I am attempting to create a sample Ensight .case file to use as a minimal example for another issue I’m having. I create an object using a source (e.g. Wavelet, or Sphere), then run D3 to break it up, then Save Data as Ensight .case. The result in my file system is a .geo file and another file for each data array, but no .case is generated. Am I doing something wrong here? I can’t seem to load the results into Paraview afterward.
It’s possible that I’ll be able to reproduce the issue I’m having with some other kind of VTKComposite data, if that’s easier to generate. I’d be happy to give it a try.
That indeed seems to be the case. I am afraid the writer has not been looked at my anyone in a while and not surprising it has some issues. Please feel free to report an issue and we can track it down. I am not familiar with the reader / writer enough to know off what it is expected to do or should do, so will definitely need some investigation.
I found an old issue from 5.4. I tagged you in it.
Is there another way to create the kind of multiblock dataset that Paraview uses for Ensight? I have a problem that shows up with ensight data, but not with .pvtu data, for instance. I was thinking maybe something like using redistributing data from a Source and saving the output, but I’m not sure what to save it as.
From what I can see, I am not sure this writer ever worked in ParaView (except maybe ages ago).
Several issues:
only supports vtkUnstructuredGrid inputs, not multiblocks – from the looks of it, it was written before VTK supported mutliblock
vtkEnSightWriter::WriteCaseFile(...) seems to write the case file but it doesn’t get called at all. I supsect it was intended to be called manually, but we don’t ever do that in writers. I suspect the writer came into being before time was introduced to VTK pipeline and hence this odd API.
Moral of the story is we need to rewrite the writer if we want to support it. From what I can see, it’s not in usable state at this point.