Le mardi 17 décembre 2013 10:29:28 UTC+1, Steven D'Aprano a écrit : > On Mon, 16 Dec 2013 23:58:15 -0800, wxjmfauth wrote: > > > > > From all the toolkits, wxPython is probably the most interesting. I used > > > all versions from 2.0 (?) up to 2.8. Then it has been decided to go > > > unicode. > > > > > > Let see in the wx interactive intepreter, it is only the top of the > > > iceberg. (Py27, wxPy294) > > > > > >>>> len('ሴЃ') > > > 5 > > > > > > What does that have to do with wxPython? It looks like you're just mis- > > using Python 2.7. > > > > In Python 2.7, 'ሴЃ' is not a Unicode string, it is a byte string. The > > exact bytes you get are not well-defined but on many systems you may get > > a UTF-8 encoded byte string: > > > > > > py> sys.version > > '2.7.4 (default, Apr 18 2013, 17:48:59) \n[GCC 4.4.5]' > > py> for b in 'ሴЃ': > > ... print hex(ord(b)), b > > ... > > 0xe1 > > 0x88 � > > 0xb4 � > > 0xd0 > > 0x83 � > > > > > > If you use a Unicode string instead: > > > > py> for c in u'ሴЃ': > > ... print hex(ord(c)), c > > ... > > 0x1234 ሴ > > 0x403 Ѓ > > > > py> for b in u'ሴЃ'.encode('utf-8'): > > ... print hex(ord(b)), b > > ... > > 0xe1 > > 0x88 � > > 0xb4 � > > 0xd0 > > 0x83 � > > > > > > > > Even if it is true that wxPython cannot handle Unicode text, you haven't > > shown it here. > > > > > > Quich answer.
That's PyShell, the wxPy interpreter. You are right in pointing all this missmatch: byte string, unicode, utf-8, mixing a "unicode" tool kit wiht a non unicode native engine, plus plenty of related things, ... You may be not aware. I contributed to wxPython from practically 2.0, 2.1. 2.2 during 6, 7, 8 yeaurs, testing, reporting about/with all versions, then I gave up. --- Yout quote. "Even if it is true that wxPython cannot handle Unicode text..." It's a confusing vision and that's not correct. It's a little bit more complicate than this. jmf -- https://mail.python.org/mailman/listinfo/python-list