Karen is right. Grab the data immediately after the insert. It should work correctly. It is designed to be multi-user friendly. Dennis McGrath
________________________________ From: [email protected] [mailto:[email protected]] On Behalf Of [email protected] Sent: Thursday, January 28, 2010 2:13 PM To: RBASE-L Mailing List Subject: [RBASE-L] - Re: where COUNT=INSERT Bob: I've used count = insert in tables the size you're talking about and haven't heard anything about poor performance. However, probably only with 2 or 3 hitting the same table at the same time. Haven't been notified of any problems where it might be grabbing the wrong insert. As for your #4, I have always done my count = insert immediately after the insert into that table, never farther down in the programming. Don't know that it wouldn't work later, but I have always selected the count immediately. Karen I am planning to use the where statement : Where COUNT=INSERT to obtain the value in a row that has just been inserted by the user. There will be 20+ users running this app continuously during the day and often inserts will be made at the "same time". The table will grow to medium size, probably around 100,000 records and remain at that size. I have used this syntax before, but not in conditions where the row count will be as high, where there will be as many concurrent users inserting records and where speed is of great importance. 1)Are there any considerations I should take using this syntax? 2)Does this use an index? 3)The speed to retrieve the value will need to be very fast and this will be running on a network. 4) I assume that the "Insert" condition is table specific. I.E. a user inserts a row in table A and then a new record in table B. The Where count = Insert for Table A would still be correct even after Table B had an insert. Any thoughts as to the above? Thanks, -Bob
