Hello,
Currently I have a sql server 2000 stored proc which executes queries
to return historical data for
a number of financial bonds in the following format.
Date OutputCol1 OutputCol2 OutputCol3
---------------------------------------------
1/1/90 3.2938 90.2930 94.019
1/2/90 3.2938 90.2930 94.019
The underlying query in the stored proc which returns the data as
above is in the following format.
Select a.CurveDate as Date,
b.TblCol1 as OutputCol1,
c.TblCol2 + d.TblCol3 + log(e.TblCol4) as OutputCol2
Each of these TblCols are retrieved from a different table based on
the joins on the Date field across all the tables.
As of now all these tables are within a single SQL server 2000
database.
We need to move all the math operations which we are currently doing
in stored proc to the middle tier(Either VB or C# class library
application). and use database just for retrieving raw data from the
tables.
Also We need to expand this functionality to get raw data for columns
from different data sources like SQL server, Oracle
sybase or a xml data file and perform math operations in the middle
tier and then return the processed data to the front end.
We were thinking of doing it in ADO.net which lets us get each of the
raw data columns(without
math experessions) into different oledbadapters and them filling each
into a dataset. I need to know if it is possible
to execute join queries on different tables of a dataset. Also I want
to perform math operations like
dataset.tables(0).columns(1) + log(dataset.tables(1).columns(2)).
If not do let me know if you have any other suggestions.
Thanks
-Mahesha