[Paraview] Run Problem using Viaual Studio 8.0

[Amy Squillacote]

Hi Andrew,

I'm copying this thread back to the ParaView 
users list to keep other people up-to-date on 
this issue.  The error around line 1205 in 
vtkClientServerStream.cxx has now been fixed both 
on the trunk and the 2.4 branch of the ParaView 
CVS repository.  Another developer at Kitware 
(who has a Visual Studio 8.0 compiler) is looking 
into the problem in 
vtkInformationVectorKey.  I'll let you know when it has been fixed.

- Amy

At 10:33 PM 3/13/2006, Andrew Maclean wrote:
>Hi Amy,
>
>It still won’t run – an exception is raised at 
>line 1205 in vtkClientserver.cxx.
>
>Thanks for fixing line 239 (I prefer your 
>solution to mine!), but in the same bug report I 
>also mentioned line1205 and, as I thought, it is not fixed.
>Look at line 1205 – in vtkClientServerStream.cxx 
>[ unsigned char* end = &*this->Internal->Data.end(); ]
>This raises a “vector iterator not 
>dereferencable” error. This is actually correct 
>behavior; the standard does not allow you to 
>dereference the end iterator. It just so happens 
>that this compiler picks up this one and no others do!
>
>
>My suggestions would be:
>
>1 ) Delete lines 1208 to 1211 and add the following at line 1202:
>if(this->Internal->Data.empty())
>   {
>      return 0;
>   }
>This does the empty vector check before anything happens.
>
>2) You can then (but I don’t like it – it’s a 
>kludge which fools the compiler) replace:
>unsigned char *end = &*this->Internal->Data.end();
>with
>unsigned char *end = begin+this->Internal->Data.size()*sizeof(unsigned char);
>
>and the parser seems to work.
>
>3) But then you then get a crash at line 138 in 
>vtkInformationVectorKey.cxx. Once again this 
>line triggers a “vector subscript out of range error”.
>
>
 this is getting complex!   I don’t know this 
>code very well. I am guessing that v is empty 
>and that info is empty also. So something is not 
>being checked for being empty in ParaView.
>
>I suspect something in vtkSMProxy::UpdateVTKObjects(); line 875.
>
>Attached is the call stack for this crash.
>
>
>4) Could you also disable the deprecated 
>warnings by adding this to the CMakeLists.txt file?
>
>
>I hope all this helps.
>
>
>Regards
>
>Andrew
>
>
>
>From: Amy Squillacote [mailto:amy.squillacote at kitware.com]
>Sent: Tuesday, 14 March 2006 00:58
>To: a.maclean at cas.edu.au; paraview at paraview.org
>Subject: Re: [Paraview] Run Problem using Viaual Studio 8.0
>
>Hi Andrew,
>
>A change was made recently (a week ago) to fix 
>this problem both on the trunk and the 2.4 
>branch of the ParaView CVS repository.  How 
>recent is your checkout of ParaView?  If it's 
>older than a week ago, please try updating the 
>source tree and rebuilding.  If your source tree 
>is up-to-date and you're still seeing the 
>problem, can you tell where the end iterator is 
>being dereferenced and will you send us a stack trace?
>
>- Amy
>
>At 01:57 AM 3/11/2006, Andrew Maclean wrote:
>
>Hi All,
>    I have been able to compile ParaView OK 
> using Visual Studio  8.0 (it compiles with the 
> usual warnings about {Microsoft} deprecated functions) but it will not run.
>I have filed a bug report (2926) but I am 
>curious to see if others are able to run 
>ParaView after compiling with Visual Studio 8.0. 
>The crash seems to be caused by dereferencing the end iterator.
>Am I the only one using Visual Studio 8.0 and ParaView?
>
>Andrew
>--
>
>___________________________________________
>Andrew J. P. Maclean
>Centre for Autonomous Systems
>The Rose Street Building J04
>The University of Sydney  2006  NSW
>AUSTRALIA
>Ph: +61 2 9351 3283
>Fax: +61 2 9351 7474
>URL: <http://www.cas.edu.au/>http://www.cas.edu.au/
>___________________________________________
>
>_______________________________________________
>ParaView mailing list
>ParaView at paraview.org
>http://www.paraview.org/mailman/listinfo/paraview
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/paraview/attachments/20060314/30392658/attachment.htm