diff options
Diffstat (limited to 'Lib/test')
-rw-r--r-- | Lib/test/test_unicode.py | 23 |
1 files changed, 11 insertions, 12 deletions
diff --git a/Lib/test/test_unicode.py b/Lib/test/test_unicode.py index 8d4bed5..5368f6e 100644 --- a/Lib/test/test_unicode.py +++ b/Lib/test/test_unicode.py @@ -495,19 +495,18 @@ else: verify(unicode('+3ADYAA-', 'utf-7', 'replace') == u'\ufffd') # UTF-8 specific encoding tests: -verify(u'\u20ac'.encode('utf-8') == \ - ''.join((chr(0xe2), chr(0x82), chr(0xac))) ) -verify(u'\ud800\udc02'.encode('utf-8') == \ - ''.join((chr(0xf0), chr(0x90), chr(0x80), chr(0x82))) ) -verify(u'\ud84d\udc56'.encode('utf-8') == \ - ''.join((chr(0xf0), chr(0xa3), chr(0x91), chr(0x96))) ) +verify(u'\u20ac'.encode('utf-8') == '\xe2\x82\xac') +verify(u'\ud800\udc02'.encode('utf-8') == '\xf0\x90\x80\x82') +verify(u'\ud84d\udc56'.encode('utf-8') == '\xf0\xa3\x91\x96') +verify(u'\ud800'.encode('utf-8') == '\xed\xa0\x80') +verify(u'\udc00'.encode('utf-8') == '\xed\xb0\x80') +verify((u'\ud800\udc02'*1000).encode('utf-8') == + '\xf0\x90\x80\x82'*1000) + # UTF-8 specific decoding tests -verify(unicode(''.join((chr(0xf0), chr(0xa3), chr(0x91), chr(0x96))), - 'utf-8') == u'\U00023456' ) -verify(unicode(''.join((chr(0xf0), chr(0x90), chr(0x80), chr(0x82))), - 'utf-8') == u'\U00010002' ) -verify(unicode(''.join((chr(0xe2), chr(0x82), chr(0xac))), - 'utf-8') == u'\u20ac' ) +verify(unicode('\xf0\xa3\x91\x96', 'utf-8') == u'\U00023456' ) +verify(unicode('\xf0\x90\x80\x82', 'utf-8') == u'\U00010002' ) +verify(unicode('\xe2\x82\xac', 'utf-8') == u'\u20ac' ) # Other possible utf-8 test cases: # * strict decoding testing for all of the |