I have refurbished the UI (still using the Fiori Elements preview) but using more field groups to allow for a more easy navigation.
Open RAP generator
The new version comes with a new UI which let you enter the basic project data upfront, such as the type of the datasource being used (table, cds_view or abstract_entity), the implementation type (managed or unmanaged), the binding type and whether the project shall be draft enabled or not.
In addition you have to provide the name of the data source that is used by the root entity, the name of the package where the repository objects shall be generated and the name of the root entity.
By providing this information upfront the generator is able to determine the abap language version of the datasource and the key layout (uuid or semantic key) so that all this information can be set upfront.
Open RAP Generator
The only thing that is then left to do on header level is to provide (optionally) a suffix or a prefix that will be added to the repository object names that are suggested by the RAP generator.
You can then continue to maintain the details of the root entity such as the mapping of those field names to their role in the RAP business objects that the generator was not able to determine automatically.
When adding a child node you are asked for the entity name and the name of the datasource so that again determinations can run before you start editing the entity details.
A new feature is also that you are now able to maintain the field names of the entities that are suggested by the generator.
In various customer events I learned that the generator is used by several customers and partners, more than I have been aware about so far.
Feedback is as always welcome and if you don't mind to share with me that you are using this tool, that would be great as well.