bpo-34527: POSIX locale enables the UTF-8 Mode (GH-8972) (GH-8974)

* The UTF-8 Mode is now also enabled by the "POSIX" locale, not only
  by the "C" locale.
* On FreeBSD, Py_DecodeLocale() and Py_EncodeLocale() now also forces
  the ASCII encoding if the LC_CTYPE locale is "POSIX", not only if
  the LC_CTYPE locale is "C".
* test_utf8_mode.test_cmd_line() checks also that the command line
  arguments are decoded from UTF-8 when the the UTF-8 Mode is enabled
  with POSIX locale or C locale.

(cherry picked from commit 5cb258950c)

Lib/test/test_utf8_mode.py

@@ -12,6 +12,7 @@ from test.support.script_helper import assert_python_ok, assert_python_failure
 
 
 MS_WINDOWS = (sys.platform == 'win32')
+POSIX_LOCALES = ('C', 'POSIX')
 
 
 class UTF8ModeTests(unittest.TestCase):
@@ -23,7 +24,7 @@ class UTF8ModeTests(unittest.TestCase):
 
     def posix_locale(self):
         loc = locale.setlocale(locale.LC_CTYPE, None)
-        return (loc == 'C')
+        return (loc in POSIX_LOCALES)
 
     def get_output(self, *args, failure=False, **kw):
         kw = dict(self.DEFAULT_ENV, **kw)
@@ -39,7 +40,9 @@ class UTF8ModeTests(unittest.TestCase):
     def test_posix_locale(self):
         code = 'import sys; print(sys.flags.utf8_mode)'
 
-        out = self.get_output('-c', code, LC_ALL='C')
+        for loc in POSIX_LOCALES:
+            with self.subTest(LC_ALL=loc):
+                out = self.get_output('-c', code, LC_ALL=loc)
                 self.assertEqual(out, '1')
 
     def test_xoption(self):
@@ -201,7 +204,9 @@ class UTF8ModeTests(unittest.TestCase):
         out = self.get_output('-X', 'utf8', '-c', code)
         self.assertEqual(out, 'UTF-8 UTF-8')
 
-        out = self.get_output('-X', 'utf8', '-c', code, LC_ALL='C')
+        for loc in POSIX_LOCALES:
+            with self.subTest(LC_ALL=loc):
+                out = self.get_output('-X', 'utf8', '-c', code, LC_ALL=loc)
                 self.assertEqual(out, 'UTF-8 UTF-8')
 
     @unittest.skipIf(MS_WINDOWS, 'test specific to Unix')
@@ -217,11 +222,17 @@ class UTF8ModeTests(unittest.TestCase):
             self.assertEqual(args, ascii(expected), out)
 
         check('utf8', [arg_utf8])
+        for loc in POSIX_LOCALES:
+            with self.subTest(LC_ALL=loc):
+                check('utf8', [arg_utf8], LC_ALL=loc)
+
         if sys.platform == 'darwin' or support.is_android:
             c_arg = arg_utf8
         else:
             c_arg = arg_ascii
-        check('utf8=0', [c_arg], LC_ALL='C')
+        for loc in POSIX_LOCALES:
+            with self.subTest(LC_ALL=loc):
+                check('utf8=0', [c_arg], LC_ALL=loc)
 
     def test_optim_level(self):
         # CPython: check that Py_Main() doesn't increment Py_OptimizeFlag

Misc/NEWS.d/next/Core and Builtins/2018-08-28-11-52-13.bpo-34527.sh5MQJ.rst

@@ -0,0 +1,2 @@
+The UTF-8 Mode is now also enabled by the "POSIX" locale, not only by the "C"
+locale.

Misc/NEWS.d/next/Core and Builtins/2018-08-28-11-53-39.bpo-34527.aBEX9b.rst

@@ -0,0 +1,3 @@
+On FreeBSD, Py_DecodeLocale() and Py_EncodeLocale() now also forces the
+ASCII encoding if the LC_CTYPE locale is "POSIX", not only if the LC_CTYPE
+locale is "C".

Modules/main.c

@@ -2102,15 +2102,25 @@ done:
 static void
 config_init_locale(_PyCoreConfig *config)
 {
-    if (_Py_LegacyLocaleDetected()) {
-        /* POSIX locale: enable C locale coercion and UTF-8 Mode */
-        if (config->utf8_mode < 0) {
-            config->utf8_mode = 1;
-        }
     if (config->coerce_c_locale < 0) {
+        /* The C locale enables the C locale coercion (PEP 538) */
+        if (_Py_LegacyLocaleDetected()) {
             config->coerce_c_locale = 1;
         }
     }
+
+#ifndef MS_WINDOWS
+    if (config->utf8_mode < 0) {
+        /* The C locale and the POSIX locale enable the UTF-8 Mode (PEP 540) */
+        const char *ctype_loc = setlocale(LC_CTYPE, NULL);
+        if (ctype_loc != NULL
+           && (strcmp(ctype_loc, "C") == 0
+               || strcmp(ctype_loc, "POSIX") == 0))
+        {
+            config->utf8_mode = 1;
+        }
+    }
+#endif
 }
 
 

Python/fileutils.c

@@ -128,7 +128,7 @@ check_force_ascii(void)
     loc = setlocale(LC_CTYPE, NULL);
     if (loc == NULL)
         goto error;
-    if (strcmp(loc, "C") != 0) {
+    if (strcmp(loc, "C") != 0 && strcmp(loc, "POSIX") != 0) {
         /* the LC_CTYPE locale is different than C */
         return 0;
     }