On Fri, Oct 27, 2017 at 6:24 AM, Robert Haas <robertmhaas@gmail.com> wrote:
> I think one of the reasons why Peter Geoghegan decided to pursue
> INSERT .. ON CONFLICT UPDATE was that, because it is non-standard SQL
> syntax, he felt free to mandate a non-standard SQL requirement, namely
> the presence of a unique index on the arbiter columns.
That's true, but what I was really insistent on, more than anything
else, was that the user would get a practical guarantee about an
insert-or-update outcome under concurrency. There could be no
"unprincipled deadlocks", nor could there be spurious unique
violations.
This is the kind of thing that the SQL standard doesn't really concern
itself with, and yet it's of significant practical importance to
users. Both Oracle and SQL Server allow these things that I
specifically set out to avoid. I think that that's mostly a good
thing, though; they do a really bad job of explaining what's what, and
don't provide for a very real need ("upsert") in some other way, but
their MERGE semantics do make sense to me.
--
Peter Geoghegan
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers