Closed samuelpmishLLNL closed 6 months ago
All modified and coverable lines are covered by tests :white_check_mark:
Comparison is base (
03d7d72
) 88.77% compared to head (c7335af
) 88.80%.
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
When using a 3D mesh, would ofBoundaryElements
work exactly like ofFaces
? And in 2D would ofBoundaryElements
work exactly as ofEdges
?
When using a 3D mesh, would
ofBoundaryElements
work exactly likeofFaces
? And in 2D wouldofBoundaryElements
work exactly asofEdges
?
They're not the same thing. For example, a 3D mesh has face elements on the interior and the boundary, but ofBoundaryFaces
will not visit interior faces when evaluating the predicate.
When using a 3D mesh, would
ofBoundaryElements
work exactly likeofFaces
? And in 2D wouldofBoundaryElements
work exactly asofEdges
?They're not the same thing. For example, a 3D mesh has face elements on the interior and the boundary, but
ofBoundaryFaces
will not visit interior faces when evaluating the predicate.
Got it, thanks! Should we maybe add ofBoundaryFaces
to the test? I primarily use Domain
for boundary attributes. Unless it is tested somewhere else in the repo I didn't notice?
maybe the ofFaces
and ofEdges
should be removed in favor of ofInteriorFaces
and ofBoundaryFaces
. I wonder, is there a real use-case where one would want to include boundary and interior faces in the same domain?
maybe the
ofFaces
andofEdges
should be removed in favor ofofInteriorFaces
andofBoundaryFaces
. I wonder, is there a real use-case where one would want to include boundary and interior faces in the same domain?
I think it could be useful for solving physics on a partial domain. E.g. an interior material interface may become the domain boundary along with boundary faces.
fixes https://github.com/LLNL/serac/issues/1077, adds more tests to verify