[Paraview] vtkKWScale - hinky?

Randy Hudson hudson at mcs.anl.gov
Wed Oct 12 14:06:16 EDT 2005


I'm resending the included email and adding some info.

In the included email, I describe aberrant behavior of the isovalue slider in the GUI of the contour filter.  I've just noticed the same behavior with the isovalue slider I use in my custom interface to select an isovalue.  Whenever I slide the slider back and forth a few times before proceeding with the button click that passes control to paraview to create the isosurface and color it, things work fine.  If I only move the slider a small amount, things usually fail.

The class my slider and the default contour slider have in common is vtkKWScale.  (I use vtkPVScale.)  I can only conclude that somehow, the isovalue selected by the slider is not guaranteed to be set, or passed to where it needs to be passed to.

Was vtkKWScale changed between paraview 2.0.0 and July 11, when I downloaded the development release I'm having trouble with?  Maybe the slider resolution is being processed differently, e.g.?


Thanks.


>Date: Mon, 10 Oct 2005 20:03:56 -0500
>To: paraview at paraview.org
>From: Randy Hudson <hudson at mcs.anl.gov>
>Subject: Contours
>
>
>In paraview 2.0.0, my custom paraview GUI produces isosurfaces by calling certain functions, such as "AcceptCallback", in the default interface of the contour filter.
>
>When I started using the paraview development release I had downloaded on July 11, the widget in my GUI quit working that sets the scalar variable by which to color the isosurface.  So, I started fiddling with the default contour widgets and learned that I have to slide the isovalue slider back and forth a couple times before I click "Accept" to force an isosurface to be created.  It doesn't work if I just instantiate the filter, add the default value with the "Add" button, and click "Accept".
>
>I'm thinking the cause of this failure might be the cause of the failure of my colormap widget.
>
>Does anyone know about this odd behavior of the contour widget and how I might get around it?
>
>Thanks.
>
>Randy.


Randy.
630 252 8690.



More information about the ParaView mailing list