Issue #19512: add _PyUnicode_CompareWithId() function

_PyUnicode_CompareWithId() is faster than PyUnicode_CompareWithASCIIString()
when both strings are equal and interned.

Add also _PyId_builtins identifier for "builtins" common string.

diff --git a/Include/object.h b/Include/object.h
index a36c5be52f50065ba0848b23c36ddaf4d97ad195..27f48a5ba9a73ae0198f6e0eac2cb22c56e25117 100644 (file)
--- a/Include/object.h
+++ b/Include/object.h
@@ -147,9 +147,10 @@ typedef struct _Py_Identifier {
 #define _Py_static_string(varname, value)  static _Py_Identifier varname = _Py_static_string_init(value)
 #define _Py_IDENTIFIER(varname) _Py_static_string(PyId_##varname, #varname)
 
-/* Common identifiers */
-PyAPI_DATA(_Py_Identifier) _PyId_path;
+/* Common identifiers (ex: _PyId_path is the string "path") */
 PyAPI_DATA(_Py_Identifier) _PyId_argv;
+PyAPI_DATA(_Py_Identifier) _PyId_builtins;
+PyAPI_DATA(_Py_Identifier) _PyId_path;
 PyAPI_DATA(_Py_Identifier) _PyId_stdin;
 PyAPI_DATA(_Py_Identifier) _PyId_stdout;
 PyAPI_DATA(_Py_Identifier) _PyId_stderr;

diff --git a/Include/unicodeobject.h b/Include/unicodeobject.h
index 6d830c0f1f3bf1e2d4ab8ecf76dadbed7719990f..20ce41d8310242565bb2920e70c480588618a03d 100644 (file)
--- a/Include/unicodeobject.h
+++ b/Include/unicodeobject.h
@@ -1996,6 +1996,11 @@ PyAPI_FUNC(int) PyUnicode_Compare(
     PyObject *right             /* Right string */
     );
 
+PyAPI_FUNC(int) _PyUnicode_CompareWithId(
+    PyObject *left,             /* Left string */
+    _Py_Identifier *right       /* Right identifier */
+    );
+
 PyAPI_FUNC(int) PyUnicode_CompareWithASCIIString(
     PyObject *left,
     const char *right           /* ASCII-encoded string */

diff --git a/Objects/typeobject.c b/Objects/typeobject.c
index 02ad0b7e576ddb81f54bb58ae4854918eb1d42a7..299c6117c0e48c7afbc6fca72ca12cb244c3d605 100644 (file)
--- a/Objects/typeobject.c
+++ b/Objects/typeobject.c
@@ -345,11 +345,10 @@ type_set_qualname(PyTypeObject *type, PyObject *value, void *context)
 static PyObject *
 type_module(PyTypeObject *type, void *context)
 {
-    PyObject *mod;
     char *s;
 
     if (type->tp_flags & Py_TPFLAGS_HEAPTYPE) {
-        mod = _PyDict_GetItemId(type->tp_dict, &PyId___module__);
+        PyObject *mod = _PyDict_GetItemId(type->tp_dict, &PyId___module__);
         if (!mod) {
             PyErr_Format(PyExc_AttributeError, "__module__");
             return 0;
@@ -358,11 +357,14 @@ type_module(PyTypeObject *type, void *context)
         return mod;
     }
     else {
+        PyObject *name;
         s = strrchr(type->tp_name, '.');
         if (s != NULL)
             return PyUnicode_FromStringAndSize(
                 type->tp_name, (Py_ssize_t)(s - type->tp_name));
-        return PyUnicode_FromString("builtins");
+        name = _PyUnicode_FromId(&_PyId_builtins);
+        Py_XINCREF(name);
+        return name;
     }
 }
 
@@ -712,7 +714,7 @@ type_repr(PyTypeObject *type)
         return NULL;
     }
 
-    if (mod != NULL && PyUnicode_CompareWithASCIIString(mod, "builtins"))
+    if (mod != NULL && _PyUnicode_CompareWithId(mod, &_PyId_builtins))
         rtn = PyUnicode_FromFormat("<class '%U.%U'>", mod, name);
     else
         rtn = PyUnicode_FromFormat("<class '%s'>", type->tp_name);
@@ -2143,7 +2145,7 @@ type_new(PyTypeObject *metatype, PyObject *args, PyObject *kwds)
             if (!valid_identifier(tmp))
                 goto error;
             assert(PyUnicode_Check(tmp));
-            if (PyUnicode_CompareWithASCIIString(tmp, "__dict__") == 0) {
+            if (_PyUnicode_CompareWithId(tmp, &PyId___dict__) == 0) {
                 if (!may_add_dict || add_dict) {
                     PyErr_SetString(PyExc_TypeError,
                         "__dict__ slot disallowed: "
@@ -2174,7 +2176,7 @@ type_new(PyTypeObject *metatype, PyObject *args, PyObject *kwds)
         for (i = j = 0; i < nslots; i++) {
             tmp = PyTuple_GET_ITEM(slots, i);
             if ((add_dict &&
-                 PyUnicode_CompareWithASCIIString(tmp, "__dict__") == 0) ||
+                 _PyUnicode_CompareWithId(tmp, &PyId___dict__) == 0) ||
                 (add_weak &&
                  PyUnicode_CompareWithASCIIString(tmp, "__weakref__") == 0))
                 continue;
@@ -3183,7 +3185,7 @@ object_repr(PyObject *self)
         Py_XDECREF(mod);
         return NULL;
     }
-    if (mod != NULL && PyUnicode_CompareWithASCIIString(mod, "builtins"))
+    if (mod != NULL && _PyUnicode_CompareWithId(mod, &_PyId_builtins))
         rtn = PyUnicode_FromFormat("<%U.%U object at %p>", mod, name, self);
     else
         rtn = PyUnicode_FromFormat("<%s object at %p>",
@@ -6336,8 +6338,8 @@ super_getattro(PyObject *self, PyObject *name)
         /* We want __class__ to return the class of the super object
            (i.e. super, or a subclass), not the class of su->obj. */
         skip = (PyUnicode_Check(name) &&
-            PyUnicode_GET_LENGTH(name) == 9 &&
-            PyUnicode_CompareWithASCIIString(name, "__class__") == 0);
+                PyUnicode_GET_LENGTH(name) == 9 &&
+                _PyUnicode_CompareWithId(name, &PyId___class__) == 0);
     }
 
     if (!skip) {
@@ -6543,8 +6545,7 @@ super_init(PyObject *self, PyObject *args, PyObject *kwds)
         for (i = 0; i < n; i++) {
             PyObject *name = PyTuple_GET_ITEM(co->co_freevars, i);
             assert(PyUnicode_Check(name));
-            if (!PyUnicode_CompareWithASCIIString(name,
-                                                  "__class__")) {
+            if (!_PyUnicode_CompareWithId(name, &PyId___class__)) {
                 Py_ssize_t index = co->co_nlocals +
                     PyTuple_GET_SIZE(co->co_cellvars) + i;
                 PyObject *cell = f->f_localsplus[index];

diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index 574b57a259f2c41ef172a51e916c69fd83497c6b..4ae73771f238fddba6d58ad1427048cb2537c346 100644 (file)
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -10565,6 +10565,15 @@ PyUnicode_Compare(PyObject *left, PyObject *right)
     return -1;
 }
 
+int
+_PyUnicode_CompareWithId(PyObject *left, _Py_Identifier *right)
+{
+    PyObject *right_str = _PyUnicode_FromId(right);   /* borrowed */
+    if (right_str == NULL)
+        return -1;
+    return PyUnicode_Compare(left, right_str);
+}
+
 int
 PyUnicode_CompareWithASCIIString(PyObject* uni, const char* str)
 {

diff --git a/Python/errors.c b/Python/errors.c
index b92911a034433bf0a6ccfd19b887d6e1a33ea07a..93e47246803bd5c24b4fa5656d216d6418f7937f 100644 (file)
--- a/Python/errors.c
+++ b/Python/errors.c
@@ -878,7 +878,7 @@ PyErr_WriteUnraisable(PyObject *obj)
             goto done;
     }
     else {
-        if (PyUnicode_CompareWithASCIIString(moduleName, "builtins") != 0) {
+        if (_PyUnicode_CompareWithId(moduleName, &_PyId_builtins) != 0) {
             if (PyFile_WriteObject(moduleName, f, Py_PRINT_RAW) < 0)
                 goto done;
             if (PyFile_WriteString(".", f) < 0)

diff --git a/Python/pythonrun.c b/Python/pythonrun.c
index e0c863811a4bb349c3618515f9d1b58ca881f16d..be41de67f2a12f928341199b0cfe69f11dd98082 100644 (file)
--- a/Python/pythonrun.c
+++ b/Python/pythonrun.c
@@ -37,6 +37,7 @@
 
 /* Common identifiers */
 _Py_Identifier _PyId_argv = _Py_static_string_init("argv");
+_Py_Identifier _PyId_builtins = _Py_static_string_init("builtins");
 _Py_Identifier _PyId_path = _Py_static_string_init("path");
 _Py_Identifier _PyId_stdin = _Py_static_string_init("stdin");
 _Py_Identifier _PyId_stdout = _Py_static_string_init("stdout");
@@ -1928,7 +1929,7 @@ print_exception(PyObject *f, PyObject *value)
             err = PyFile_WriteString("<unknown>", f);
         }
         else {
-            if (PyUnicode_CompareWithASCIIString(moduleName, "builtins") != 0)
+            if (_PyUnicode_CompareWithId(moduleName, &_PyId_builtins) != 0)
             {
                 err = PyFile_WriteObject(moduleName, f, Py_PRINT_RAW);
                 err += PyFile_WriteString(".", f);