MontruJdbc_D5E1: If it works, take it apart... (a serious one, ColumnInfo refactor)
The initial ColumnInfo class was useful for prototyping because in the DB response, there is no difference between attributes and links. However, in Dust that is fundamentally different and as the related code got stronger, there were many conditional segments in the code. I have created separate PopulatedFactory instances for them, now I make ColumnInfo class abstract, and move specific behaviors into the derived Att/Link specific classes. Then, I also moved the temporal, String-based attribute access service that I use in the Connector to utility level for later reuse in similar prototyping tasks.