Re: WIP fix proposal for bug #6123 - Mailing list pgsql-hackers

From Kevin Grittner
Subject Re: WIP fix proposal for bug #6123
Date
Msg-id 4E29AAA5020000250003F727@gw.wicourts.gov
Whole thread Raw
In response to Re: WIP fix proposal for bug #6123  ("Kevin Grittner" <Kevin.Grittner@wicourts.gov>)
List pgsql-hackers
"Kevin Grittner" <Kevin.Grittner@wicourts.gov> wrote:

> I believe that we can get DELETE behavior which is every bit as
> sensible as INSERT behavior with a very small change.

> I think the right thing is to throw an error if the old row for a
> BEFORE UPDATE is updated by the same transaction and the trigger
> function ultimately returns a non-NULL value.

And to make this a bit less hand-wavy, a rough patch attached.  I
expect the error message could use some word-smithing, and it could
use comments; but it seemed like something concrete might speed
things along.

-Kevin


Attachment

pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: cataloguing NOT NULL constraints
Next
From: Josh Berkus
Date:
Subject: Re: storing TZ along timestamps