Re: INHERITS doesn't offer enough functionality

From: Chris <chris(at)bitmead(dot)com>
To: marten(at)toppoint(dot)de
Cc: Hiroshi Inoue <Inoue(at)tpf(dot)co(dot)jp>, Oliver Elphick <olly(at)lfix(dot)co(dot)uk>, Alfred Perlstein <bright(at)wintelcom(dot)net>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: INHERITS doesn't offer enough functionality
Date: 2000-10-23 21:13:28
Message-ID: 39F4A9F8.B24C1835@bitmead.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Marten Feldtmann wrote:

> > You'll still have to do 6 queries in postgres because it does not return
> > fields in sub-classes.
>
> Practically this is not such a big problem as one might think.

> WHEN you have a persistance framework you tell your framework,
> that every attribut is located (mapped or stored or however you
> may see it) in the superclass and then your top class (table)
> helds all attributes your "lowest" subclass has.

I don't understand what you're saying. There is no query which will
bring back a set of objects of different types without truncating the
sub-class fields. Therefore it's a big problem for persistance
frameworks that use inheritance.

> I can only tell you, what will be an improvement for me as
> a persistance framework programmer and will not help me.
>
> What will not help me:
>
> * that the database generates OID
>
> * that the database generates "clsss" OID (one may want to
> have that, because to recognize which table the data
> comes from..)

You don't seem to be thinking much in terms of an Object Data Management
Group style persistence framework. That's a shame since it's becoming
increasingly important. Sun seems to be endorsing it for Java in some
way too.

>
> * special features to solve very special problems
>
> What will help me:
>
> * all the stuff to reduce the number (!) of queries send
> to database to get my data
>
> * a way to insert VERY quickly a larger amount of data
> into a table.
>
> * a good, speedy database
>
> Marten

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2000-10-23 21:22:34 Re: Navigating time-warps in the CVS tree (was re the rule system)
Previous Message Krzysztof Kowalczyk 2000-10-23 20:44:17 Mailing list archives available?