On Feb 9, 2008, at 9:24 AM, Barry Scott wrote:
>
> On Feb 7, 2008, at 09:12, Erwin Marsi wrote:
>
>>
>> On Feb 6, 2008, at 10:04 PM, Barry Scott wrote:
>>
>>>
>>> print repr(wx)
>>>
>>> will show you where wx was loaded from.
>>
>> That gives
>>
>> <module 'wx' from
'/Applications/WorkBench.app/Contents/Resources/
>> wx/__init__.pyc'>
>>
>> which seems OK, but inconsistent with the original
error message:
>>
>> Traceback (most recent call last):
>> File "../../Source/wb_exceptions.py",
line 45, in __call__
>> File "../../Source/wb_tree_panel.py",
line 448, in OnTreeSelChanged
>> File "../../Source/wb_tree_panel.py",
line 467, in changedSelection
>> File
"/Library/Frameworks/Python.framework/Versions/2.5/lib/
>>
python2.5/site-packages/wx-2.6-mac-unicode/wx/_misc.py"
, line 978,
>> in __init__
>> TypeError: new_BusyInfo() takes exactly 1 argument
(2 given)
>>
>
> Interesting. So the correct wx is load but not the
correct wx._misc.
>
> I assume that if you
>
> print repr(wx._misc)
>
> after the other print we see the bad module loaded?
No
$ cat /tmp/workbench.tmp
PySVN WorkBench starting
<module 'wx' from
'/Applications/WorkBench.app/Contents/Resources/wx/
__init__.pyc'>
<module 'wx._misc' from
'/Applications/WorkBench.app/Contents/
Resources/wx/_misc.pyc'>
>
> If so its a matter of reading wx/__init__.py to figure
out how it
> does this trick.
There's only the byte-compiled version __init__.pyc in
Modules.zip,
but it does contain the following suspicious substring:
startswithendswith(((so/Library/Frameworks/Python.framework/
Versions/
2.5/lib/python2.5/site-packages/wx-2.6-mac-unicode/wx/
__init__.py<module>s0
Erwin
------------------------------------------------------------
---------
To unsubscribe, e-mail: users-unsubscribe pysvn.tigris.org
For additional commands, e-mail: users-help pysvn.tigris.org
|