The Dolly is moving the camera towards the focal point when we zoom in. So, if the focal point is within the data set (which is most likely the case), we will never cross the focal point and move outside the dataset.
Yes, I agree that I can look to create my own custom interactor style.
To add some context to this topic, I started off by looking at this thread:
https://vtk.org/Bug/view.php?id=7823
I am facing the same issue reported in this thread (near plane clipping in parallel projection)
Now, the fix reported there was to allow the camera to have negative near plane clipping value so that it can SEE the objects behind it in the parallel projection mode. Also, that fix was not incorporated into the VTK source code.
The other way for me to fix the clipping issue was to MOVE the camera when zooming in parallel just like in perspective projection. In this case, the clipping planes are always in front of the camera.
I will, as suggested create my own interactor style but if you would have to pick the better of the two methods which one do you feel would not have any side-effects?
Thanks,
Santosh