Make measure volume / area warn if surface is clipped

[Tom Goddard]

The measure volume and area commands should probably warn if clip cap submodels are found because the reported values include clipped parts of the surface contrary to what a user is likely to expect.

From: Tom Goddard
Subject: Re: [chimerax-users] volume amount estimation
Date: December 21, 2023 at 6:16:09 PM PST
To: Elaine Meng

Hi Elaine,

The coordinates defining the cap are floating point numbers and they don’t exactly lie in a plane so the volume and area come out to be non-zero because of code’s closing of holes by stitching surfaces to span any hole boundaries. On the other hand cap models are marked as caps so it would probably be best to exclude them from the measure command. While I could filter out those, I think I won’t because if you use clip planes you probably are going to think they effect the volume and they don’t. So perhaps this noise about caps will alert the user that caps are present and cause them to investigate. I guess better would be if I see clip cap submodels I should spew a warning that the reported volume and area of the surface includes clipped invisible parts.

Tom

On Dec 21, 2023, at 6:01 PM, Elaine Meng wrote:

Possibly a trivial micro-mini-bug is that planar caps are reported as enclosing a tiny amount of volume. I.e. when I had global clipping planes on a volume isosurface #2:

measure volume #2
Enclosed volume for surface (#2.1) = 1.547e+04
Enclosed volume for cap near (#2.1.1) = 6.825e-07 with 2 surface holes
Enclosed volume for cap far (#2.1.2) = 0.002643 with 1 surface holes
Total enclosed volume for 3 surfaces = 1.547e+04 with 3 surface holes

When the planes were not intersecting the isosurface, however, they were reported as enclosing 0 volume.