T
tim
Hi,
I recently created a generic data access layer so I can use it in many
applications. It is designed to return datasets, datareaders, scalar
values... I recently have come across in an application that uses the data
access layer needing a datatable instead of a dataset. What I have been
doing is returning the dataset with only one datatable and then accessing
the datatable in code (ds.tables[0]). I have been thinking about changing
my data access layer to return datatables. However with that said, is it
worth the redesign? Will I see better performance? Do you think it is a
bad practice to fill a dataset with only one table?
Thanks
I recently created a generic data access layer so I can use it in many
applications. It is designed to return datasets, datareaders, scalar
values... I recently have come across in an application that uses the data
access layer needing a datatable instead of a dataset. What I have been
doing is returning the dataset with only one datatable and then accessing
the datatable in code (ds.tables[0]). I have been thinking about changing
my data access layer to return datatables. However with that said, is it
worth the redesign? Will I see better performance? Do you think it is a
bad practice to fill a dataset with only one table?
Thanks