meshlib - Creating merged water-tight mesh and cleaning data - zero-area triangles

[GlenEaton]

Hi MeshInspector Team,
Thank you for the great software and library tools for working with meshes.
I am currently working on a project in python where I am merging multiple mesh segments together with a boolean union to create a watertight mesh.
As part of the process I have to do some cleaning of the data along the way.
Currently I have used:
mrmeshpy.fixMultipleEdges(mesh_obj)
mrmeshpy.uniteCloseVertices(mesh_obj, 0)
to fix some of the issues during the merging.
Everything is working and coming out well, however when I check the merged mesh in MeshInspector, some of the files have zero_area triangles.
I have tried to add:
mrmeshpy.fixMeshDegeneracies(mesh_obj, params) - (using 0.0000001 value)
during the processing but it made the result worse and caused segments that aren't close together to interpolate between them.

I think I need to run it at the end on the final mesh, not during the generation process as per the other corrections which are fixed during the merging process (iteratively when error identified).

Would mrmeshpy.fixMeshDegeneracies(mesh_obj, params) be the correct process to run to fix zero-area triangles and do I need to run a fill holes or something else afterwards to ensure the mesh is watertight still?
Should I run it at the end of the merge only or is there a way to identify if there are zero-area triangles after each merge iteration?

Any advice or suggestions would be greatly appreciated.
Regards,
Glen.

[Grantim]

Hello!

fixMeshDegeneracies designed to fix such issues, and it should preserve mesh connectivity, so it is kind of strange (or probably buggy) that mesh loses watertightness.

Could you please provide exact script you are using and meshes to reproduce the result? (if you don't want to share data publicly, you can send it to or email [email protected])

Also, please note that we have function for uniting several meshes at once:
https://meshlib.io/documentation/Py/namespacemrmeshpy.html#a5d8cd610c03300e6d33bddf16ab56960

[GlenEaton]

Hi Grantim,
I will email you the code, although its a bit long and for specific type of data. I will see if I can email you a data sample also. Can I email you "ply" or "zip" file. The sample data is only 16kb.
Regards,
Glen.

[Grantim]

Waiting for your email, thanks!