[Paraview] [Catalyst] Bug in PV 5.4.0-RC3

[Renato Elias]

Just to report a bug detected in version 5.4.0-RC3


1) If a load a set of parallel data stored in xdmf file format
2) extract surface
3) apply a clip to isolate one of the surface
4) color this surface with a variable, let's say, pressure
and
5) save the coprocessing script using the "autoscale" for the variable used
to color the surface clipped, Catalyst will crash when running in parallel.

I've tested it a ton of times and it only happen with the clip filter
specifically. In my case, I was running a flow around a sphere and the
clipped data (using the sphere tool) was the sphere itself where I'd like
to color with the pressure but I only managed to make a valid python script
after selecting a solid color for the sphere...

the basica pipeline was

1) load the xdmf
2) create ghost cells
3) merge blocks
4) extract surface
5) clip (the sphere)
6) stream tracer
7) tube (colored with velocity)

In the 5th step, I've tried:

a). Color with pressure and save with autoscale enabled. It crashes
b). Color with pressure and save with autoscale disabled. It works fine
c). Color with a solid color and using autoscale for the tube colored with
velocity. It works fine

I hope it helps to reproduce the problem

Regards

Renato
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20170531/cb6320ea/attachment.html>