> There are basically two ways to attack the problem of a database-wide > unique ID: > > * Use a single int8 sequence for the whole database; > > * Use the combination of table OID and row OID (or, perhaps, an int8 > sequence for the row identifier, if you need more than a billion or > so rows in the table). There's a third way: int8 serial, lower 48 bits for row id, upper 16 bits appended with your own table id. -- Scott Ribe scott_ribe@xxxxxxxxxxxxxxx http://www.killerbytes.com/ (303) 722-0567 voice
