Not quite, Rachel. It keeps incrementing, so testing for > 1 only works the first time. When you retrieve any row, its Sys_Rowver value could be any integer. When you save the row back, it will be that value + 1. A row that's been edited 20 times might have a sys_rowver of 21. In web applications, when a user can be editing the detail, my SELECT command that gets the data for the html forms also gets the sys_rowver. Then, when the user presses "Save changes", the UPDATE command can have the equivalent of "IDColumn = .vIdColumn AND sys_rowver = .vSys_Rowver" If it doesn't find the row, (and SQLCODE says so), it means somebody else probably updated it in between. So the application can take appropriate action: "Somebody else just changed the Zip Code to 56085. Do you want to overwrite that change with 56058?" etc.
On Fri, Mar 13, 2009 at 12:17 PM, Rachael Malberg < [email protected]> wrote: > DITTO!! this cool!! is there a sql statement we could do? something > like... > > update PermTable set PTCol=TTCol from PermTable t1, TempTable T2 where > t1.ID=t2.ID and t2.SYS_ROWVER>1 > > or just cursor threw the recs? > > ----- Original Message ----- > > *From:* Dennis McGrath <[email protected]> > *To:* RBASE-L Mailing List <[email protected]> > *Sent:* Friday, March 13, 2009 8:46 AM > *Subject:* [RBASE-L] - Re: How to detect a dirty table? > > My sentiments too! > > > > Dennis McGrath > > > ------------------------------ > > *From:* [email protected] [mailto:[email protected]] *On Behalf Of * > [email protected] > *Sent:* Friday, March 13, 2009 8:43 AM > *To:* RBASE-L Mailing List > *Subject:* [RBASE-L] - Re: How to detect a dirty table? > > > > Doug: Cool! For years I've considered that column useless, and would > usually set it permanently off. Never considered turning it on and off to > track whether changes are made to a temp table. I'm glad you brought it up > because I never would have thought of it on my own! > > Karen > > > > Bingo Karen! And it isn't as complicated (for my needs) as you suggested. > > SYS_ROWVER basically does what Emmitt suggested with triggers but all the > heavy lifting is done by R:Base: it adds the SYS_ROWVER column, sets it to 1 > for all rows, creates the trigger and increments SYS_ROWVER each time a col > in a row is changed. So any changed row will have a SYS_ROWVER value >1, I > don't even need to compare it to the original table. woohoo! > > > >
