Re: pgsql: Fix assorted missing infrastructure for ON CONFLICT. - Mailing list pgsql-committers

Alvaro Herrera <alvherre@2ndquadrant.com> writes:
> Peter Geoghegan wrote:
>> What does this mean?
>>
>> +           /* XXX broken */
>> if (attno < 0)
>> elog(ERROR, "system column in index");

> My guess is that it means we do support indexes in system columns (oid
> in particular) and that instead of an ugly error this should do
> something else.  Maybe silently ignore the index.

I left that for a second patch, which is now pushed.

            regards, tom lane