python · vstinner · Mar 19, 2021 · Mar 19, 2021
diff --git a/Doc/glossary.rst b/Doc/glossary.rst
@@ -705,12 +705,15 @@ Glossary
       This issue can be solved with locks or by using the EAFP approach.
 
    locale encoding
-      On Unix, it is the encoding of the LC_CTYPE locale. It can be set with
-      ``locale.setlocale(locale.LC_CTYPE, new_locale)``.
+      On Unix, it is the encoding of the :const:`LC_CTYPE <locale.LC_CTYPE>`
+      locale. It can be set with ``locale.setlocale(locale.LC_CTYPE,
+      new_locale)``.
 
       On Windows, it is is the ANSI code page (ex: ``cp1252``).
 
-      ``locale.getpreferredencoding(False)`` can be used to get the locale
+      Use :func:`locale.getpreferredencoding(False)
+      <locale.getpreferredencoding>` to get the locale encoding and
+      :func:`locale.get_current_locale_encoding` to get the *current* locale
       encoding.
 
       Python uses the :term:`filesystem encoding and error handler` to convert

diff --git a/Doc/library/locale.rst b/Doc/library/locale.rst
@@ -302,6 +302,24 @@ The :mod:`locale` module defines the following exception and functions:
    determined.
 
 
+.. function:: get_current_locale_encoding()
+
+   Get the current :term:`locale encoding`:
+
+   * On Windows, return the current ANSI code page (ex: ``"cp1252"``) for the
+     operating system.
+   * Return ``"UTF-8"`` if ``nl_langinfo(CODESET)`` returns an empty string.
+   * Otherwise, return ``nl_langinfo(CODESET)`` result.
+
+   On Unix, the current locale encoding is the encoding of the
+   :const:`LC_CTYPE` locale.
+
+   Use :func:`locale.getpreferredencoding(False) <locale.getpreferredencoding>`
+   to get the locale encoding.
+
+   .. versionadded:: 3.10
+
+
 .. function:: getlocale(category=LC_CTYPE)
 
    Returns the current setting for the given locale category as sequence containing
@@ -331,6 +349,9 @@ The :mod:`locale` module defines the following exception and functions:
    The :ref:`Python preinitialization <c-preinit>` configures the LC_CTYPE
    locale. See also the :term:`filesystem encoding and error handler`.
 
+   Use :func:`locale.get_current_locale_encoding` to get the *current* locale
+   encoding.
+
    .. versionchanged:: 3.7
       The function now always returns ``UTF-8`` on Android or if the
       :ref:`Python UTF-8 Mode <utf8-mode>` is enabled.

diff --git a/Doc/library/sys.rst b/Doc/library/sys.rst
@@ -614,7 +614,7 @@ always available.
 .. function:: getdefaultencoding()
 
    Return the name of the current default string encoding used by the Unicode
-   implementation.
+   implementation: ``"utf-8"``.
 
 
 .. function:: getdlopenflags()

@@ -728,6 +728,13 @@ linecache
 When a module does not define ``__loader__``, fall back to ``__spec__.loader``.
 (Contributed by Brett Cannon in :issue:`42133`.)
 
+locale
+------
+
+Added :func:`locale.get_current_locale_encoding` to get the current
+:term:`locale encoding`.
+(Contributed by Victor Stinner in :issue:`43552`.)
+
 os
 --
 

diff --git a/Include/internal/pycore_fileutils.h b/Include/internal/pycore_fileutils.h
@@ -51,6 +51,7 @@ PyAPI_FUNC(int) _Py_GetLocaleconvNumeric(
 PyAPI_FUNC(void) _Py_closerange(int first, int last);
 
 PyAPI_FUNC(wchar_t*) _Py_GetLocaleEncoding(void);
+PyAPI_FUNC(wchar_t*) _Py_GetCurrentLocaleEncoding(void);
 PyAPI_FUNC(PyObject*) _Py_GetLocaleEncodingObject(void);
 
 #ifdef __cplusplus

diff --git a/Lib/locale.py b/Lib/locale.py
@@ -620,21 +620,44 @@ def resetlocale(category=LC_ALL):
     _setlocale(category, _build_localename(getdefaultlocale()))
 
 
+try:
+    from _locale import get_current_locale_encoding
+except ImportError:
+    try:
+        from _locale import nl_langinfo, CODESET
+
+        # nl_langinfo(CODESET) implementation
+        def get_current_locale_encoding():
+            result = _locale.nl_langinfo(_locale.CODESET)
+            if not result:
+                # On macOS, nl_langinfo(CODESET) can return an empty string
+                # when the setting has an invalid value.  Default to UTF-8 in
+                # that case because UTF-8 is the default charset on macOS and
+                # the caller expects a non-empty string.
+                result = 'UTF-8'
+            return result
+    except ImportError:
+        # getdefaultlocale() implementation.
+        # On Windows, _locale.getdefaultlocale()[1] is the ANSI code page.
+        def get_current_locale_encoding():
+            encoding = getdefaultlocale()[1]
+            if encoding is None:
+                # LANG not set, default conservatively to ASCII
+                encoding = 'ascii'
+            return encoding
+
 try:
     from _locale import _get_locale_encoding
 except ImportError:
     def _get_locale_encoding():
-        if hasattr(sys, 'getandroidapilevel'):
+        if hasattr(sys, 'getandroidapilevel') or sys.platform == 'vxworks':
             # On Android langinfo.h and CODESET are missing, and UTF-8 is
             # always used in mbstowcs() and wcstombs().
+            # Always use UTF-8 on VxWorks.
             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'
-        return encoding
+        return get_current_locale_encoding()
 
 try:
     CODESET

diff --git a/Lib/test/test_locale.py b/Lib/test/test_locale.py
@@ -536,6 +536,12 @@ def test_getpreferredencoding(self):
             # If encoding non-empty, make sure it is valid
             codecs.lookup(enc)
 
+    def test_get_current_locale_encoding(self):
+        encoding = locale.get_current_locale_encoding()
+        self.assertIsInstance(encoding, str)
+        self.assertGreater(len(encoding), 0, encoding)
+        codecs.lookup(encoding)
+
     def test_strcoll_3303(self):
         # test crasher from bug #3303
         self.assertRaises(TypeError, locale.strcoll, "a", None)

diff --git a/Misc/NEWS.d/next/Library/2021-03-19-10-43-54.bpo-43552.10M8Dt.rst b/Misc/NEWS.d/next/Library/2021-03-19-10-43-54.bpo-43552.10M8Dt.rst
@@ -0,0 +1,2 @@
+Added :func:`locale.get_current_locale_encoding` to get the current
+:term:`locale encoding`. Patch by Victor Stinner.
diff --git a/Modules/_localemodule.c b/Modules/_localemodule.c
@@ -772,15 +772,47 @@ _locale_bind_textdomain_codeset_impl(PyObject *module, const char *domain,
 #endif  // HAVE_LIBINTL_H
 
 
+/*[clinic input]
+_locale.get_current_locale_encoding
+
+Get the current locale encoding:
+
+* On Windows, return the current ANSI code page (ex: ``"cp1252"``)
+  for the operating system.
+* Return "UTF-8" if nl_langinfo(CODESET) returns an empty string.
+* Otherwise, return nl_langinfo(CODESET) result.
+[clinic start generated code]*/
+
+static PyObject *
+_locale_get_current_locale_encoding_impl(PyObject *module)
+/*[clinic end generated code: output=fce82957b117a446 input=2c7a800e7cf93287]*/
+{
+    wchar_t *encoding = _Py_GetCurrentLocaleEncoding();
+    if (encoding == NULL) {
+        PyErr_NoMemory();
+        return NULL;
+    }
+
+    PyObject *str = PyUnicode_FromWideChar(encoding, -1);
+    PyMem_RawFree(encoding);
+    return str;
+}
+
+
 /*[clinic input]
 _locale._get_locale_encoding
 
-Get the current locale encoding.
+Get the locale encoding:
+
+* "UTF-8" on Android and VxWorks;
+* "UTF-8" if the Python UTF-8 Mode is enabled;
+* ANSI code page on Windows;
+* nl_langinfo(CODESET) otherwise.
 [clinic start generated code]*/
 
 static PyObject *
 _locale__get_locale_encoding_impl(PyObject *module)
-/*[clinic end generated code: output=e8e2f6f6f184591a input=513d9961d2f45c76]*/
+/*[clinic end generated code: output=e8e2f6f6f184591a input=4d3ed54cd5278cf2]*/
 {
     return _Py_GetLocaleEncodingObject();
 }
@@ -812,6 +844,7 @@ static struct PyMethodDef PyLocale_Methods[] = {
 #endif
 #endif
     _LOCALE__GET_LOCALE_ENCODING_METHODDEF
+    _LOCALE_GET_CURRENT_LOCALE_ENCODING_METHODDEF
   {NULL, NULL}
 };
 

diff --git a/Modules/clinic/_localemodule.c.h b/Modules/clinic/_localemodule.c.h
diff --git a/Python/fileutils.c b/Python/fileutils.c
@@ -857,30 +857,19 @@ _Py_EncodeLocaleEx(const wchar_t *text, char **str,
 }
 
 
-// Get the current locale encoding name:
+// Get the current locale encoding:
 //
-// - Return "UTF-8" if _Py_FORCE_UTF8_LOCALE macro is defined (ex: on Android)
-// - Return "UTF-8" if the UTF-8 Mode is enabled
-// - On Windows, return the ANSI code page (ex: "cp1250")
-// - Return "UTF-8" if nl_langinfo(CODESET) returns an empty string.
-// - Otherwise, return nl_langinfo(CODESET).
+// * On Windows, return the current ANSI code page (ex: ``"cp1252"``)
+//   for the operating system.
+// * Return "UTF-8" if nl_langinfo(CODESET) returns an empty string.
+// * Otherwise, return nl_langinfo(CODESET) result.
 //
 // Return NULL on memory allocation failure.
 //
-// See also config_get_locale_encoding()
+// Result must be freed by PyMem_RawFree().
 wchar_t*
-_Py_GetLocaleEncoding(void)
+_Py_GetCurrentLocaleEncoding(void)
 {
-#ifdef _Py_FORCE_UTF8_LOCALE
-    // On Android langinfo.h and CODESET are missing,
-    // 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];
     unsigned int ansi_codepage = GetACP();
@@ -904,6 +893,34 @@ _Py_GetLocaleEncoding(void)
     return wstr;
 #endif  // !MS_WINDOWS
 
+}
+
+
+// Get the locale encoding:
+//
+// - Return "UTF-8" if _Py_FORCE_UTF8_LOCALE macro is defined (ex: on Android)
+// - Return "UTF-8" if the UTF-8 Mode is enabled
+// - Return _Py_GetCurrentLocaleEncoding() otherwise.
+//
+// Return NULL on memory allocation failure.
+//
+// Result must be freed by PyMem_RawFree().
+//
+// See also config_get_locale_encoding()
+wchar_t*
+_Py_GetLocaleEncoding(void)
+{
+#ifdef _Py_FORCE_UTF8_LOCALE
+    // On Android langinfo.h and CODESET are missing,
+    // 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");
+    }
+
+    return _Py_GetCurrentLocaleEncoding();
 #endif  // !_Py_FORCE_UTF8_LOCALE
 }
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		Added :func:`locale.get_current_locale_encoding` to get the current
		:term:`locale encoding`. Patch by Victor Stinner.