Igor Dejanović

Discussion continued on #297

Good idea but not sure where this should fit. Should it be done in the core textX or should we do that in textX-LS? We should definitely have some generic...

Actually, we have two classes of users, DSL designers and DSL users. We are targeting both with our tools.

Maybe we need a call in textX to properly invalidate cache based on the path that has been changed. In that way we leave to the user of the library...

I think you are right. It looks like a potential problem. The metamodel should be consistent with the parsing results. Since `NotS3` instance is never instantiated during `Base` parsing (i.e....

No problem. :) Yes, It seems that you are right. I'll have to do some more investigation.

Definitely agree on this. :) Would be nice to have a complete example/tutorial with imports and scoping. It would make multi-(meta)-model and scoping features more visible to the users.

> Maybe this unittest could be migrated to the examples folder? test_multi_metamodel_types_data_flow1 Probably we should start with multi-model and scoping tutorial. I agree that entity example would be a good...

Report rule name only if the match failed on the left-most match in the rule definition.

Yes, you are definitely right. Getting line/col is not very intuitive. One would have to do something like `get_model(obj)._tx_parser.pos_to_linecol(obj._tx_position)` which is not either easy nor readable. The main idea of...