You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I found failures in the test ./tests/primal_intersect_test --gtest_filter=*triangle_segment_intersection.
The specific case has the label "beyond corner" in file primal_intersect.cpp. We test a triangle-segment intersection where the intersection is right at a corner of the triangle. The results depend on the round-off error, but the test expects an intersection. The test should be robust enough to allow for intersection or not, and maybe check that the closest distance is within some epsilon for these cases.
I don't think this failure would affect applications, but to head off false positives, it should be fixed. Hence, I'm labeling it low priority. I have the configuration that can reproduce this bug, but it's a horrendously long Spack-generated cmake command.
The text was updated successfully, but these errors were encountered:
I found failures in the test
./tests/primal_intersect_test --gtest_filter=*triangle_segment_intersection
.The specific case has the label "beyond corner" in file
primal_intersect.cpp
. We test a triangle-segment intersection where the intersection is right at a corner of the triangle. The results depend on the round-off error, but the test expects an intersection. The test should be robust enough to allow for intersection or not, and maybe check that the closest distance is within some epsilon for these cases.I don't think this failure would affect applications, but to head off false positives, it should be fixed. Hence, I'm labeling it low priority. I have the configuration that can reproduce this bug, but it's a horrendously long Spack-generated cmake command.
The text was updated successfully, but these errors were encountered: