bpo-46541: Discover the global strings. (gh-31346)

Instead of manually enumerating the global strings in generate_global_objects.py, we extrapolate the list from usage of _Py_ID() and _Py_STR() in the source files.

This is partly inspired by gh-31261.

https://bugs.python.org/issue46541

Author: ericsnowcurrently

Include/internal/pycore_global_strings.h

@@ -28,21 +28,19 @@ extern "C" {
 /* The following is auto-generated by Tools/scripts/generate_global_objects.py. */
 struct _Py_global_strings {
     struct {
-        STRUCT_FOR_STR(empty, "")
-        STRUCT_FOR_STR(dot, ".")
-        STRUCT_FOR_STR(comma_sep, ", ")
-        STRUCT_FOR_STR(percent, "%")
-        STRUCT_FOR_STR(dbl_percent, "%%")
-
-        // "anonymous" labels
         STRUCT_FOR_STR(anon_dictcomp, "<dictcomp>")
         STRUCT_FOR_STR(anon_genexpr, "<genexpr>")
         STRUCT_FOR_STR(anon_lambda, "<lambda>")
         STRUCT_FOR_STR(anon_listcomp, "<listcomp>")
         STRUCT_FOR_STR(anon_module, "<module>")
         STRUCT_FOR_STR(anon_setcomp, "<setcomp>")
         STRUCT_FOR_STR(anon_string, "<string>")
+        STRUCT_FOR_STR(comma_sep, ", ")
+        STRUCT_FOR_STR(dbl_percent, "%%")
+        STRUCT_FOR_STR(dot, ".")
         STRUCT_FOR_STR(dot_locals, ".<locals>")
+        STRUCT_FOR_STR(empty, "")
+        STRUCT_FOR_STR(percent, "%")
     } literals;
 
     struct {
@@ -330,6 +328,7 @@ struct _Py_global_strings {
 #define _Py_STR(NAME) \
      (_Py_SINGLETON(strings.literals._ ## NAME._ascii.ob_base))
 
+#define _Py_DECLARE_STR(name, str)
 
 #ifdef __cplusplus
 }

Include/internal/pycore_runtime_init.h

@@ -644,20 +644,19 @@ extern "C" {
         \
         .strings = { \
             .literals = { \
-                INIT_STR(empty, ""), \
-                INIT_STR(dot, "."), \
-                INIT_STR(comma_sep, ", "), \
-                INIT_STR(percent, "%"), \
-                INIT_STR(dbl_percent, "%%"), \
-                \
                 INIT_STR(anon_dictcomp, "<dictcomp>"), \
                 INIT_STR(anon_genexpr, "<genexpr>"), \
                 INIT_STR(anon_lambda, "<lambda>"), \
                 INIT_STR(anon_listcomp, "<listcomp>"), \
                 INIT_STR(anon_module, "<module>"), \
                 INIT_STR(anon_setcomp, "<setcomp>"), \
                 INIT_STR(anon_string, "<string>"), \
+                INIT_STR(comma_sep, ", "), \
+                INIT_STR(dbl_percent, "%%"), \
+                INIT_STR(dot, "."), \
                 INIT_STR(dot_locals, ".<locals>"), \
+                INIT_STR(empty, ""), \
+                INIT_STR(percent, "%"), \
             }, \
             .identifiers = { \
                 INIT_ID(Py_Repr), \

Objects/typeobject.c

@@ -4546,6 +4546,7 @@ object_new(PyTypeObject *type, PyObject *args, PyObject *kwds)
             Py_DECREF(sorted_methods);
             return NULL;
         }
+        _Py_DECLARE_STR(comma_sep, ", ");
         joined = PyUnicode_Join(&_Py_STR(comma_sep), sorted_methods);
         method_count = PyObject_Length(sorted_methods);
         Py_DECREF(sorted_methods);

Objects/weakrefobject.c

@@ -458,12 +458,12 @@ proxy_checkref(PyWeakReference *proxy)
         return res; \
     }
 
-#define WRAP_METHOD(method, special) \
+#define WRAP_METHOD(method, SPECIAL) \
     static PyObject * \
     method(PyObject *proxy, PyObject *Py_UNUSED(ignored)) { \
             UNWRAP(proxy); \
             Py_INCREF(proxy); \
-            PyObject* res = PyObject_CallMethodNoArgs(proxy, &_Py_ID(special)); \
+            PyObject* res = PyObject_CallMethodNoArgs(proxy, &_Py_ID(SPECIAL)); \
             Py_DECREF(proxy); \
             return res; \
         }

Python/_warnings.c

@@ -186,8 +186,8 @@ check_matched(PyInterpreterState *interp, PyObject *obj, PyObject *arg)
     return rc;
 }
 
-#define GET_WARNINGS_ATTR(interp, attr, try_import) \
-    get_warnings_attr(interp, &_Py_ID(attr), try_import)
+#define GET_WARNINGS_ATTR(interp, ATTR, try_import) \
+    get_warnings_attr(interp, &_Py_ID(ATTR), try_import)
 
 /*
    Returns a new reference.

Python/ast_opt.c

@@ -268,6 +268,8 @@ parse_literal(PyObject *fmt, Py_ssize_t *ppos, PyArena *arena)
     PyObject *str = PyUnicode_Substring(fmt, start, pos);
     /* str = str.replace('%%', '%') */
     if (str && has_percents) {
+        _Py_DECLARE_STR(percent, "%");
+        _Py_DECLARE_STR(dbl_percent, "%%");
         Py_SETREF(str, PyUnicode_Replace(str, &_Py_STR(dbl_percent),
                                          &_Py_STR(percent), -1));
     }

Python/compile.c

@@ -667,6 +667,7 @@ compiler_set_qualname(struct compiler *c)
                 || parent->u_scope_type == COMPILER_SCOPE_ASYNC_FUNCTION
                 || parent->u_scope_type == COMPILER_SCOPE_LAMBDA)
             {
+                _Py_DECLARE_STR(dot_locals, ".<locals>");
                 base = PyUnicode_Concat(parent->u_qualname,
                                         &_Py_STR(dot_locals));
                 if (base == NULL)
@@ -2022,6 +2023,7 @@ compiler_mod(struct compiler *c, mod_ty mod)
 {
     PyCodeObject *co;
     int addNone = 1;
+    _Py_DECLARE_STR(anon_module, "<module>");
     if (!compiler_enter_scope(c, &_Py_STR(anon_module), COMPILER_SCOPE_MODULE,
                               mod, 1)) {
         return NULL;
@@ -2876,6 +2878,7 @@ compiler_lambda(struct compiler *c, expr_ty e)
         return 0;
     }
 
+    _Py_DECLARE_STR(anon_lambda, "<lambda>");
     if (!compiler_enter_scope(c, &_Py_STR(anon_lambda), COMPILER_SCOPE_LAMBDA,
                               (void *)e, e->lineno)) {
         return 0;
@@ -5347,6 +5350,7 @@ static int
 compiler_genexp(struct compiler *c, expr_ty e)
 {
     assert(e->kind == GeneratorExp_kind);
+    _Py_DECLARE_STR(anon_genexpr, "<genexpr>");
     return compiler_comprehension(c, e, COMP_GENEXP, &_Py_STR(anon_genexpr),
                                   e->v.GeneratorExp.generators,
                                   e->v.GeneratorExp.elt, NULL);
@@ -5356,6 +5360,7 @@ static int
 compiler_listcomp(struct compiler *c, expr_ty e)
 {
     assert(e->kind == ListComp_kind);
+    _Py_DECLARE_STR(anon_listcomp, "<listcomp>");
     return compiler_comprehension(c, e, COMP_LISTCOMP, &_Py_STR(anon_listcomp),
                                   e->v.ListComp.generators,
                                   e->v.ListComp.elt, NULL);
@@ -5365,6 +5370,7 @@ static int
 compiler_setcomp(struct compiler *c, expr_ty e)
 {
     assert(e->kind == SetComp_kind);
+    _Py_DECLARE_STR(anon_setcomp, "<setcomp>");
     return compiler_comprehension(c, e, COMP_SETCOMP, &_Py_STR(anon_setcomp),
                                   e->v.SetComp.generators,
                                   e->v.SetComp.elt, NULL);
@@ -5375,6 +5381,7 @@ static int
 compiler_dictcomp(struct compiler *c, expr_ty e)
 {
     assert(e->kind == DictComp_kind);
+    _Py_DECLARE_STR(anon_dictcomp, "<dictcomp>");
     return compiler_comprehension(c, e, COMP_DICTCOMP, &_Py_STR(anon_dictcomp),
                                   e->v.DictComp.generators,
                                   e->v.DictComp.key, e->v.DictComp.value);

Python/pythonrun.c

@@ -515,6 +515,7 @@ parse_syntax_error(PyObject *err, PyObject **message, PyObject **filename,
         goto finally;
     if (v == Py_None) {
         Py_DECREF(v);
+        _Py_DECLARE_STR(anon_string, "<string>");
         *filename = &_Py_STR(anon_string);
         Py_INCREF(*filename);
     }
@@ -1562,6 +1563,7 @@ PyRun_StringFlags(const char *str, int start, PyObject *globals,
     if (arena == NULL)
         return NULL;
 
+    _Py_DECLARE_STR(anon_string, "<string>");
     mod = _PyParser_ASTFromString(
             str, &_Py_STR(anon_string), start, flags, arena);