _PySys_GetAttr(tstate, &_Py_ID(stdout));
         if (file == NULL) {
             PyErr_SetString(PyExc_RuntimeError, "lost sys.stdout");
             return NULL;
@@ -2020,7 +2004,7 @@ builtin_print_impl(PyObject *module, PyObject *args, PyObject *sep,
     }
 
     if (flush) {
-        PyObject *tmp = _PyObject_CallMethodIdNoArgs(file, &PyId_flush);
+        PyObject *tmp = PyObject_CallMethodNoArgs(file, &_Py_ID(flush));
         if (tmp == NULL) {
             return NULL;
         }
@@ -2050,9 +2034,13 @@ static PyObject *
 builtin_input_impl(PyObject *module, PyObject *prompt)
 /*[clinic end generated code: output=83db5a191e7a0d60 input=5e8bb70c2908fe3c]*/
 {
-    PyObject *fin = _PySys_GetObjectId(&PyId_stdin);
-    PyObject *fout = _PySys_GetObjectId(&PyId_stdout);
-    PyObject *ferr = _PySys_GetObjectId(&PyId_stderr);
+    PyThreadState *tstate = _PyThreadState_GET();
+    PyObject *fin = _PySys_GetAttr(
+        tstate, &_Py_ID(stdin));
+    PyObject *fout = _PySys_GetAttr(
+        tstate, &_Py_ID(stdout));
+    PyObject *ferr = _PySys_GetAttr(
+        tstate, &_Py_ID(stderr));
     PyObject *tmp;
     long fd;
     int tty;
@@ -2079,7 +2067,7 @@ builtin_input_impl(PyObject *module, PyObject *prompt)
     }
 
     /* First of all, flush stderr */
-    tmp = _PyObject_CallMethodIdNoArgs(ferr, &PyId_flush);
+    tmp = PyObject_CallMethodNoArgs(ferr, &_Py_ID(flush));
     if (tmp == NULL)
         PyErr_Clear();
     else
@@ -2088,7 +2076,7 @@ builtin_input_impl(PyObject *module, PyObject *prompt)
     /* We should only use (GNU) readline if Python's sys.stdin and
        sys.stdout are the same as C's stdin and stdout, because we
        need to pass it those. */
-    tmp = _PyObject_CallMethodIdNoArgs(fin, &PyId_fileno);
+    tmp = PyObject_CallMethodNoArgs(fin, &_Py_ID(fileno));
     if (tmp == NULL) {
         PyErr_Clear();
         tty = 0;
@@ -2101,7 +2089,7 @@ builtin_input_impl(PyObject *module, PyObject *prompt)
         tty = fd == fileno(stdin) && isatty(fd);
     }
     if (tty) {
-        tmp = _PyObject_CallMethodIdNoArgs(fout, &PyId_fileno);
+        tmp = PyObject_CallMethodNoArgs(fout, &_Py_ID(fileno));
         if (tmp == NULL) {
             PyErr_Clear();
             tty = 0;
@@ -2127,8 +2115,8 @@ builtin_input_impl(PyObject *module, PyObject *prompt)
         size_t len;
 
         /* stdin is a text stream, so it must have an encoding. */
-        stdin_encoding = _PyObject_GetAttrId(fin, &PyId_encoding);
-        stdin_errors = _PyObject_GetAttrId(fin, &PyId_errors);
+        stdin_encoding = PyObject_GetAttr(fin, &_Py_ID(encoding));
+        stdin_errors = PyObject_GetAttr(fin, &_Py_ID(errors));
         if (!stdin_encoding || !stdin_errors ||
                 !PyUnicode_Check(stdin_encoding) ||
                 !PyUnicode_Check(stdin_errors)) {
@@ -2139,7 +2127,7 @@ builtin_input_impl(PyObject *module, PyObject *prompt)
         stdin_errors_str = PyUnicode_AsUTF8(stdin_errors);
         if (!stdin_encoding_str || !stdin_errors_str)
             goto _readline_errors;
-        tmp = _PyObject_CallMethodIdNoArgs(fout, &PyId_flush);
+        tmp = PyObject_CallMethodNoArgs(fout, &_Py_ID(flush));
         if (tmp == NULL)
             PyErr_Clear();
         else
@@ -2148,8 +2136,8 @@ builtin_input_impl(PyObject *module, PyObject *prompt)
             /* We have a prompt, encode it as stdout would */
             const char *stdout_encoding_str, *stdout_errors_str;
             PyObject *stringpo;
-            stdout_encoding = _PyObject_GetAttrId(fout, &PyId_encoding);
-            stdout_errors = _PyObject_GetAttrId(fout, &PyId_errors);
+            stdout_encoding = PyObject_GetAttr(fout, &_Py_ID(encoding));
+            stdout_errors = PyObject_GetAttr(fout, &_Py_ID(errors));
             if (!stdout_encoding || !stdout_errors ||
                     !PyUnicode_Check(stdout_encoding) ||
                     !PyUnicode_Check(stdout_errors)) {
@@ -2234,7 +2222,7 @@ builtin_input_impl(PyObject *module, PyObject *prompt)
         if (PyFile_WriteObject(prompt, fout, Py_PRINT_RAW) != 0)
             return NULL;
     }
-    tmp = _PyObject_CallMethodIdNoArgs(fout, &PyId_flush);
+    tmp = PyObject_CallMethodNoArgs(fout, &_Py_ID(flush));
     if (tmp == NULL)
         PyErr_Clear();
     else
@@ -2285,7 +2273,7 @@ builtin_round_impl(PyObject *module, PyObject *number, PyObject *ndigits)
             return NULL;
     }
 
-    round = _PyObject_LookupSpecial(number, &PyId___round__);
+    round = _PyObject_LookupSpecial(number, &_Py_ID(__round__));
     if (round == NULL) {
         if (!PyErr_Occurred())
             PyErr_Format(PyExc_TypeError,
@@ -2346,7 +2334,7 @@ builtin_sorted(PyObject *self, PyObject *const *args, Py_ssize_t nargs, PyObject
     if (newlist == NULL)
         return NULL;
 
-    callable = _PyObject_GetAttrId(newlist, &PyId_sort);
+    callable = PyObject_GetAttr(newlist, &_Py_ID(sort));
     if (callable == NULL) {
         Py_DECREF(newlist);
         return NULL;
@@ -2378,7 +2366,7 @@ builtin_vars(PyObject *self, PyObject *args)
         Py_XINCREF(d);
     }
     else {
-        if (_PyObject_LookupAttrId(v, &PyId___dict__, &d) == 0) {
+        if (_PyObject_LookupAttr(v, &_Py_ID(__dict__), &d) == 0) {
             PyErr_SetString(PyExc_TypeError,
                 "vars() argument must have __dict__ attribute");
         }