Denormalization of Tables

[VictoriaJones]

Help!!

I need to denormalise a new table in order to match it to a table from an old database, in order to validate the new data.

I currently have a list of "function id's" associated with both a series of "Products" (up to 3) and "Risks" (up to 4, plus a descriptor of "OTHER" choice), The choices of Products and Risks are from a drop down box which have been entered directly from the form, (except the "other" option which is from a text box).

In the old database, the products and risks were stored in different fields, such as Product 1, Product 2 and Product 3, and the same for the Risk Categories, within the same table. However in the new db they are all stored in one field "Product Category" and "Risk Category", in different tables, with their associated Function ID - thus creating multiple rows/records in the query/recordset.

So the question is - how can I get the new data separated out into 3 fields, or 4 for the Risk Data, from the esisting 1.

I hope this makes sense, and any help offered would be gratefully received!!;-)

 

[IForgot]

Are you really needing to make your new tables less efficient through de-normalizing them?

Or are you merely looking to transfer data from the old tables into the new ones and looking to make the task more simple?

If the latter, I would not recommend changing the new tables, but instead work the old table data into a new format to support the transfer/validate process. This new format may be just a temporary set of tables and need not impact the original application and might be created using an SQL query or other means. Note - it might be a multi-step process to get the old data into the proper format and not be done with one single operation.

Good Luck,
I_Forgot

 

[VictoriaJones]

Sorry, neither of the above!!!

I have inherited both db's, and we are currently using the older (inefficient) set of tables. However we need to be able to compare the data that has been uploaded from Sybase into the new table format. For this I need to take (even a temporary) recordset, that conforms to the old structure, so that I can reconcile the two data sets.

Due to a number of reasons I can't just normalise the old data set (very long set of reasons that I wont bore you with now!), but need to find a (relatively?!) quick way of creating a denormalised recordset from the new structure.

Any thoughts?
Victoria