From: | Andres Freund <andres(at)anarazel(dot)de> |
---|---|
To: | Peter Geoghegan <pg(at)heroku(dot)com> |
Cc: | Pg Hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Minor ON CONFLICT related fixes |
Date: | 2015-05-12 02:11:24 |
Message-ID: | 20150512021124.GV12950@alap3.anarazel.de |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
HI,
Don't have time to go through this in depth.
On 2015-05-11 18:53:22 -0700, Peter Geoghegan wrote:
> Note that the patch proposes to de-support CREATE RULE with an
> alternative action involving ON CONFLICT DO UPDATE. Such a rule seems
> particularly questionable to me, and I'm pretty sure it was understood
> that only ON CONFLICT DO NOTHING should be supported as an action for
> a rule (recall that INSERT statements with ON CONFLICT can, in
> general, never target relations with rules). At least, I believe
> Heikki said that.
> Deparsing with an inference clause is now correctly supported. However,
> user-defined rules with ON CONFLICT DO UPDATE are now formally
> disallowed/unsupported. It seemed there would be significant complexity
> involved in making this work correctly with the EXCLUDED.*
> pseudo-relation, which was not deemed worthwhile. Such a user-defined
> rule seems very questionable anyway.
Do I understand correctly that you cut this out because there
essentially was a ruleutils bug with with EXCLUDED? If so, I don't find
that acceptable. I'm pretty strictly convincded that independent of rule
support, we shouldn't add things to insert queries that get_query_def
can't deparse.
Greetings,
Andres Freund
From | Date | Subject | |
---|---|---|---|
Next Message | Peter Geoghegan | 2015-05-12 02:22:47 | Re: Minor ON CONFLICT related fixes |
Previous Message | Peter Geoghegan | 2015-05-12 02:02:52 | Re: Minor ON CONFLICT related fixes |