Enhance ElviraAlgorithm so it can output plane offset and point mesh. #1570
Conversation
…a field and a pointmesh option that lets it output point meshes.
… Added some 3D tests.
BradWhitlock
requested review from
kennyweiss,
rhornung67,
publixsubfan,
gunney1,
bmhan12 and
Arlie-Capps
There was a problem hiding this comment.
Thanks for updating the documentation!
src/axom/mir/ElviraAlgorithm.hpp
Outdated
| // The last fragment's normals are just (1,0,0). | ||
| //const double *normalPtr = fragmentVectorsView.data() + (fragmentIndex * numVectorComponents); | ||
| // It seems more useful to emit the opposite of the last fragment's normal instead. |
There was a problem hiding this comment.
construction debris? Or need to keep?
There was a problem hiding this comment.
I reworked that comment so it retains what I wanted without being an old line of code.
| output is comprised of shapes that result from several cuts of the input zones, the resulting | ||
| topology is not necessarily water-tight and it consists of polygons for 2D and polyhedra for 3D. | ||
| output is typically comprised of shapes that result from several cuts of the input zones, | ||
| the resulting topology is not necessarily water-tight and it consists of polygons for 2D |
There was a problem hiding this comment.
Readability suggestion: break sentence so we have "...is not necessarily water-tight. It consists of polygons..."
| The ELVIRA algorithm also supports a mode where it instead creates a mesh consisting of | ||
| points at the centers of the polygonal/polyhedral zone centers and emits the points in | ||
| place of the other zones. This output mode is used to save work when the algorithm is |
There was a problem hiding this comment.
I'm confused by this sentence: please clarify. I think you mean something like
"The ELVIRA algorithm also supports a mode where it creates points instead of a polygonal/polyhedral mesh. The points are located where the centroids of the zone faces would have been. This output mode is used..."
There was a problem hiding this comment.
Or does pointmesh produce points only at the centroids of new fragment cut planes/facets?
There was a problem hiding this comment.
Ack, I read it more carefully. Do you mean "A point is created at the centroid of each zone." ?
There was a problem hiding this comment.
I tried again. pointmesh makes a point at the origin of every clipping plane. A zone with 3 materials would make 3 clipping planes (2 are at the same place with opposite normals). A zone with 1 material does not get clipped so the point goes at the zone centroid. This mode is for when wants the plane origin and normal and is not that interested in the clipped zone fragments - so we make points instead.
There was a problem hiding this comment.
Thank you! Much clearer.
| +---------------------------------+------------------------------------------------------+ | ||
| | Option | Description | | ||
| +=================================+======================================================+ |
There was a problem hiding this comment.
Please state the default behavior. I am guessing that if plane or pointmesh is omitted, the behavior is the same as specifying 0. What is the default for point_tolerance?
There was a problem hiding this comment.
I added defaults in the table.
This PR resolves #1565.
axom::mir::ElviraAlgorithmclass so it can output plane normal and offset for the planes used to clip zone fragments. These fields are output when the "plane" option is set to values other than zero.axom::mir::ElviraAlgorithmto accept a "pointmesh" option that causes the algorithm to generate a mesh consisting of points located at the zone fragment clipping plane origins instead of making the usual polygonal or polyhedral meshes. This skips more costly work involved in generating the usual geometry when the user only wants the clipping planes.