Yo Craig

Thanks for your reply.

>Mark is correct. Remember that a view is always located on the user's PC, so it makes no sense to do locking on the view. When you use pesimistic locking, it will lock the tables...not the view.

Let me get this right. You set the view properties to use pessimistic locking and it has the effect of locking the records in the underlying table on which the view is based. I assume this is correct as you can not initially apply the pessimistic locking to the table as in the first instance, you are unaware of the records that will be selected in the view until the query is executed on which the view is based. If I am correct so far, then would you be good enough to explain what happens if one of the records that is selected by the view's query is actually locked at table level by another user. The locked record may be available in the view on a "read only" basis but a lock could not be applied as one is applied elsewhere by another user.

Therefore, how can I be certain that when I populate my view with pessimistic locking applied, that the records selected from the underlying table can all be locked whilst they are in use by the view? Will the view report some form of error condition that pessimistic locking could not be fully achieved on the selected data because one or more of the table records were already locked by another process when the view's query was executed?

I very much look forward to your advices.