[Paraview] memory leak in writer
Natalie Happenhofer
nataliehapp at hotmail.com
Fri Jul 17 13:41:58 EDT 2009
ok, I tried cvs -d :pserver:anoncvs at www.paraview.org:/cvsroot/ParaView-3-6-1 , but it says there is no such repository.
N
> Date: Fri, 17 Jul 2009 13:34:34 -0400
> Subject: Re: [Paraview] memory leak in writer
> From: utkarsh.ayachit at kitware.com
> To: nataliehapp at hotmail.com
>
> You can check out the CVS tag: ParaView-3-6 for the 3.6 branch or
> ParaView-3-6-1 for the release tag.
>
> Utkarsh
>
> On Fri, Jul 17, 2009 at 1:31 PM, Natalie
> Happenhofer<nataliehapp at hotmail.com> wrote:
> > But the source code is also not available yet, is it?
> >
> > Natalie
> >
> >> Date: Fri, 17 Jul 2009 13:22:20 -0400
> >> Subject: Re: [Paraview] memory leak in writer
> >> From: utkarsh.ayachit at kitware.com
> >> To: nataliehapp at hotmail.com
> >> CC: paraview at paraview.org
> >>
> >> You can't download 3.6 binaries yet. We should have them posted by
> >> early next week.
> >>
> >> Utkarsh
> >>
> >> On Fri, Jul 17, 2009 at 1:13 PM, Natalie
> >> Happenhofer<nataliehapp at hotmail.com> wrote:
> >> > I have written it months ago and it worked well for small files and
> >> > timeseries for about 30 files - I have just come across it trying to
> >> > write a
> >> > time series of 200 files.
> >> >
> >> > I can already download ParaView 3.6? Great, I did not know this!
> >> >
> >> > thx,
> >> > Natalie
> >> >
> >> >> Date: Fri, 17 Jul 2009 08:47:37 -0400
> >> >> Subject: Re: [Paraview] memory leak in writer
> >> >> From: utkarsh.ayachit at kitware.com
> >> >> To: nataliehapp at hotmail.com
> >> >> CC: paraview at paraview.org
> >> >>
> >> >> Out of curiosity, why are you writing your own wrapper? ParaView 3.6
> >> >> onwards supports writing time series for legacy vtk files.
> >> >>
> >> >> Utkarsh
> >> >>
> >> >> On Fri, Jul 17, 2009 at 7:53 AM, Natalie
> >> >> Happenhofer<nataliehapp at hotmail.com> wrote:
> >> >> > Hi!
> >> >> > I´ve written a Wrapper for the Legacy VTK files writer, so that it
> >> >> > writes a
> >> >> > whole timeseries if I want it to. Lamentably, applying it, it needs
> >> >> > more
> >> >> > and
> >> >> > more memory, so I am looking for a memory leak.
> >> >> >
> >> >> > Here is the bit of code I suspect:
> >> >> >
> >> >> > vtkInformation* inInfo = inputVector[0]->GetInformationObject(0);
> >> >> > vtkDataObject* input = inInfo->Get(vtkDataObject::DATA_OBJECT());
> >> >> >
> >> >> > vtkSmartPointer<vtkDataObject> inputCopy;
> >> >> > inputCopy.TakeReference(input->NewInstance());
> >> >> > inputCopy->ShallowCopy(input);
> >> >> >
> >> >> > vtksys_ios::ostringstream fname;
> >> >> > if (this->WriteAllTimeSteps)
> >> >> > {
> >> >> > vtkstd::string path =
> >> >> > vtksys::SystemTools::GetFilenamePath(this->FileName);
> >> >> > vtkstd::string fnamenoext =
> >> >> >
> >> >> > vtksys::SystemTools::GetFilenameWithoutLastExtension(this->FileName);
> >> >> > vtkstd::string ext =
> >> >> > vtksys::SystemTools::GetFilenameLastExtension(this->FileName);
> >> >> > if((this -> CurrentTimeIndex) < 10) fname << path << "/" <<
> >> >> > fnamenoext
> >> >> > << ".000" <<this->CurrentTimeIndex << ext;
> >> >> > else if((this -> CurrentTimeIndex) < 100) fname << path << "/" <<
> >> >> > fnamenoext << ".00" << this -> CurrentTimeIndex << ext;
> >> >> > else if((this -> CurrentTimeIndex) < 1000) fname << path << "/"
> >> >> > <<
> >> >> > fnamenoext << ".0" << this -> CurrentTimeIndex << ext;
> >> >> > else fname << path << "/" << fnamenoext << this ->
> >> >> > CurrentTimeIndex
> >> >> > <<
> >> >> > ext;
> >> >> > }
> >> >> > else
> >> >> > {
> >> >> > fname << this->FileName;
> >> >> > }
> >> >> >
> >> >> > //Here I call the usual VTK Writer and set the input
> >> >> >
> >> >> > vtkGenericDataObjectWriter* Writer =
> >> >> > vtkGenericDataObjectWriter::New();
> >> >> > Writer -> SetInput(inputCopy);
> >> >> > Writer -> SetFileName(fname.str().c_str());
> >> >> > if((this -> FileType) == 1) Writer -> SetFileTypeToASCII();
> >> >> > else Writer -> SetFileTypeToBinary();
> >> >> >
> >> >> > Writer -> Update();
> >> >> >
> >> >> > if (this->WriteAllTimeSteps)
> >> >> > {
> >> >> > this->CurrentTimeIndex++;
> >> >> > if (this->CurrentTimeIndex == this->NumberOfTimeSteps)
> >> >> > {
> >> >> > // Tell the pipeline to stop looping.
> >> >> >
> >> >> >
> >> >> > request->Remove(vtkStreamingDemandDrivenPipeline::CONTINUE_EXECUTING());
> >> >> > this->CurrentTimeIndex = 0;
> >> >> > }
> >> >> > }
> >> >> >
> >> >> > Writer -> Delete();
> >> >> > return 1;
> >> >> >
> >> >> > Is there anything I forget to delete?
> >> >> >
> >> >> >
> >> >> > thx for answers
> >> >> > Natalie
> >> >> >
> >> >> >
> >> >> >
> >> >> > ________________________________
> >> >> > Nuevo Internet Explorer 8 ¡Es Gratis y optimizado para MSN!
> >> >> > _______________________________________________
> >> >> > Powered by 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
> >> >> >
> >> >> >
> >> >
> >> > ________________________________
> >> > ¿Quieres conocerte mejor? ¡Conoce lo que Windows Live tiene
> >> > especialmente
> >> > para ti!
> >> > _______________________________________________
> >> > Powered by 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
> >> >
> >> >
> >
> > ________________________________
> > ¿Quieres conocerte mejor? ¡Conoce lo que Windows Live tiene especialmente
> > para ti!
_________________________________________________________________
Enamórate este invierno con MSN Amor y Amistad
http://amor.latam.msn.com/channel/index.aspx?trackingid=202694
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.paraview.org/pipermail/paraview/attachments/20090717/a7e01e1f/attachment-0001.htm>
More information about the ParaView
mailing list