add not-yet-supported Unicode test just so it doesn't get lost.

diff --git a/Lib/test/test_csv.py b/Lib/test/test_csv.py
index 18f8851e70c277b0c99cfe5b96494db121434584..c0ad645d899038c1bdd8504195046ac3d619c497 100644 (file)
--- a/Lib/test/test_csv.py
+++ b/Lib/test/test_csv.py
@@ -1,3 +1,4 @@
+# -*- coding: iso-8859-1 -*-
 # Copyright (C) 2001,2002 Python Software Foundation
 # csv package unit tests
 
@@ -640,6 +641,25 @@ else:
             # if writer leaks during write, last delta should be 5 or more
             self.assertEqual(delta < 5, True)
 
+# commented out for now - csv module doesn't yet support Unicode
+if 0:
+    from StringIO import StringIO
+    import csv
+
+    class TestUnicode(unittest.TestCase):
+        def test_unicode_read(self):
+            import codecs
+            f = codecs.EncodedFile(StringIO("Martin von Löwis,"
+                                            "Marc André Lemburg,"
+                                            "Guido van Rossum,"
+                                            "François Pinard\r\n"),
+                                   data_encoding='iso-8859-1')
+            reader = csv.reader(f)
+            self.assertEqual(list(reader), [[u"Martin von Löwis",
+                                             u"Marc André Lemburg",
+                                             u"Guido van Rossum",
+                                             u"François Pinardn"]])
+
 def test_main():
     mod = sys.modules[__name__]
     test_support.run_unittest(