Commit

Author: StanFromIreland

Lib/test/test_c_locale_coercion.py

@@ -478,6 +478,31 @@ def test_PYTHONCOERCECLOCALE_set_to_one(self):
                              text=True)
         self.assertEqual(cmd.stdout.rstrip(), loc)
 
+    def test_unsupported_locale_fallback_to_utf8(self):
+        locales = [
+            "zh_TW.euctw",
+            "hy_AM.armscii8",
+            "ka_GE.georgianps",
+            "C"
+        ]
+
+        for locale in locales:
+            with self.subTest(locale=locale):
+                env = dict(os.environ, LC_ALL=locale, PYTHONUTF8="0")
+
+                result = subprocess.run(
+                    [sys.executable, "-c", "import sys; print(sys.getfilesystemencoding())"],
+                    env=env,
+                    capture_output=True,
+                    text=True,
+                    timeout=10)
+
+                self.assertEqual(result.returncode, 0)
+                if locale == "C":
+                    self.assertEqual(result.stdout.strip(), "ascii")
+                else:
+                    self.assertEqual(result.stdout.strip(), "utf-8")
+
 
 def tearDownModule():
     support.reap_children()

Misc/NEWS.d/next/Core_and_Builtins/2025-09-29-19-10-31.gh-issue-91992.KtCoxh.rst

@@ -0,0 +1 @@
+Fallback to UTF-8 if an unsupported locale is provided.

Objects/unicodeobject.c

@@ -16455,10 +16455,23 @@ config_get_codec_name(wchar_t **config_encoding)
 
     PyObject *name_obj = NULL;
     PyObject *codec = _PyCodec_Lookup(encoding);
-    PyMem_RawFree(encoding);
 
-    if (!codec)
-        goto error;
+    if (!codec) {  // Fallback to UTF-8 if the codec is not found
+        PyMem_RawFree(encoding);
+        PyErr_Clear();
+
+        wchar_t *utf8_encoding = _PyMem_RawWcsdup(L"utf-8");
+        if (utf8_encoding == NULL) {
+            PyErr_NoMemory();
+            return -1;
+        }
+
+        PyMem_RawFree(*config_encoding);
+        *config_encoding = utf8_encoding;
+        return 0;
+    }
+
+    PyMem_RawFree(encoding);
 
     name_obj = PyObject_GetAttrString(codec, "name");
     Py_CLEAR(codec);