I have registered bug 87 after verifying that the LIMIT clause does not meet the SQL92 spec.
However this error points to the larger question of unique row ids in the database, and by implication the unique IDs of all messages handled by Kannel as Stipe mentions below. Naturally database work against (properly indexed) unique ids will be far quicker than the existing system. Kind regards Rory On 19/09/03, Stipe Tolj ([EMAIL PROTECTED]) wrote: > > Hi Andreas. My point is that the LIMIT clauses are not optional for > > Postgres. You simply can't have them. > > ok, so this is a bug at least for PostgreSQL?! (I thought LIMIT would > be SQL92 sompliant?? hmmm.) > > Ok, Rory, can you do a bug report to http://bugs.kannel.org/ so we > keep track of this. > > > Is there a way we could do this update on a unique id rather than the > > timestamp (i.e. Kannel uses a value from the db rather than provide the > > key?). For large dbs this could be optimised to provide very quick > > performaance. > > yep, that's an issue we have to do. SMS that are MO or MT should have > unique IDs (even accross multiple Kannel instances). I have discussed > this some time ago with Alexander in a thread. > > We have to enforce this. > > @alex: should be possible to implement the UID (unique ID) to > msg->sms.id (changed into an OCTSTR) in all modules, do we?! -- Rory Campbell-Lange <[EMAIL PROTECTED]> <www.campbell-lange.net>
