Hi Bonnie,

The strange part of this (to me) is that you are using pessimistic row buffering which, as I read it, should work exactly as you expect without special steps of any kind.

That leads to the question... are you really really sure that you are using pessimistic row buffering on the table in question (and, of course, that the form field control sources *do* point to the actual table record fields)???

I see that some are trying to encourage you to go with an 'Edit" button but that, so far at least, you are resisting. I hope you persist with this.
This is one of countless interface "issues" Alan Cooper exposes in his book "About Face - The Essentials of user interface design". Thsi book was much lauded in the VFP community about 2 years ago. Yet virtually *none* of the issues raised has been even marginally addressed in Windows (interface) or the VFP interface capabilities, even though there has been a Win98 and a VFP 6 *since* its publication. I get the impression that we really like to read books (or at least promote them) but that we pay little service to the messages in those (much lauded) books.

Good luck,

Jim N

>Sylvain,
>Nope, not gonna work. I've tried several variations on this theme and tried placing code in several different methods (When, InteractiveChange), but nothing seems to trap it accurately.
>
>HELP! There's gotta be a way to do this! It's not a complicated task. Any other suggestions????
>
>Bonnie
>
>>How about a loop that goes through each control and compare the current value of the control to the value in the ControlSource? Something like (haven't tested it):
>>
>>

llDirty = .F.
>>
>>FOR EACH Controls IN THISFORM.Controls
>>   IF Controls.Value <> (Controls.ControlSource)
>>      llDirty = .T.
>>      EXIT
>>   ENDIF
>>ENDFOR
>>
>>IF llDirty
>>   *-- Record was modified. Maybe advise the user before doing a Refresh()
>>ELSE
>>   *-- Everything OK, start the edit mode
>>ENDIF

>>HTH
>>
>>>I know it might be confusing, that is why I was looking for some way to detect that the record had been changed so that I could notify the user of that prior to doing a refresh(). Any other ideas on how to trap for this?