<div dir="ltr"><div>Thanks for the inputs, I have yet to run the code through valgrind, but I will try this.</div><div> </div><div>Using vtkExtractUnstructuredGrid works without errors. Perhaps I was miss using vtkExtractGeometry, as the input dataset is an unstructured grid.</div>
</div><div class="gmail_extra"><br><br><div class="gmail_quote">On Fri, May 10, 2013 at 1:26 PM, Laurence Griffiths <span dir="ltr"><<a href="mailto:lgriffiths.aero@gmail.com" target="_blank">lgriffiths.aero@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="ltr"><div>Hi All,</div><div>I'm using vtkExtractGeometry on a large ensight dataset (vtkUnstructuredGrid - 60 million cells, approximately). </div>
<div> </div><div>ExtractData=vtk.vtkExtractGeometry()</div><div>
ExtractData.SetInput(MultiBlock.GetOutput().GetBlock(0))</div><div>ExtractData.SetImplicitFunction(ClipBox)</div><div>ExtractData.ExtractInsideOn()</div><div>ExtractData.ExtractBoundaryCellsOn()</div><div>ExtractData.Update()</div>
<div> </div><div>ClipBox is an instance of vtk.vtkBox()</div><div> </div><div>On small datasets it works fine, however on large datasets I seem to get Bands of cells missing internally in the dataset. (seen by by saving the volume, opening in paraview, and taking a slice).</div>
<div> </div><div>If i take a slice of the raw dataset (i.e. the vtkUnstructuredGrid) before passing to vtkExtractGeometry - I have no internal cells missing</div><div> </div><div>If i take a slice in vtk, I get a repeatable Bad Alloc error.</div>
<div> </div><div>Anyone else had simillar problems to this? Or Know a work around?</div><div> </div><div>Cheers,</div><div>Laurence.</div></div>
</blockquote></div><br></div>