Re[2]: [SOGo] Exception in processing of unicode strings

[Helge Hess]

On 21.11.2007, at 18:37, Wolfgang Sourdeau wrote:
> SOPE is full of calls to getCString which lacks in correctness when  
> all your encodings are not the same between your web application  
> and your server environment.


Thats true, SOPE probably has a few sections which are not Unicode  
ready (however note that not all getCString's are wrong!, it can't be  
that many).
I suppose all modern Linuxes are C99 now, so we should probably use  
mbstowcs() and companions when possible (this allows us to convert  
from Linux platform encoding to wchar_t).

Anyways, the error is pretty specific:

   EXCEPTION: <NSException: 8659d38>
   NAME:NSCharacterConversionException
   REASON:Can't get cStringLength from Unicode string.

We really need to find the section which throws this exception, place  
a breakpoint and see how its triggered.

Ivanov: would it be OK for you to fetch a fresh GDB, compile it with  
Objective-C support and retry?

Thanks,
   Helge
-- 
Helge Hess
http://www.helgehess.eu/