`StructureData.cell` in `dir()` but raises `AttributeError` about aiida-atomistic HOT 2 OPEN

yes, for now I have not deleted direct attributes like cell, pbc and so on, which are of the old StructureData. But this are no more working and are not related to the properties. I just left them there because I am not sure we really want to put everything under the properties attribute. In that case, the true structuredata node will be contained in the properties attribute, and nothing else is left outside (apart the Data methods)

from aiida-atomistic.

I just left them there because I am not sure we really want to put everything under the properties attribute. In that case, the true structuredata node will be contained in the properties attribute, and nothing else is left outside (apart the Data methods)

Yes, this is also an observation I was making, and would make certain operations more cumbersome than they have to be. My first inclination is that I would really like to have structure.cell. But then if we treat all properties equally, all of them should have a corresponding attribute. And maybe that's ok, but then what's the point of the properties namespace?

I'm not even considering the implementation here. Just the API.

from aiida-atomistic.