[Paraview] Problem with Pixie files: build 3.14.1 and 3.98.0

Claudio Zanni zanni at oato.inaf.it
Mon Nov 12 06:06:03 EST 2012


Dear Utkarsh,

    thanks for trying to look into it.

    These are some datasets distributed with VisIt to test the Pixie output

blockedhttp://portal.nersc.gov/svn/visit/trunk/data/pixie_test_data.tar.gz

    As you will notice Paraview 3.14 or newer can not open the file 
pixie_expressions.h5,
    while it correctly opens all other files (pixie3d3.h5, ...), where 
the only difference
    between the files is the number of timesteps contained in each of them.

    The following files are written by the code that we develop 
(http://plutocode.ph.unito.it/):

    blockedhttps://dl.dropbox.com/u/18683007/data.2tstep.h5
    blockedhttps://dl.dropbox.com/u/18683007/data.1tstep.h5

     Paraview 3.14 or newer can open data.2tstep.h5 but hangs when 
trying to open data.1tstep.h5.
     These files contain 2 and 1 timesteps respectively, but their 
structure is exactly the same
     (groups, attributes, datasets, etc etc ...)

     Previous versions of Paraview (down to 3.10) and VisIt (up to the 
current release 2.5.2) can
     correctly open all the files. Therefore I suspect that the problem 
is somewhere in the bridge
     from VisIt to Paraview of the two most recent releases.
     I have always used the pre-compiled binaries for Mac OS X 64-bit Intel.

     Thanks again,
     Claudio

On 11/12/12 12:13 AM, Utkarsh Ayachit wrote:
> Claudio,
>
> Do you have any sample dataset that I can use to test this/the fix?
>
> Thanks,
> Utkarsh
>
>
> On Sun, Nov 11, 2012 at 4:07 PM, Utkarsh Ayachit 
> <utkarsh.ayachit at kitware.com <mailto:utkarsh.ayachit at kitware.com>> wrote:
>
>     Claudio,
>
>     I'll take a look at get back.
>
>     Utkarsh
>
>
>     On Fri, Nov 9, 2012 at 9:54 AM, Claudio Zanni <zanni at oato.inaf.it
>     <mailto:zanni at oato.inaf.it>> wrote:
>
>         I already had this problem with Paraview 3.14.1 but I received
>         no answer to a first email,
>         so I tried to investigate on my own since the problem is still
>         present in release 3.98.0.
>
>         Paraview can not open a Pixie file (it's a HDF5 format
>         included through the VisIt Database
>         Bridge) when the file includes one timestep only: Paraview
>         just hangs indefinitely.
>         Everything works fine when multiple timesteps are contained in
>         the file.
>
>         I suspect that the problem is connected to lines 618-634 in
>         Utilities/VisItBridge/AvtAlgorithms/vtkAvtFileFormatAlgorithm.cxx:
>
>               //in some case the times and cycles have all zero values.
>               //This is caused by a file reader that generates the
>             time value
>               //once the reader moves to that timestep.
>               //That kind of behaviour is not possible currently in
>             ParaView. Instead
>               //we will force the reader to generate the time values
>             for each timestep
>               //by cycling through everytime step but not requesting
>             any data.
>               if(hasTime && timesteps[0] == timesteps[timesteps.size()-1])
>                 {
>                 //we have hit a timestep range that needs to be cycled
>                 for(int i=0; i < timesteps.size();++i)
>                   {
>                   this->ActivateTimestep(i);
>                   //Nek and other readers don't update the time info
>             intill you
>                   //call gettimes.
>                   this->AvtFile->FormatGetTimes(timesteps);
>                   }
>                 }
>
>         The condition is true not only when the same value (0?) has
>         been assigned to all
>         the timesteps but also when there is one timestep only
>         (timesteps.size() = 1).
>         Besides these lines were not present in Paraview 3.10/3.12 and
>         these releases
>         were able to open Pixie files containing one timestep only.
>
>         Is it possible to fix this problem? That would be highly
>         appreciated.
>         Regards,
>         Claudio
>
>         -- 
>         Claudio Zanni
>         INAF - Osservatorio Astronomico di Torino
>         Via Osservatorio 20
>         10025 Pino Torinese (TO), Italy
>         tel. +39 011 8101907 <tel:%2B39%20011%208101907>
>         fax +39 011 8101930 <tel:%2B39%20011%208101930>
>
>         _______________________________________________
>         Powered by www.kitware.com <http://www.kitware.com>
>
>         Visit other Kitware open-source projects at
>         http://www.kitware.com/opensource/opensource.html
>
>         Please keep messages on-topic and check the ParaView Wiki at:
>         http://paraview.org/Wiki/ParaView
>
>         Follow this link to subscribe/unsubscribe:
>         http://www.paraview.org/mailman/listinfo/paraview
>
>
>


-- 
Claudio Zanni
INAF - Osservatorio Astronomico di Torino
Via Osservatorio 20
10025 Pino Torinese (TO), Italy
tel. +39 011 8101907
fax  +39 011 8101930

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20121112/2bee0d1c/attachment.htm>


More information about the ParaView mailing list