For developing a DB application, .NET provides the FillSchema method for populating certain aspects of a DataSet's schema. This includes the PK column, allow NULL and several other properties.
However, FillSchema will not populate the AutoIncrementStep, AutoIncrementSeed or FK information.
(Speaking now to just the FK issue)
Is it just me, or does this seem like one of the stupidest design decisions of all times? This means that as a developer, I must maintain the relationship in two separate places (in the DB and in my code) using two different tools!
Similarly, some comments I have read suggest that using FillSchema is inefficient and developers should programmatically define the PK within the DataSet. Again, this seems incredibly stupid.
I would like to hope that there is a rational explanation for this, but I cannot fathom it. Is there one?
----
Gerry Roston
gerry@pairofdocs.net
However, FillSchema will not populate the AutoIncrementStep, AutoIncrementSeed or FK information.
(Speaking now to just the FK issue)
Is it just me, or does this seem like one of the stupidest design decisions of all times? This means that as a developer, I must maintain the relationship in two separate places (in the DB and in my code) using two different tools!
Similarly, some comments I have read suggest that using FillSchema is inefficient and developers should programmatically define the PK within the DataSet. Again, this seems incredibly stupid.
I would like to hope that there is a rational explanation for this, but I cannot fathom it. Is there one?
----
Gerry Roston
gerry@pairofdocs.net
