Closed valikos closed 11 years ago
There will be a virtus plugin for ROM soon
Oh, that's great, thanks What about the schema if I understand correctly it is needed to describe the tuples for relational algebra We have two definitions: attributes for schema and attributes name list from schema for model I correctly think?
@valikos yes, for now you would define the in-memory store as well as the mapping to your objects.
We do think that at some point it will be possible to perform inference so that we can minimize the coupling between layers. I don't think we would be doing this implicitly though, it would probably be explicit.
For example, in theory if you were to define a model using virtus, along with a mapper, then there would probably be enough information for the in-memory schema to be defined.
An other idea, when it is ready, would be to infer the schema from an existing database. Aside from the obvious benefit of saving some typing to write the code for the schema, it should be possible to check the mapper against the schema and warn/raise if there are any mismatches. If there was a virtus model, we could probably take this further and assert all three layers are compatible. This way if there are incompatible changes at one of the layers we could know up-front rather than at runtime.
For now we're requiring that things be declared explicitly for now, but it should be possible to cut down on code in the future.
@dkubb thanks a lot for the explanation
Hello there!
I have a question about attributes We have a schema
Is there a way to take the attributes and pass them to the model? Because we can have a lot of attributes
Or whether in ROM support of Virtus for extend model?