diff --git a/Lib/test/test_unicode.py b/Lib/test/test_unicode.py
index 5e96ba8b63b..94035ef597e 100644
--- a/Lib/test/test_unicode.py
+++ b/Lib/test/test_unicode.py
@@ -1380,20 +1380,23 @@ class UnicodeTest(string_tests.CommonTest,
         self.assertEqual("{}".format(s), '__str__ overridden')
 
     def test_from_format(self):
-        # Ensure that PyUnicode_FromFormat() raises an error for a non-ascii
-        # format string.
         from _testcapi import format_unicode
 
         # ascii format, non-ascii argument
         text = format_unicode(b'ascii\x7f=%U', 'unicode\xe9')
         self.assertEqual(text, 'ascii\x7f=unicode\xe9')
 
-        # non-ascii format, ascii argument
+        # non-ascii format, ascii argument: ensure that PyUnicode_FromFormat()
+        # raises an error for a non-ascii format string.
         self.assertRaisesRegexp(ValueError,
             '^PyUnicode_FromFormatV\(\) expects an ASCII-encoded format '
             'string, got a non-ASCII byte: 0xe9$',
             format_unicode, b'unicode\xe9=%s', 'ascii')
 
+        # other tests
+        text = format_unicode(b'%%A:%A', 'abc\xe9\uabcd\U0010ffff')
+        self.assertEqual(text, r"%A:'abc\xe9\uabcd\U0010ffff'")
+
     # Test PyUnicode_AsWideChar()
     def test_aswidechar(self):
         from _testcapi import unicode_aswidechar
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index 0e2f95018a5..7564b67a215 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -1074,6 +1074,7 @@ PyUnicode_FromFormatV(const char *format, va_list vargs)
             }
             case 'S':
             case 'R':
+            case 'A':
             {
                 Py_UNICODE *ucopy;
                 Py_ssize_t usize;