Here is what I want to do:
Fill a dataset with one adapter using one connection. Use the same dataset to Update using a second adapter and a second connection. The only problem is that I cannot find a way to trick the dataset into thinking that all of the rows in its datatables were added and hence need added to the second connection. The RowState property is readonly.
Two practical applications:
1) I have an online and offline db so our salespeople can work on the road without being connected to our office. I want them to be able to copy all orders for identical tables in an online and offline (local) db.
2) We have standard models that we sell. Each order uses these standard specs as a "template", essentially copying a query of the models table -> the orders table.
Now, I have pulled this off by creating a second dataset, then iterating through each row in the first dataset and creating a identical rows in the second dataset. However, it seems like you should be able to skip (1) the redundant memory usage and (2) the time it takes to do this.
Any ideas?
Fill a dataset with one adapter using one connection. Use the same dataset to Update using a second adapter and a second connection. The only problem is that I cannot find a way to trick the dataset into thinking that all of the rows in its datatables were added and hence need added to the second connection. The RowState property is readonly.
Two practical applications:
1) I have an online and offline db so our salespeople can work on the road without being connected to our office. I want them to be able to copy all orders for identical tables in an online and offline (local) db.
2) We have standard models that we sell. Each order uses these standard specs as a "template", essentially copying a query of the models table -> the orders table.
Now, I have pulled this off by creating a second dataset, then iterating through each row in the first dataset and creating a identical rows in the second dataset. However, it seems like you should be able to skip (1) the redundant memory usage and (2) the time it takes to do this.
Any ideas?