bpo-43687: Py_Initialize() creates singletons earlier (GH-25147)

Reorganize pycore_interp_init() to initialize singletons before the
the first PyType_Ready() call. Fix an issue when Python is configured
using --without-doc-strings.

Include/internal/pycore_long.h

@@ -18,8 +18,8 @@ static inline PyObject* __PyLong_GetSmallInt_internal(int value)
     assert(-_PY_NSMALLNEGINTS <= value && value < _PY_NSMALLPOSINTS);
     size_t index = _PY_NSMALLNEGINTS + value;
     PyObject *obj = (PyObject*)interp->small_ints[index];
-    // _PyLong_GetZero() and _PyLong_GetOne() must not be called
-    // before _PyLong_Init() nor after _PyLong_Fini()
+    // _PyLong_GetZero(), _PyLong_GetOne() and get_small_int() must not be
+    // called before _PyLong_Init() nor after _PyLong_Fini().
     assert(obj != NULL);
     return obj;
 }

Include/internal/pycore_pylifecycle.h

@@ -50,9 +50,11 @@ PyAPI_FUNC(int) _Py_IsLocaleCoercionTarget(const char *ctype_loc);
 /* Various one-time initializers */
 
 extern PyStatus _PyUnicode_Init(PyInterpreterState *interp);
+extern PyStatus _PyUnicode_InitTypes(void);
 extern PyStatus _PyBytes_Init(PyInterpreterState *interp);
 extern int _PyStructSequence_Init(void);
 extern int _PyLong_Init(PyInterpreterState *interp);
+extern int _PyLong_InitTypes(void);
 extern PyStatus _PyTuple_Init(PyInterpreterState *interp);
 extern PyStatus _PyFaulthandler_Init(int enable);
 extern int _PyTraceMalloc_Init(int enable);
@@ -64,9 +66,10 @@ extern PyStatus _PySys_ReadPreinitWarnOptions(PyWideStringList *options);
 extern PyStatus _PySys_ReadPreinitXOptions(PyConfig *config);
 extern int _PySys_UpdateConfig(PyThreadState *tstate);
 extern PyStatus _PyExc_Init(PyInterpreterState *interp);
-extern PyStatus _PyErr_Init(void);
+extern PyStatus _PyErr_InitTypes(void);
 extern PyStatus _PyBuiltins_AddExceptions(PyObject * bltinmod);
-extern int _PyFloat_Init(void);
+extern void _PyFloat_Init(void);
+extern int _PyFloat_InitTypes(void);
 extern PyStatus _Py_HashRandomization_Init(const PyConfig *);
 
 extern PyStatus _PyTypes_Init(void);

Objects/floatobject.c

@@ -1968,7 +1968,7 @@ PyTypeObject PyFloat_Type = {
     .tp_vectorcall = (vectorcallfunc)float_vectorcall,
 };
 
-int
+void
 _PyFloat_Init(void)
 {
     /* We attempt to determine if this machine is using IEEE
@@ -2016,14 +2016,18 @@ _PyFloat_Init(void)
 
     double_format = detected_double_format;
     float_format = detected_float_format;
+}
 
+int
+_PyFloat_InitTypes(void)
+{
     /* Init float info */
     if (FloatInfoType.tp_name == NULL) {
         if (PyStructSequence_InitType2(&FloatInfoType, &floatinfo_desc) < 0) {
-            return 0;
+            return -1;
         }
     }
-    return 1;
+    return 0;
 }
 
 void

Objects/longobject.c

@@ -5719,17 +5719,20 @@ _PyLong_Init(PyInterpreterState *interp)
 
         interp->small_ints[i] = v;
     }
+    return 0;
+}
 
-    if (_Py_IsMainInterpreter(interp)) {
-        /* initialize int_info */
-        if (Int_InfoType.tp_name == NULL) {
-            if (PyStructSequence_InitType2(&Int_InfoType, &int_info_desc) < 0) {
-                return 0;
-            }
+
+int
+_PyLong_InitTypes(void)
+{
+    /* initialize int_info */
+    if (Int_InfoType.tp_name == NULL) {
+        if (PyStructSequence_InitType2(&Int_InfoType, &int_info_desc) < 0) {
+            return -1;
         }
     }
-
-    return 1;
+    return 0;
 }
 
 void

Objects/structseq.c

@@ -579,7 +579,8 @@ int _PyStructSequence_Init(void)
     if (_PyUnicode_FromId(&PyId_n_sequence_fields) == NULL
         || _PyUnicode_FromId(&PyId_n_fields) == NULL
         || _PyUnicode_FromId(&PyId_n_unnamed_fields) == NULL)
+    {
         return -1;
-
+    }
     return 0;
 }

Objects/unicodeobject.c

@@ -15676,18 +15676,6 @@ PyTypeObject PyUnicode_Type = {
 PyStatus
 _PyUnicode_Init(PyInterpreterState *interp)
 {
-    /* XXX - move this array to unicodectype.c ? */
-    const Py_UCS2 linebreak[] = {
-        0x000A, /* LINE FEED */
-        0x000D, /* CARRIAGE RETURN */
-        0x001C, /* FILE SEPARATOR */
-        0x001D, /* GROUP SEPARATOR */
-        0x001E, /* RECORD SEPARATOR */
-        0x0085, /* NEXT LINE */
-        0x2028, /* LINE SEPARATOR */
-        0x2029, /* PARAGRAPH SEPARATOR */
-    };
-
     struct _Py_unicode_state *state = &interp->unicode;
     if (unicode_create_empty_string_singleton(state) < 0) {
         return _PyStatus_NO_MEMORY();
@@ -15695,23 +15683,39 @@ _PyUnicode_Init(PyInterpreterState *interp)
 
     if (_Py_IsMainInterpreter(interp)) {
         /* initialize the linebreak bloom filter */
+        const Py_UCS2 linebreak[] = {
+            0x000A, /* LINE FEED */
+            0x000D, /* CARRIAGE RETURN */
+            0x001C, /* FILE SEPARATOR */
+            0x001D, /* GROUP SEPARATOR */
+            0x001E, /* RECORD SEPARATOR */
+            0x0085, /* NEXT LINE */
+            0x2028, /* LINE SEPARATOR */
+            0x2029, /* PARAGRAPH SEPARATOR */
+        };
         bloom_linebreak = make_bloom_mask(
             PyUnicode_2BYTE_KIND, linebreak,
             Py_ARRAY_LENGTH(linebreak));
+    }
 
-        if (PyType_Ready(&PyUnicode_Type) < 0) {
-            return _PyStatus_ERR("Can't initialize unicode type");
-        }
+    return _PyStatus_OK();
+}
 
-        if (PyType_Ready(&EncodingMapType) < 0) {
-             return _PyStatus_ERR("Can't initialize encoding map type");
-        }
-        if (PyType_Ready(&PyFieldNameIter_Type) < 0) {
-            return _PyStatus_ERR("Can't initialize field name iterator type");
-        }
-        if (PyType_Ready(&PyFormatterIter_Type) < 0) {
-            return _PyStatus_ERR("Can't initialize formatter iter type");
-        }
+
+PyStatus
+_PyUnicode_InitTypes(void)
+{
+    if (PyType_Ready(&PyUnicode_Type) < 0) {
+        return _PyStatus_ERR("Can't initialize unicode type");
+    }
+    if (PyType_Ready(&EncodingMapType) < 0) {
+         return _PyStatus_ERR("Can't initialize encoding map type");
+    }
+    if (PyType_Ready(&PyFieldNameIter_Type) < 0) {
+        return _PyStatus_ERR("Can't initialize field name iterator type");
+    }
+    if (PyType_Ready(&PyFormatterIter_Type) < 0) {
+        return _PyStatus_ERR("Can't initialize formatter iter type");
     }
     return _PyStatus_OK();
 }

Python/errors.c

@@ -1192,7 +1192,7 @@ static PyStructSequence_Desc UnraisableHookArgs_desc = {
 
 
 PyStatus
-_PyErr_Init(void)
+_PyErr_InitTypes(void)
 {
     if (UnraisableHookArgsType.tp_name == NULL) {
         if (PyStructSequence_InitType2(&UnraisableHookArgsType,

Python/pylifecycle.c

@@ -628,38 +628,16 @@ pycore_create_interpreter(_PyRuntimeState *runtime,
 
 
 static PyStatus
-pycore_init_types(PyInterpreterState *interp)
+pycore_init_singletons(PyInterpreterState *interp)
 {
     PyStatus status;
-    int is_main_interp = _Py_IsMainInterpreter(interp);
 
-    status = _PyGC_Init(interp);
-    if (_PyStatus_EXCEPTION(status)) {
-        return status;
-    }
-
-    // Create the empty tuple singleton. It must be created before the first
-    // PyType_Ready() call since PyType_Ready() creates tuples, for tp_bases
-    // for example.
-    status = _PyTuple_Init(interp);
-    if (_PyStatus_EXCEPTION(status)) {
-        return status;
-    }
-
-    if (is_main_interp) {
-        status = _PyTypes_Init();
-        if (_PyStatus_EXCEPTION(status)) {
-            return status;
-        }
-    }
-
-    if (!_PyLong_Init(interp)) {
+    if (_PyLong_Init(interp) < 0) {
         return _PyStatus_ERR("can't init longs");
     }
 
-    status = _PyUnicode_Init(interp);
-    if (_PyStatus_EXCEPTION(status)) {
-        return status;
+    if (_Py_IsMainInterpreter(interp)) {
+        _PyFloat_Init();
     }
 
     status = _PyBytes_Init(interp);
@@ -667,22 +645,58 @@ pycore_init_types(PyInterpreterState *interp)
         return status;
     }
 
+    status = _PyUnicode_Init(interp);
+    if (_PyStatus_EXCEPTION(status)) {
+        return status;
+    }
+
+    status = _PyTuple_Init(interp);
+    if (_PyStatus_EXCEPTION(status)) {
+        return status;
+    }
+
+    return _PyStatus_OK();
+}
+
+
+static PyStatus
+pycore_init_types(PyInterpreterState *interp)
+{
+    PyStatus status;
+    int is_main_interp = _Py_IsMainInterpreter(interp);
+
+    if (is_main_interp) {
+        if (_PyStructSequence_Init() < 0) {
+            return _PyStatus_ERR("can't initialize structseq");
+        }
+
+        status = _PyTypes_Init();
+        if (_PyStatus_EXCEPTION(status)) {
+            return status;
+        }
+
+        if (_PyLong_InitTypes() < 0) {
+            return _PyStatus_ERR("can't init int type");
+        }
+
+        status = _PyUnicode_InitTypes();
+        if (_PyStatus_EXCEPTION(status)) {
+            return status;
+        }
+    }
+
+    if (is_main_interp) {
+        if (_PyFloat_InitTypes() < 0) {
+            return _PyStatus_ERR("can't init float");
+        }
+    }
+
     status = _PyExc_Init(interp);
     if (_PyStatus_EXCEPTION(status)) {
         return status;
     }
 
-    if (is_main_interp) {
-        if (!_PyFloat_Init()) {
-            return _PyStatus_ERR("can't init float");
-        }
-
-        if (_PyStructSequence_Init() < 0) {
-            return _PyStatus_ERR("can't initialize structseq");
-        }
-    }
-
-    status = _PyErr_Init();
+    status = _PyErr_InitTypes();
     if (_PyStatus_EXCEPTION(status)) {
         return status;
     }
@@ -693,22 +707,15 @@ pycore_init_types(PyInterpreterState *interp)
         }
     }
 
-    if (_PyWarnings_InitState(interp) < 0) {
-        return _PyStatus_ERR("can't initialize warnings");
-    }
-
-    status = _PyAtExit_Init(interp);
-    if (_PyStatus_EXCEPTION(status)) {
-        return status;
-    }
-
     return _PyStatus_OK();
 }
 
 
 static PyStatus
-pycore_init_builtins(PyInterpreterState *interp)
+pycore_init_builtins(PyThreadState *tstate)
 {
+    PyInterpreterState *interp = tstate->interp;
+
     PyObject *bimod = _PyBuiltin_Init(interp);
     if (bimod == NULL) {
         goto error;
@@ -744,6 +751,7 @@ pycore_init_builtins(PyInterpreterState *interp)
     }
     interp->import_func = Py_NewRef(import_func);
 
+    assert(!_PyErr_Occurred(tstate));
     return _PyStatus_OK();
 
 error:
@@ -755,29 +763,49 @@ error:
 static PyStatus
 pycore_interp_init(PyThreadState *tstate)
 {
+    PyInterpreterState *interp = tstate->interp;
     PyStatus status;
     PyObject *sysmod = NULL;
 
-    status = pycore_init_types(tstate->interp);
+    // Create singletons before the first PyType_Ready() call, since
+    // PyType_Ready() uses singletons like the Unicode empty string (tp_doc)
+    // and the empty tuple singletons (tp_bases).
+    status = pycore_init_singletons(interp);
+    if (_PyStatus_EXCEPTION(status)) {
+        return status;
+    }
+
+    // The GC must be initialized before the first GC collection.
+    status = _PyGC_Init(interp);
+    if (_PyStatus_EXCEPTION(status)) {
+        return status;
+    }
+
+    status = pycore_init_types(interp);
     if (_PyStatus_EXCEPTION(status)) {
         goto done;
     }
 
+    if (_PyWarnings_InitState(interp) < 0) {
+        return _PyStatus_ERR("can't initialize warnings");
+    }
+
+    status = _PyAtExit_Init(interp);
+    if (_PyStatus_EXCEPTION(status)) {
+        return status;
+    }
+
     status = _PySys_Create(tstate, &sysmod);
     if (_PyStatus_EXCEPTION(status)) {
         goto done;
     }
 
-    assert(!_PyErr_Occurred(tstate));
-
-    status = pycore_init_builtins(tstate->interp);
+    status = pycore_init_builtins(tstate);
     if (_PyStatus_EXCEPTION(status)) {
         goto done;
     }
 
-    assert(!_PyErr_Occurred(tstate));
-
-    const PyConfig *config = _PyInterpreterState_GetConfig(tstate->interp);
+    const PyConfig *config = _PyInterpreterState_GetConfig(interp);
     if (config->_install_importlib) {
         /* This call sets up builtin and frozen import support */
         if (init_importlib(tstate, sysmod) < 0) {