Remember that a group in Rhino is more like a selection. It is not an actual object. I don’t want to merge objects that are in a Rhino group into one, as that will not work.
But you should be able to use the collections to select the groups you want to select.
this is how they should come in, you can select them by right clicking the collection and choosing select objects
edit: ahhh, nathan was faster again 
I am not able to switch link material slot from data to object at once in Blender.
Shift + G collection selection is not useful to change material slot from data to object. Every time when I re-import rhino mesh, I need to re-arrange materials. If I want to use 2 different material at once in blender when I update model it turns back beginning. Thats why I use object slot in Blender 
If there would be merge option my problem might be solved.
I’ll have to think that over for a bit. Again, a group isn’t an object, so I am hesitant to merge objects in groups. Especially since you can group different objects that already have different materials.
Perhaps you just need a way to quickly assign a certain material to a selection of objects.
Can you explain what setting the material to Object does/how it would help? Afaik it only changes the assignment behaviour in case of linked duplicates. For the instance collections there is no way to override materials, correct me if I’m wrong here…
Library overrides would be the correct way to do this, but alas library overrides is still incomplete. There are quite a bit of open issues and unimplemented bits and pieces: https://developer.blender.org/T53500
Again, my purpose is model in rhino and Archiviz in Blender but the workflow is not going well. If I re-upload the mesh, the maps in blender disappear.
Ah thats because Uvmaps are still not supported. Basic uv support is in the works, but custom mappings are another thing…
I sent you a private message. Could you check it?
@nathan: do we have any re-import strategy for these cases?
The strategy hasn’t been fully fleshed out, it certainly can use more support.
The Runtime Serial Number is not something to look at. You need the Id (GUID). This we already save as custom object data. We should use this information in determining what already is in the file, what not. This rather than object name, especially considering in Rhino many objects can have the same name (but not in Blender).
i know we tag objects with the guid, but it doesn’t give us any information about geometry changes, does it? so we’d basically just check if it’s already there and then act based on the options selected to update or discard materials/mappings etc? Probably best to move this discussion to a git issue 
Right, my point is that we could calculate some extra data that we can use to track.
For instance CRC of the geometry, texture coordinates, etc. It’ll take processing power, but should allows us to track better what goes on.
sounds like another thing that will wait for rhino inside 
maybe we could consider an option to only import new objects in the meantime?
Sounds like a plan.
If the addon re-upload the mesh and keep the UV map coordinates, the problem will be solved.
For architecture projects it works pretty perfect when you combine it with the script. By that I mean the usual situation where you get a finished rhino model and you need to move it to blender for the renderings. I am still testing it but I find it very promising. Thank you so much
Honestly what would be really nice here is some way to know that the data for a RhinoObject has changed, either in the Rhinodoc or at the RhinoObject level. Tracking changed objects with RhinoCommon is no trivial task and involves any number of events, including knowing that moving an object is the same as deleting and adding a new one (fundamentally anyway). It would be nice if the document could tag the object in some way as changed or “versioned” and save that in the file.
This could be done with a “helper plugin” through RhinoCommon but I would find it useful in other cases as well if rhino did this natively in some way without adding extra bloat to the file size.
maybe something like… buzzword buzzword blockchain?? 
