Overflow when trying to assign Float to Double

Question

Please excuse me if this has already been anwered.
I'm trying to pull a long decimal (0.110494842038439 to be exact) from an
Oracle database table where it's stored as a Float.
Dim z As Double = rsAgg("st_dev")
Returns the error "Arithmetic operation resulted in an overflow."
Dim z As Oracle.DataAccess.Types.OracleDecimal = rsAgg.GetDecimal(2)
Seems like it should do it, but that one gives the same error.
Dim z As Oracle.DataAccess.Types.OracleDecimal = rsAgg.GetValue(2)
Same thing.
I'm running out of ideas, hopefully someone can point me in the right
direction.
Thanks!
-Dan

Answer 1

I know that using the SafeMapping feature you could get the value back either as a string or as a binary representation.
Converting from a string value is, as we both agree, inefficient especially for thousands of values in a loop.
The binary representation is easily converted to an OracleNumber using the constructor that takes an array of bytes. The problem is that I can't safely cast the OracleNumber to a decimal (System.Decimal) when the OracleNumber has a greater precision than a decimal will hold.
Now I could write my own function that operates on the byte array and converts the relevant portion to a decimal ignoring the extra precision. But it would be so much more clean, and definitely less buggy, if this were done for me in the OracleDataReader itself.
Just to clarify, the reason I want these values as decimals is because I want them in a DataSet or other structure that I want to serialise to a remote client - where the ODP is not installed. In which case I am restricted to native .NET types.