triggers, views and rules (not instead) - Mailing list pgsql-hackers

From Zeugswetter Andreas SARZ
Subject triggers, views and rules (not instead)
Date
Msg-id 219F68D65015D011A8E000006F8590C6010A51E8@sdexcsrv1.sd.spardat.at
Whole thread Raw
Responses Re: [HACKERS] triggers, views and rules (not instead)  (jwieck@debis.com (Jan Wieck))
List pgsql-hackers
Jan wrote:

    The only things not working for copy are rewrite rules. But I
    think  we  should  restrict rules to the view handling in the
    future and move forward by implementing  a  pure  and  really
    powerful procedural language.

Hm, it looks like you are not really a fan of the rewrite system,
eventhough you seem to have the most insight in these matters. I wonder why?

Why I like the rewrite system is:
    1. select rewrite     -- select trigger would be no good (optimizer)
    2. The client can be really dumb, like MS Access or some other
standard ODBC tool
        which does not know anything about funcs procs and the like
        (even without using passthrough)
    3. it is a lot more powerful than views
    4. it allows the optimizer to get involved (this is where triggers
fail per definition)
    5. once understood it is very easy to use
        easier than trigger with c stored procedure at least

I guess if triggers could also trigger simple select statements, I could do
most of what I want using triggers except of course the select stuff.
But as I said I like the rules system very much, especially after your
recent
fixes Jan :-) So please stick to supporting all 3: triggers, views and
rules. Wow :-)

Andreas

pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: [HACKERS] Subselects and NOTs
Next
From: Bruce Momjian
Date:
Subject: Re: [HACKERS] Another one I thought should work...