[Paraview] [Catalyst] multi-block names are not written to file

[Andy Bauer]

Hi,

This appears to be a bug in VTK. I have a potential fix at
https://gitlab.kitware.com/acbauer/vtk/blob/parallelmultiblockdatawriter_write_block_names/IO/ParallelXML/vtkXMLPMultiBlockDataWriter.cxx.
You can track the fix at
https://gitlab.kitware.com/vtk/vtk/merge_requests/471.

Thanks for reporting this.

Cheers,
Andy

On Mon, Aug 3, 2015 at 4:48 AM, Deij-van Rijswijk, Menno <M.Deij at marin.nl>
wrote:

> Hi,
>
> I’m trying to set up a co-processing pipeline by supplying it with a
> multi-block dataset, of which the individual unstructured grids have names.
>
> TL;DR: the block names are not written to XML files when using a Python
> co-processing pipeline.
>
>
> In detail:
> The block names are supplied like so
>
> ///
> vtkInformation* md =
> multiBlockGrid->GetMetaData(static_cast<unsigned int>(0));
> if (md)
> {
>     md->Set(vtkCompositeDataSet::NAME(), "Interior");
> }
> ///
>
> The grid is added as a producer with the name "grid". Then, in the Python
> pipeline, I simply write the full grid using the following code:
>
> ///
> grid = coprocessor.CreateProducer( datadescription, "grid" )
> ParallelMultiBlockDataSetWriter1 = coprocessor.CreateWriter(
> XMLMultiBlockDataWriter, "fullgrid_%t.vtm", 100 )
> ///
>
> This writes the fullgrid_0.vtm file, directory and each block in a
> separate vtu file. All is good, except that the block names that were set
> are not written in the file in the "name" attribute.
>
> I have tried to write an XML file directly using XMLMultiBlockDataWriter,
> but that did not work due to the implementation of a mapped data array
> which has no implementation for NewIterator(). Weird - how does the Python
> pipeline write the file? I don't know, but it works.
>
> To still check the names having been set, I emulate the first few steps of
> the XMLMultiBlockDataWriter, to see if the block names come through like so:
>
> ///
> vtkSmartPointer<vtkDataObjectTreeIterator> iter;
> iter.TakeReference(
> vtkDataObjectTree::SafeDownCast(multiBlockGrid)->NewTreeIterator());
> iter->VisitOnlyLeavesOff();
> iter->TraverseSubTreeOff();
> iter->SkipEmptyNodesOff();
>
> int index = 0;
> int RetVal = 0;
> for (iter->InitTraversal(); !iter->IsDoneWithTraversal();
> iter->GoToNextItem(), index++)
> {
> vtkDataObject* curDO = iter->GetCurrentDataObject();
> const char *name = NULL;
> if (iter->HasCurrentMetaData())
>   {
>   name = iter->GetCurrentMetaData()->Get(vtkCompositeDataSet::NAME());
>   cout << "Name found by pseudo-write code: " << (name ? name : "NULL") <<
> endl;
>   }
> else
>   {
>     cout << "No metadata found on iterator" << endl;
>   }
> }
> ///
>
> And this clearly shows that the names are found as they are written to
> stdout.
>
> So, the question is: how can I get the Python co-processing pipeline to
> write the block names?
>
> Thanks and best wishes,
>
>
> Menno
>
>
> dr. ir. Menno A. Deij-van Rijswijk
> Researcher / Software Engineer
> Maritime Simulation & Software Group
> E mailto:M.Deij at marin.nl
> T +31 317 49 35 06
>
>
> MARIN
> 2, Haagsteeg, P.O. Box 28, 6700 AA Wageningen, The Netherlands
> T +31 317 49 39 11, F +31 317 49 32 45, I www.marin.nl
>
> _______________________________________________
> 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
>
> Search the list archives at: http://markmail.org/search/?q=ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://public.kitware.com/mailman/listinfo/paraview
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20150803/f790526a/attachment.html>