Re: [HACKERS] FOREIGN KEY revisited

On Fri, 28 Aug 1998, Thomas G. Lockhart wrote:
> Back to your question: Postgres probably does not remove trigger
> functions if a table is destroyed, but that isn't a problem if the
> trigger function is generic code which will be reused anyway.

Even if you could create a trigger to remove the triggers used to
implement FOREIGN KEYs, how would you remove the trigger that removes the
triggers? :)

What is needed is some way of identifying a particular trigger with a
table and dropping it if the table goes away.

This could be sufficiently genericized to allow any dependant object to be
removed when its dependancy is droped.

> Vadim has been thinking about how to do foreign keys, but I can't
> remember if it was via triggers or some other means.

I'm the lone PostgreSQL advocate at a rather large backbone provider. I'd
love to keep using postgres, but lack of FOREIGN KEYs is probably the
biggest nail in the coffin. Oracle would be overkill for this project but
it would make DB maintainence much easier as I wouldn't have to keep track
of all these triggers I'm using to implement FOREIGN KEYs. The
applications that use the DB are considered 'hostile' and I need to do all
the constraint checking I can to make sure users don't do stupid things so
not using FOREIGN KEYs isn't a solution.

--
| Matthew N. Dodd |This space | '78 Datsun 280Z | FreeBSD/NetBSD/Sprite/VMS |
| winter(at)jurai(dot)net |is for rent| '84 Volvo 245DL | ix86,sparc,m68k,pmax,vax |
| http://www.jurai.net/~winter | Are you k-rad elite enough for my webpage? |