I have a feeling this is a simple yes or no answer but I am going to ask it anyways.
I am using a sequence to populate a given ClientID column in my table. This ClientID however can be persisted across multiple records within the table, each one representing a different version of the record, of which only a single is "Active".
My concern is that the first time I preform a revision on the record, which will "Delete" the record by setting it's Deleted bit, and create a new record in the same table with the new values. The new record will have a new ClientID value given to it where it should have persisted the previous ClientID value.
UID ClientID Name Deleted
20 1 Vendor A 0
21 2 Vendor B 0
If I were to add a new record (non-revision), I would expect the following:
UID ClientID Name Deleted
20 1 Vendor A 0
21 2 Vendor B 0
22 3 Vendor C 0
But if I were to want to rename (Vendor B) to say ABC Corp, I would want to mark record 22 as deleted and create a new record for "ABC Corp" with a ClientID of 2. (see Below)
UID ClientID Name Deleted
20 1 Vendor A 0
21 2 Vendor B 1
22 3 Vendor C 0
23 2 ABC Corp 0 <-- FK relationships are still in tact.
But with the sequence in place, I am expecting to see this instead...
UID ClientID Name Deleted
20 1 Vendor A 0
21 2 Vendor B 1
22 3 Vendor C 0
23 4 ABC Corp 0 <-- Incorrect ClientID (Loss of FK relationship)
Can you selectively override a computed column calculation and insert your own value in it's place?