On Thursday 24 November 2011, 12:19:09 Victor Varvariuc wrote:
> #!/usr/bin/env python3
>
> from PyQt4 import QtGui, QtCore, uic
>
> class Form(QtGui.QDialog):
>
> def __init__(self, parentWidget):
> super().__init__(parentWidget)
> self.setupUi()
>
> def setupUi(self):
> uic.loadUi('test.ui', self)
>
> def ัะตัั(self):
> '''This method has non ASCII name causing this error:
> Traceback (most recent call last):
> File "test.py", line 20, in <module>
> form = Form(None)
> File "test.py", line 9, in __init__
> self.setupUi()
> File "test.py", line 12, in setupUi
> uic.loadUi('test.ui', self)
> File "/usr/lib/python3/dist-packages/PyQt4/uic/__init__.py", line
> 221, in loadUi
> return DynamicUILoader().loadUi(uifile, baseinstance)
> File "/usr/lib/python3/dist-packages/PyQt4/uic/Loader/loader.py",
> line 71, in loadUi
> return self.parse(filename, basedir)
> File "/usr/lib/python3/dist-packages/PyQt4/uic/uiparser.py", line
> 925, in parse
> elem = document.find(tagname)
> File "/usr/lib/python3.2/xml/etree/ElementTree.py", line 726, in
> find return self._root.find(path, namespaces)
> File "/usr/lib/python3.2/xml/etree/ElementTree.py", line 363, in
> find return ElementPath.find(self, path, namespaces)
> File "/usr/lib/python3.2/xml/etree/ElementPath.py", line 285, in
> find return next(iterfind(elem, path, namespaces))
> File "/usr/lib/python3.2/xml/etree/ElementPath.py", line 249, in
> iterfind if path[-1:] == "/":
> UnicodeEncodeError: 'ascii' codec can't encode characters in position
> 0-3: ordinal not in range(128)
> '''
>
> if __name__ == '__main__':
> app = QtGui.QApplication([])
> form = Form(None)
> form.show()
> app.exec()
>
>
> Python3 supports unicode names - which are working. Does Qt support
> unicode names or it's a PyQt bug (see the example above)? Qt Designer
> also doesn't allow me to enter non-ascii names for objects.
In one word: don't.
Since PyQt has to respect the constraints of Qt (C++), what do you
expect? After rewriting uic and qt designer, there's even more trouble
ahead of your road (qt meta system, translation, ...)
I have a hard time to find the upsides from this move in python3.
What would you think about some wonderful python open source
application, that is written with all kind of _labels_ in hanzi, kanji
or hanja? I guess, it would prevent you from modifying anything in that
code. Similar, a lot of people in the world are lost with cyrillic
labels.
Imagine, Phil would write his code in some gaelic language? Even if
(mostly) ascii based, only a small population would be able to
understand the code. To get this straight: even written with proper
english labels, comments and docuemtation, it's hard to grok
completely. Try it.
Unicode labels are a silly form of obfuscation to me.
If you like that, why don't you start hacking in APL, then?
Pete
_______________________________________________
PyQt mailing list PyQt@riverbankcomputing.com
http://www.riverbankcomputing.com/mailman/listinfo/pyqt
