Re: pg_depend

From: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
To: Peter Eisentraut <peter_e(at)gmx(dot)net>
Cc: PostgreSQL-development <pgsql-hackers(at)postgreSQL(dot)org>
Subject: Re: pg_depend
Date: 2001-07-16 20:11:28
Message-ID: 200107162011.f6GKBSp02389@candle.pha.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

> Bruce Momjian writes:
>
> > I have found that many TODO items would benefit from a pg_depend table
> > that tracks object dependencies. TODO updated.
>
> I'm not so convinced on that idea. Assume you're dropping object foo.
> You look at pg_depend and see that objects 145928, 264792, and 1893723
> depend on it. Great, what do you do now?
>
> Every system catalog (except the really badly designed ones) already
> contains dependency information. What might help is that we make the
> internal API for altering and dropping any kind of object more consistent
> and general so that they can call each other in the dependency case.
> (E.g., make sure none of them require whereToSendOutput or parser state as
> an argument.)

Yes, it is not simple. The table is just one part of it. Code has to
do lookups and have cascade/failure options based on what it finds.

Things can get quite complicated, especially circular dependencies. It
needs a general overhaul and has to hit every area. We need a central
location to keep all this info.

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Eisentraut 2001-07-16 20:40:31 Re: NetBSD 1.5.1(HP300)
Previous Message Peter Eisentraut 2001-07-16 20:03:50 Re: [HACKERS] Translators wanted