From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | nconway(at)klamath(dot)dyndns(dot)org (Neil Conway) |
Cc: | pgsql-hackers(at)postgresql(dot)org, mojo(at)thewickedtribe(dot)net |
Subject: | Re: rules and default values |
Date: | 2002-03-31 00:26:01 |
Message-ID: | 3977.1017534361@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
nconway(at)klamath(dot)dyndns(dot)org (Neil Conway) writes:
> In other words, when the insert statement on the view is transformed by
> the rule, the "default value" columns are replaced by explicit NULL
> values (which is the default value for the columns of the pseudo-table
> created by CREATE VIEW). Is this the correct behavior?
It's correct, from the point of view of the rule rewriter, but that
doesn't make the behavior useful.
What'd make sense to me is to allow defaults to be attached to the
view columns, say by doing ALTER TABLE ADD DEFAULT on the view.
Unfortunately that won't do much in the current implementation,
because such defaults will never get applied (the planner certainly
won't see them as applicable).
Maybe inserting defaults should be the first phase of rewriting, just
before rule substitution, rather than being left to the planner as it
is now. We took it out of the parser for good reasons, but perhaps
we moved it too far downstream.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Neil Conway | 2002-03-31 01:11:13 | Re: compile bug in HEAD? |
Previous Message | Neil Conway | 2002-03-30 23:53:46 | rules and default values |