[3.7] bpo-29571: Fix test_re.test_locale_flag() (GH-12178)

vstinner · web-flow · commit 279657bac285 · 2019-03-05T16:17:43.000+01:00
Use locale.getpreferredencoding() rather than locale.getlocale() to
get the locale encoding. With some locales, locale.getlocale()
returns the wrong encoding.

For example, on Fedora 29, locale.getlocale() returns ISO-8859-1
encoding for the "en_IN" locale, whereas
locale.getpreferredencoding() reports the correct encoding: UTF-8.

On Windows, set temporarily the LC_CTYPE locale to the user preferred
encoding to ensure that it uses the ANSI code page, to be consistent
with locale.getpreferredencoding().
diff --git a/Lib/test/test_re.py b/Lib/test/test_re.py
@@ -1516,8 +1516,18 @@ def test_ascii_and_unicode_flag(self):
         self.assertRaises(re.error, re.compile, r'(?au)\w')
 
     def test_locale_flag(self):
-        import locale
-        _, enc = locale.getlocale(locale.LC_CTYPE)
+        # On Windows, Python 3.7 doesn't call setlocale(LC_CTYPE, "") at
+        # startup and so the LC_CTYPE locale uses Latin1 encoding by default,
+        # whereas getpreferredencoding() returns the ANSI code page. Set
+        # temporarily the LC_CTYPE locale to the user preferred encoding to
+        # ensure that it uses the ANSI code page.
+        oldloc = locale.setlocale(locale.LC_CTYPE, None)
+        locale.setlocale(locale.LC_CTYPE, "")
+        self.addCleanup(locale.setlocale, locale.LC_CTYPE, oldloc)
+
+        # Get the current locale encoding
+        enc = locale.getpreferredencoding(False)
+
         # Search non-ASCII letter
         for i in range(128, 256):
             try:
diff --git a/Misc/NEWS.d/next/Tests/2019-03-05-13-48-39.bpo-29571.ecGuKR.rst b/Misc/NEWS.d/next/Tests/2019-03-05-13-48-39.bpo-29571.ecGuKR.rst
@@ -0,0 +1,6 @@
+Fix ``test_re.test_locale_flag()``:  use ``locale.getpreferredencoding()``
+rather than ``locale.getlocale()`` to get the locale encoding. With some
+locales, ``locale.getlocale()`` returns the wrong encoding. On Windows, set
+temporarily the ``LC_CTYPE`` locale to the user preferred encoding to ensure
+that it uses the ANSI code page, to be consistent with
+``locale.getpreferredencoding()``.
