My concern there would be the type of update statement generated. Would it
try to include as "original" column data the column as it looked before the
second modification? If the update statement includes only the Primary Key
it would be OK but I thought it tried to include all changed columns by
default?
Thanks,
James K.
"Tim Ellison" <ti*******@earthlink.net> wrote in message
news:O6*************@TK2MSFTNGP09.phx.gbl...
Try adding the row, then calling AcceptChanges() on the row, then
modifying one of the columns.
--
Regards,
Tim Ellison
"-=JLK=-" <No***********@nospam.com> wrote in message
news:uJ**************@TK2MSFTNGP11.phx.gbl... All,
Is there a way to add a new row to a dataset but make it think it is
changed data and should perform an "Update" rather than insert when the
update() method is called?
Thanks,
JamesK