We're going to be upgrading our package to the SQL Development Bundle where we're currently only using SQL Compare.
Now i've downloaded the trail to start playing around with some things and one of the things I find very interesting is the Data Generator.
Sadly i'm trying to generate some data on one of our databases and i'm getting no joy at the moment because the way our databases are designed. For example we have the following design for addresses.
Address (base table, which holds an IsPostBoxOffice boolean, defining the type)
PostOfficeBoxAddress (holding only the PO box number)
GeographicalAddress (holding a house number, street)
Now, in order to make sure a PO only references a Address which has IsPostBoxOffice as true, this field combined with the ID is added to a Unique Constraint. This means in the PO table we have a IsPostBoxOffice field which is computed (always 1) and both the ID and IsPostBoxOffice field are referencing the Address table.
This ensures anyrecord entered in the specific subtables (Geographical & PostOfficeBox) will ALWAYS reference a correct record in the Address table.
This design might not be hugely common but it is (by my knowledge) the only way to design this efficiently.
Sadly the generator get's nothing but FK reference exceptions
Naturally I could edit the design and hope no one is dumb enough to make mistakes, but I kinda like having constraints to stop some of the stupidity in the world.
Is there a way to fix this with a custom generator, is this something that might need to be fixed in the Data Generator tool completely? What could I do to make this tool as usefull for us as I believe it could be!