I have several reports that I already have made on Desktop and published in service, my Data source is a custom connector we made. The connector is intended to be updated to have more and more (but also some can be removed) endpoints that are converted in tables as the time goes by.
But I’m wondering how we should handle the forwards/backward compatibility, what would happen if a report made with the version 1.0 of the connector tries to refresh let’s say with the version 3.0 of the connector that removes certain tables that were part of the queries in the report.
What is the common or proper way to handle this kind of scenario? It can be made from the source code of the connector, or the Data analyst must fix the broken queries and update the reports manually?
