[RBASE-L] - Re: Can't read updated table column

[Doug Hamilton]

Karen - I owe ya big time!  That was it. I was just getting ready to unload and input the db as a desperate last resort. I'd thought PROPERTY... REFRESH, POST, CLOSE, OPEN and permutations thereof would've worked forced a re-read of the table. The fact that the BROWSE command made it work was a bit of a tipoff for me but it would've been a while before I stumbled on using the UPDATE. While not a bug, it's an obscure anomoly IMHO.  Does it still exist in 9.1? Thanks again to Steve, Jan, Javier & Dennis. Doug karentel...@cs.com wrote: Doug:  What you said just reminded me of something...  It's possible that RBase thinks it has the answer in its cache so it doesn't go back out and retrieve the new value.  I would think the refresh, etc., would work...  But in the past when I noticed this happen, I would do this before my 'select into':        update <tablename> set <anycolumn> = <samecolumn> where limit = 1 That would always flush the memory cache and force Rbase to get the value. I've never understood why some 'select into' commands would have this problems, while the vast majority never did... Karen I tried the REFRESH before the SELECT INTO, the POST after the UPDATE, and various combinations. I also tried: SET VAR vBatesNbrLock = BatesNbrLock +   IN Control  + WHERE ControlDescr = 'Data' It didn't work. But here's a curious...I put a BRO ALL FROM CONTROL at the beginning of the code to make sure the table was updated.  That apparently forces a read of the table and the code works.