G
Guest
In perusing Microsoft’s Data Access Architecture guide I found a bit of a conundrum on concurrency. The document outlines 4 possible methods for updating data using a STDS.
1. Including only the primary key column
2. Including all columns in the WHERE claus
3. Including unique key columns and the timestamp column
4. Including unique key columns and the modified column
Of these four only the third one is really recommended by the document, but it also says that ADO.NET does not support this approach. We know that we want to use STDS in our project and it will exist in a disconnected data environment, and data integrity is paramount. So my question is do we have to implement this manually, or does someone have a “silver bullet†on this issue?
1. Including only the primary key column
2. Including all columns in the WHERE claus
3. Including unique key columns and the timestamp column
4. Including unique key columns and the modified column
Of these four only the third one is really recommended by the document, but it also says that ADO.NET does not support this approach. We know that we want to use STDS in our project and it will exist in a disconnected data environment, and data integrity is paramount. So my question is do we have to implement this manually, or does someone have a “silver bullet†on this issue?