On Fri, Nov 11, 2005 at 06:00:32PM -0500, Rod Taylor wrote:
> So? That is what save points are for. You can even skip the select for
> update if you don't mind dead tuples from the attempted insert.
> SELECT ... FOR UPDATE;
> IF not exists THEN
> SAVEPOINT;
> INSERT ;
> IF UNIQUE VIOLATION THEN
> /* Someone else inserted between the SELECT and our INSERT */
> ROLLBACK TO SAVEPOINT;
> UPDATE;
> ELSE
> RELEASE SAVEPOINT;
> FI
> ELSE
> UPDATE;
> FI
Isn't there still a race between INSERT and UPDATE?
Low probability, for sure, as it would have had to not exist, then
exist, then not exist, but still possible.
I'd like a REPLACE that could be safe, or at least cause a COMMIT to
fail, for this reason.
Cheers,
mark
--
mark@mielke.cc / markm@ncf.ca / markm@nortel.com __________________________
. . _ ._ . . .__ . . ._. .__ . . . .__ | Neighbourhood Coder
|\/| |_| |_| |/ |_ |\/| | |_ | |/ |_ |
| | | | | \ | \ |__ . | | .|. |__ |__ | \ |__ | Ottawa, Ontario, Canada
One ring to rule them all, one ring to find them, one ring to bring them all and in the darkness
bindthem...
http://mark.mielke.cc/