I hate touchpads!!! I should have written "It probably says something about deferrable constraint use in the real world, that they first got introduced approximately a decade ago and they still get forgotten (and sometimes surprise people)" Niall On Tue, Jun 24, 2008 at 2:10 PM, Niall Litchfield < niall.litchfield@xxxxxxxxx> wrote: > The old rule still applies. In case of indexes, see if Richard Foote has > written anything first :) > http://richardfoote.wordpress.com/2008/06/04/primary-keys-and-non-unique-indexes-whats-really-happening/ > - > I've linked to that page as it contains the links to all 3 of his original > posts. There are some minor (or in some cases not so minor) differences > between the two. It probably says something about deferrable constraint use > that they first got approximately a decade ago and they still get forgotten > (and sometimes surprise people). > > > > > > On Tue, Jun 24, 2008 at 1:48 PM, Mark W. Farnham <mwf@xxxxxxxx> wrote: > >> I haven't had a chance to test whether a unique or primary key >> constraint enforced by a non-unique index performs differently from one >> enforced by an index created as unique, so I've always been reticent to >> suggest it. Has anyone else tested this thoroughly? >> >> >> -- >> Niall Litchfield >> Oracle DBA >> http://www.orawin.info >> >> -- Niall Litchfield Oracle DBA http://www.orawin.info