Open pfebrer opened 5 months ago
This could be cool, but I think it has very low priority. There are soo many issues open that I think we should try to finish some of them ;)
I also can't really see who will use it? Much of the functional methods are also done in other codes, with slightly different API's, so my bet is that if somebody is using ase
, then they would stick with that. :)
I would see this being useful if some function is not implemented in ASE, or it has an API that is more complicated to acheive a certain goal. If someone asks how to acheive that thing in Matter Modelling, then you can answer: "use this function in sisl". And you will have introduced sisl to them with minimal friction :)
I don't really mind if it is a small pr, but I don't think you should spend tons of time on this, I would still prefer some help on thinking about the categories, and how we can solve that issue ;)
I will leave the issue here, I don't want to work on it for now, it was just an idea.
E.g. in the case of
Geometry
there are implementations to convert from/to ASE atoms. So with a simple wrapper it should be easy to support:I think it would be quite cool. It would be similar to how
numpy
can apply operations on many objects, although the implementation would be different in the sense that is based onsisl
implementing the converters.