On Sun, Oct 30, 2005 at 11:47:41PM +0800, Lincoln Yeoh wrote: > At 06:29 AM 10/30/2005 -0800, David Fetter wrote: > > >On Fri, Oct 28, 2005 at 09:57:03PM -0400, blackwater dev wrote: > >> In MySQL, I can use the replace statement which either updates > >> the data there or inserts it. Is there a comporable syntax to > >> use in postgreSQL? > > > >Not really, but here's an example which doesn't have the brokenness > >of MySQL's REPLACE INTO and doesn't have the race conditions that > >some others' proposals have. > > > >http://developer.postgresql.org/docs/postgres/plpgsql-control-structures.html#PLPGSQL-ERROR-TRAPPING > > Erm, doesn't it have the same race conditions? No, don't believe it does. Have you found some? > If you have the appropriate uniqueness constraint, you'll be fine > whatever you do as long as you're not doing something too stupid. > > If it is possible to lock on something that already exists or has > "yet to exist" then maybe you can do such an insert/update. > > e.g. "SELECT .... FOR INSERT WHERE field1=x" > > How about customizable user locking? e.g. lock string "tablename > field1=x" > > Anyway, I used to lock tables before doing selects before > inserts/updates. That's a real performance killer. Cheers, D -- David Fetter david@xxxxxxxxxx http://fetter.org/ phone: +1 510 893 6100 mobile: +1 415 235 3778 Remember to vote! ---------------------------(end of broadcast)--------------------------- TIP 9: In versions below 8.0, the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match
