Archived issue #0025249

Some problematic cases from mesh tests have been uncovered during refactoring of BRepMesh

Open CASCADEOCCT:Meshfeedback3 public notes

Search issues

Description

During refactoring of main classes of BRepMesh package performed in context of #0023106 some bad cases have been faced during testing campaign.

Attachments contain faces extracted from that cases. In general, these shapes seems invalid, however BRepMesh should try to fix some common problems on its level to produce mesh anyway.

Public activity

3 archived notes

Participants are labeled by their role within this record.

01Commenter 1
Problem described in issue is reproduced on current state of OCCT.
02Author
Dear Mikhail,

All cases provided in context of this issue are invalid, e.g. contain wires with gaps covered by vertex tolerance, problematic geometry or topology.

However, the most of the cases attached here are meshed anyhow on current master with several exceptions.

test_mesh_standard_incmesh_V:
face139.brep
face156.brep
face79.brep

test_bugs_vis_buc60858:
face10.brep

Please make a decision regarding this issue, should we consider that the goals have been achieved as much as possible and close it?
03Commenter 3
I agree with conclusion about the tests bugs vis buc60858, mesh standard_incmesh V4.

But the test mesh standard_incmesh C9 is bad and TODO refers to the closed bug #0023105. If the problem with free nodes is inevitable with such input shape then we should mark this error as REQUIRED.

Related records