I do have some comments inline below.
I'm not sure I agree with this. What would a "complete modeling experience" include? If it includes everything required for, say, editing the models in the UI, I think it would be going too far. To me, the point of the façade (or any language façade for that matter) is to provide an interface to the core of the language, and specifically with the necessary API to create model elements and navigate them, and perhaps support for loading and serializing, but that's about it, rather than providing the infrastructure to support everything else. Papyrus-RT is not only the tooling. Codegen is part of it too, and since the beginning, a key requirement has been that Codegen must be independent of tooling. Furthermore, as Patrik has pointed out, there are concerns about the use of the façade in a non-OSGi context. So I would be concerned about a complete modelling experience being supported at this level. I assume that the needs of tooling are more complex that those of codegen, but if this is the case, I think that the core should include only the minimal common needed functionality, rather than the larger. If all the functionality needed by tooling is added to this façade, wouldn't we end up in the same situation that we had were codegen could not use the API?
So I would be in favour of keeping the façade as light-weight as possible. If it doesn't meet the needs of tooling, then perhaps tooling needs a layer around this façade, but separate.
I think the important thing now is to define what is a "complete modelling experience", and decide if that's what the façade is supposed to support.
_______________________________________________
papyrus-rt-dev mailing list
papyrus-rt-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/papyrus-rt-dev