Thank you @Coffee77 

This is exactly what I want to achieve, that while in my relational database I have multiple tables (current + historical), I want to setup a pipeline that only one table exists in databricks for all the records, but I am not sure what the best practice is to achieve this.

Furthermore, when a record is “archived” in my relational database, it will be seen by CDC as “delete from current table, then insert to historical table”, and that totally messed up my current pipeline. Ideally, I would like to see my databricks table to have a column to indicate the record is either “current” or “historical”.