vtkMapper.setResolveCoincidentTopologyPolygonOffsetParameters errors

[PaulHax]

Context

When calling the static mapper vtkMapper.setResolveCoincidentTopologyPolygonOffsetParameters, there is an error because the "class" does not have a publicAPI with a modified() function.

Results

Does not error now when calling vtkMapper.setResolveCoincidentTopologyPolygonOffsetParameters.

Some other tests are failing after adding the test in this PR. Seems that calling setResolveCoincidentTopologyPolygonOffsetParameters with the initial results of getResolveCoincidentTopologyPolygonOffsetParameters does not completely undo the effect?

Changes

The error fix involved adding a noOp modified function. Alternatives could be to check if publicAPI has modified in macros.js or upgrade the Mapper static object with macros.js' function obj(publicAPI = {}, model = {}) "constructor."

PR and Code Checklist

• semantic-release commit messages
• Run npm run reformat to have correctly formatted code

Testing

• This change adds or fixes unit tests
• Tested environment:
  • vtk.js:
  • OS:
  • Browser:

[PaulHax]

I guess the open questions/todos on this issue/PR:

• Do we want to keep supporting the vtkMapper.setResolveCoincidentTopologyPolygonOffsetParameters function? We are using it in VolView: https://github.com/Kitware/VolView/blob/main/src/main.js#L34
• Is this "fix" acceptable: adding the noOp modified function?
• How to test vtkMapper.setResolveCoincidentTopologyPolygonOffsetParameters without affecting other tests?

[finetjul]

Do we want to keep supporting the vtkMapper.setResolveCoincidentTopologyPolygonOffsetParameters function? We are using it in VolView: https://github.com/Kitware/VolView/blob/main/src/main.js#L34

I think so, there is also a Static ResolveCoincidentTopology in VTK C++.

[finetjul]

Shouldn't you expose a "static" MTime ? so that it can be considered when deciding to building/updating the shader ?

[PaulHax]

Shouldn't you expose a "static" MTime ? so that it can be considered when deciding to building/updating the shader ?

I think that makes sense as the global/static factor/offset are added with instance values as a uniform for the mappers shaders. How to do that? Upgrade the static object to be a full macros.obj, then each mapper registers a onModified for the static object and updates its own MTime when the static object changes?

How to test vtkMapper.setResolveCoincidentTopologyPolygonOffsetParameters without affecting other tests?

Do you have any insight on that?

[finetjul]

I think that makes sense as the global/static factor/offset are added with instance values as a uniform for the mappers shaders. How to do that? Upgrade the static object to be a full macros.obj, then each mapper registers a onModified for the static object and updates its own MTime when the static object changes?

Yes, that makes sense

[finetjul]

How to test vtkMapper.setResolveCoincidentTopologyPolygonOffsetParameters without affecting other tests?
Do you have any insight on that?

I'm not sure what the issue is TBH