-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 02/28/07 10:31, Joshua D. Drake wrote: >>>> I would imagine that other DBMSes also enforce uniqueness by means of >>>> indexes, because it'd be awful darn expensive to enforce the constraint >>>> without one; but I'm only guessing here, not having looked. Can anyone >>>> point to a real system that enforces unique constraints without an >>>> underlying index? >> >> In Rdb/VMS (which does not use MVCC), PK (and it's alias UNIQUE) >> constraints are independent of whether you have a unique index on >> the table. > > PK is NOT an alias for UNIQUE. Yes it does have the same functional > operation but it is technically incorrect to consider them the same. In Rdb/VMS, which I was describing, the PK and UNIQUE constraints (which are *not* the same as unique index) *are* >> Now, 99.44% of the time you will *not* have a PK constraint, but >> simply a unique index. > > Then you have designed your database incorrectly. Or... *you* don't understand Rdb, and the circumstances in which it is used. In Rdb, a defining a PK has no automagic side effects (Which I heartily approve of). The DBA is responsible for knowing the data and determining the best (of multiple) way to ensuring that *that* set of data is. So, if you would already have put a unique index on that table, there's no reason to also put a PK constraint on in (unless there will also be an FK reference). -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFF5ba+S9HxQb37XmcRApEOAJ9x6pco5kgqc2alEVGlEHRyOyC2WQCfQCp4 JaXCNqn0UgJGl91Kb4Suq54= =tAbk -----END PGP SIGNATURE-----
