1 Reply Latest reply: Aug 3, 2013 4:27 PM by dbomp RSS

    Data Modeler accepts Duplicate Columns and silently changes the name

    user12570281

      When adding a column to a multi-column-table it sometimes happens that by chance, by typo or otherwise an existing column is added to the table.

       

      E.g. when I add a new column C_DUP and a column named C_DUP is already in the table the DM generates a name like C_DUP_2 and puts the column into the table. This avoids - or better masks - the underlying design error and is rarely useful.

       

      Is it possible to change this behavior ?

        • 1. Re: Data Modeler accepts Duplicate Columns and silently changes the name
          dbomp

          In the relational or logical model?

           

           

          My problem is that when engineering from the logical to relational model, it creates duplicates over and over again.  So, if the tables in both the logical and relational models have twelve attributes/columns, and one in the logical model has a "preferred abbreviation" called "active_flag", and its relational equivalent (which was created by engineering in the first place) has a column called "active_flag", after engineering the relational table will have thirteen columns, a new one at the end called "active_flag_2".  If I engineer again, I get an "active_flag_3".  My model has about 100 tables now, and I've turned off engineering on a third of them because of all the duplicates I get that I have to keep deleting.  (Curiously, "active_flag" is a common one to get duplicated.)  (3.3.0.747)