bpo-47000: Add locale.getencoding()

Doc/glossary.rst

@@ -710,8 +710,9 @@ Glossary
 
       On Windows, it is the ANSI code page (ex: ``cp1252``).
 
-      ``locale.getpreferredencoding(False)`` can be used to get the locale
-      encoding.
+      On Android and VxWorks, return ``"UTF-8"``.
+
+      ``locale.getencoding()`` can be used to get the locale encoding.
 
       Python uses the :term:`filesystem encoding and error handler` to convert
       between Unicode filenames and bytes filenames.

Doc/library/locale.rst

@@ -334,10 +334,26 @@ The :mod:`locale` module defines the following exception and functions:
    locale. See also the :term:`filesystem encoding and error handler`.
 
    .. versionchanged:: 3.7
-      The function now always returns ``UTF-8`` on Android or if the
+      The function now always returns ``"UTF-8"`` on Android or if the
       :ref:`Python UTF-8 Mode <utf8-mode>` is enabled.
 
 
+.. function:: getencoding()
+
+   Get the current :term:`locale encoding`:
+
+   * On Android and VxWorks, return ``"UTF-8"``.
+   * On Unix, return the encoding of the current :data:`LC_CTYPE` locale.
+     Return ``"UTF-8"`` if ``nl_langinfo(CODESET)`` returns an empty string:
+     for example, if the current LC_CTYPE locale is not supported.
+   * On Windows, return the ANSI code page.
+
+   This function is same to ``getpreferredencoding(False)`` except this
+   function ignore the :ref:`UTF-8 Mode <utf8-mode>`.
+
+   .. versionadded:: 3.11
+
+
 .. function:: normalize(localename)
 
    Returns a normalized locale code for the given locale name.  The returned locale

Doc/whatsnew/3.11.rst

@@ -274,6 +274,12 @@ inspect
 * Add :func:`inspect.ismethodwrapper` for checking if the type of an object is a
   :class:`~types.MethodWrapperType`. (Contributed by Hakan Çelik in :issue:`29418`.)
 
+locale
+------
+
+* Add :func:`locale.getencoding` that is same to
+  ``locale.getpreferredencoding(False)`` but ignores :ref:`UTF-8 Mode <utf8-mode>`.
+
 math
 ----
 

Lib/locale.py

@@ -28,7 +28,7 @@
            "setlocale", "resetlocale", "localeconv", "strcoll", "strxfrm",
            "str", "atof", "atoi", "format", "format_string", "currency",
            "normalize", "LC_CTYPE", "LC_COLLATE", "LC_TIME", "LC_MONETARY",
-           "LC_NUMERIC", "LC_ALL", "CHAR_MAX"]
+           "LC_NUMERIC", "LC_ALL", "CHAR_MAX", "getencoding"]
 
 def _strcoll(a,b):
     """ strcoll(string,string) -> int.
@@ -637,27 +637,27 @@ def resetlocale(category=LC_ALL):
 
 
 try:
-    from _locale import _get_locale_encoding
+    from _locale import getencoding
 except ImportError:
-    def _get_locale_encoding():
+    def getencoding():
         if hasattr(sys, 'getandroidapilevel'):
             # On Android langinfo.h and CODESET are missing, and UTF-8 is
             # always used in mbstowcs() and wcstombs().
             return 'UTF-8'
-        if sys.flags.utf8_mode:
-            return 'UTF-8'
         encoding = getdefaultlocale()[1]
         if encoding is None:
-            # LANG not set, default conservatively to ASCII
-            encoding = 'ascii'
+            # LANG not set, default to UTF-8
+            encoding = 'UTF-8'
         return encoding
 
 try:
     CODESET
 except NameError:
     def getpreferredencoding(do_setlocale=True):
         """Return the charset that the user is likely using."""
-        return _get_locale_encoding()
+        if sys.flags.utf8_mode:
+            return 'UTF-8'
+        return getencoding()
 else:
     # On Unix, if CODESET is available, use that.
     def getpreferredencoding(do_setlocale=True):
@@ -667,15 +667,15 @@ def getpreferredencoding(do_setlocale=True):
             return 'UTF-8'
 
         if not do_setlocale:
-            return _get_locale_encoding()
+            return getencoding()
 
         old_loc = setlocale(LC_CTYPE)
         try:
             try:
                 setlocale(LC_CTYPE, "")
             except Error:
                 pass
-            return _get_locale_encoding()
+            return getencoding()
         finally:
             setlocale(LC_CTYPE, old_loc)
 

Misc/NEWS.d/next/Library/2022-04-06-11-54-53.bpo-47000.2nmAR1.rst

@@ -0,0 +1 @@
+Add :func:`locale.getencoding`.

Modules/_io/textio.c

@@ -1145,7 +1145,14 @@ _io_TextIOWrapper___init___impl(textio *self, PyObject *buffer,
         }
     }
     if (encoding == NULL && self->encoding == NULL) {
-        self->encoding = _Py_GetLocaleEncodingObject();
+        if (_PyRuntime.preconfig.utf8_mode) {
+            _Py_DECLARE_STR(utf_8, "utf-8");
+            self->encoding = &_Py_STR(utf_8);
+            Py_INCREF(self->encoding);
+        }
+        else {
+            self->encoding = _Py_GetLocaleEncodingObject();
+        }
         if (self->encoding == NULL) {
             goto error;
         }

Modules/_localemodule.c

@@ -773,14 +773,14 @@ _locale_bind_textdomain_codeset_impl(PyObject *module, const char *domain,
 
 
 /*[clinic input]
-_locale._get_locale_encoding
+_locale.getencoding
 
 Get the current locale encoding.
 [clinic start generated code]*/
 
 static PyObject *
-_locale__get_locale_encoding_impl(PyObject *module)
-/*[clinic end generated code: output=e8e2f6f6f184591a input=513d9961d2f45c76]*/
+_locale_getencoding_impl(PyObject *module)
+/*[clinic end generated code: output=86b326b971872e46 input=6503d11e5958b360]*/
 {
     return _Py_GetLocaleEncodingObject();
 }
@@ -811,7 +811,7 @@ static struct PyMethodDef PyLocale_Methods[] = {
     _LOCALE_BIND_TEXTDOMAIN_CODESET_METHODDEF
 #endif
 #endif
-    _LOCALE__GET_LOCALE_ENCODING_METHODDEF
+    _LOCALE_GETENCODING_METHODDEF
   {NULL, NULL}
 };
 

Python/fileutils.c

@@ -93,6 +93,12 @@ _Py_device_encoding(int fd)
 
     return PyUnicode_FromFormat("cp%u", (unsigned int)cp);
 #else
+    if (_PyRuntime.preconfig.utf8_mode) {
+        _Py_DECLARE_STR(utf_8, "utf-8");
+        PyObject *encoding = &_Py_STR(utf_8);
+        Py_INCREF(encoding);
+        return encoding;
+    }
     return _Py_GetLocaleEncodingObject();
 #endif
 }
@@ -890,10 +896,6 @@ _Py_GetLocaleEncoding(void)
     // and UTF-8 is always used in mbstowcs() and wcstombs().
     return _PyMem_RawWcsdup(L"UTF-8");
 #else
-    const PyPreConfig *preconfig = &_PyRuntime.preconfig;
-    if (preconfig->utf8_mode) {
-        return _PyMem_RawWcsdup(L"UTF-8");
-    }
 
 #ifdef MS_WINDOWS
     wchar_t encoding[23];

Python/initconfig.c

@@ -1779,7 +1779,13 @@ static PyStatus
 config_get_locale_encoding(PyConfig *config, const PyPreConfig *preconfig,
                            wchar_t **locale_encoding)
 {
-    wchar_t *encoding = _Py_GetLocaleEncoding();
+    wchar_t *encoding;
+    if (preconfig->utf8_mode) {
+        encoding = _PyMem_RawWcsdup(L"UTF-8");
+    }
+    else {
+        encoding = _Py_GetLocaleEncoding();
+    }
     if (encoding == NULL) {
         return _PyStatus_NO_MEMORY();
     }