Python 2 can encode/decode surrogates to utf-8. Add a test for this.
diff --git a/Lib/test/test_unicode.py b/Lib/test/test_unicode.py
index 0452f47..ae4c355 100644
--- a/Lib/test/test_unicode.py
+++ b/Lib/test/test_unicode.py
@@ -667,11 +667,17 @@
# see http://www.unicode.org/versions/Unicode5.2.0/ch03.pdf
# (table 3-7) and http://www.rfc-editor.org/rfc/rfc3629.txt
#for cb in map(chr, range(0xA0, 0xC0)):
- #sys.__stdout__.write('\\xED\\x%02x\\x80\n' % ord(cb))
#self.assertRaises(UnicodeDecodeError,
#('\xED'+cb+'\x80').decode, 'utf-8')
#self.assertRaises(UnicodeDecodeError,
#('\xED'+cb+'\xBF').decode, 'utf-8')
+ # but since they are valid on Python 2 add a test for that:
+ for cb, surrogate in zip(map(chr, range(0xA0, 0xC0)),
+ map(unichr, range(0xd800, 0xe000, 64))):
+ encoded = '\xED'+cb+'\x80'
+ self.assertEqual(encoded.decode('utf-8'), surrogate)
+ self.assertEqual(surrogate.encode('utf-8'), encoded)
+
for cb in map(chr, range(0x80, 0x90)):
self.assertRaises(UnicodeDecodeError,
('\xF0'+cb+'\x80\x80').decode, 'utf-8')