[Paraview] best strategy for multi-block data?

[j s]

Hi,

I am currently using the multiblock dataset (vtm file) format to have
multiple unstructured regions (vtu files).  Not all of the fields are
available in all of the regions.  When I try to use a calculator to
calculate a field, I am getting errors for fields which don't exist in
each region.

So when I try to calculate E = Electric_Field_x * iHat +
Electric_Field_y *jHat, it complains about missing Acceptors.

What is the best possible way to avoid these problems?  Is there a
preferred file format?  The whole reason I selected the multiblock
format, was so that I don't have to store all the fields on all
regions.

This is paraview 3.3. for Ubuntu Linux