Re: [HACKERS] Re-Name Attributes on Inheritance

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Clark Evans <clark(dot)evans(at)manhattanproject(dot)com>
Cc: hackers(at)postgreSQL(dot)org
Subject: Re: [HACKERS] Re-Name Attributes on Inheritance
Date: 1999-03-10 15:47:05
Message-ID: 17615.921080825@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Clark Evans <clark(dot)evans(at)manhattanproject(dot)com> writes:
> How difficult would it be to allow attribute
> names to be aliased when inheriting:
> ...
> Thoughts? Is this _that_ bad of an idea.

It seems like a horrible idea to me ;-). The point of inheritance
is that whatever else your derived class may be, it *IS A* parent-
class object as well, and anything that works on the parent class
will work on the subclass. In your example,
SELECT more_attrib FROM base where oldtag = something;
would work but
SELECT more_attrib FROM derived where oldtag = something;
would fail. That's not my idea of a derived class.

It's not clear exactly what you want to accomplish here, but I
wonder whether inheritance is the right model for the relationships
among the tables in your database at all. It seems like you are
trying to force-fit some other kind of relationship into the
inheritance model.

regards, tom lane

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 1999-03-10 16:02:02 Re: [HACKERS] Re: map
Previous Message The Hermit Hacker 1999-03-10 15:41:35 Re: [HACKERS] Developers globe