I unloaded Razzak's TrigTestDB. While the procedures are loaded before any data, the procedures are not added to tables as triggers until the end, so this looks safe.
That is, unless I am missing something. ________________________________ From: [email protected] [mailto:[email protected]] On Behalf Of Dennis McGrath Sent: Wednesday, January 13, 2010 1:25 PM To: RBASE-L Mailing List Subject: [RBASE-L] - Re: date tracking question Hmm, I'll have to check when an unload/load sequence actually puts the triggers in the table. If it puts it in before loading the data, ouch. If it puts it in after all data is loaded then no problem. ________________________________ From: [email protected] [mailto:[email protected]] On Behalf Of Lawrence Lustig Sent: Wednesday, January 13, 2010 1:21 PM To: RBASE-L Mailing List Subject: [RBASE-L] - Re: date tracking question << I want this value to survive a reload or an unload/load. It believe this rules out a computed column Is a trigger the only way to accomplish this? >> I believe that LOADing data during an unload/load is going to generate the same INSERT trigger execution that you would need to timestamp the record in the first place. In that case, a trigger wouldn't help, unless you communicate to the trigger using a global variable that a load is in process. -- Larry
