Re: UPSERT wiki page, and SQL MERGE syntax

Peter Geoghegan <pg@heroku.com> wrote:
> On Fri, Oct 10, 2014 at 11:44 AM, Kevin Grittner <kgrittn@ymail.com> wrote:

>> [discussion on committers rejecting the notion of a syntax 
>>  involving specification of an index name]

> as I've said many times, there are problems with naming
> columns directly that need to be addressed. There are corner-cases.
> Are you going to complain about those once I go implement something?

If I don't like what I see, yes.  I think it will be entirely
possible for you to write something along those lines that I won't
complain about.

> I think we can just throw an error in these edge-cases, but let's
> decide if we're comfortable with that.

If the list of columns does not allow a choice of a suitable
full-table unique index on only non-null columns, an error seems
appropriate.  What other problem cases do you see?

--
Kevin Grittner
EDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company