Re: Surrogate primary key plus unique constraint vs. natural primary key: data integrity?

Hi Wolfgang,

your concerns seem to have a contradiction:

> I've always been used to using exclusively natural (comosite) primary
> keys, to warrant data integrity. Now computer science people (I'm just
> an engineer) keep telling me that concerning data integrity, it would be
> absolutely equivalent to use a surrogate primary key together with a
> unique constraint on the natural key.

Note that they say "absolutely equivalent". Though I have never heard of a "relative equivalence" this means that what applies to A applies to B and vice versa.

> Or could anyone give an example where the approach of using a surrogate
> primary key together with a unique constraint on the natural key would
> "break" data integrity (especially referential integrity) which could be
> avoided by using a natural (composite) primary key?

When the approaches are, in fact, equivalent such an example cannot exist.

Hope that helped.

Norbert
an engineer, too :) Received on Fri Mar 08 2013 - 20:37:07 CET