Обсуждение: [HACKERS] CREATE STATISTICS statistic_type documentation

Jeff Janes <jeff.janes@gmail.com> writes:
> The docs for CREATE STATISTICS does not say what happens if the
> statistic_type clause is omitted.  It should probably say that the default
> action is to create both ndistinct and dependencies.

Hmm, I coulda sworn that it did say that somewhere.

Where would you expect to find that?
        regards, tom lane

Jeff Janes <jeff.janes@gmail.com> writes:
> On Thu, May 25, 2017 at 9:28 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Where would you expect to find that?

> Probably at the end of the paragraph:
> "A statistic type to be computed in this statistics object. Currently
> supported types are ndistinct, which enables n-distinct statistics, and
> dependencies, which enables functional dependency statistics. For more
> information, see Section 14.2.2 and Section 68.2."

Fair enough.

> Something like "If omitted, both ndistinct and dependencies will be
> included."

> If we invent more types in the future, would we expect those to be
> defaulted to as well?

I might be wrong, but my impression is that the plan is to default
to gathering all possible stats types.  So I'd probably write the
addendum that way rather than naming those two types specifically.
        regards, tom lane

Tom Lane wrote:
> Jeff Janes <jeff.janes@gmail.com> writes:
> > On Thu, May 25, 2017 at 9:28 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:

> > If we invent more types in the future, would we expect those to be
> > defaulted to as well?
> 
> I might be wrong, but my impression is that the plan is to default
> to gathering all possible stats types.  So I'd probably write the
> addendum that way rather than naming those two types specifically.

Yes -- that plan has been hardcoded in ruleutils.c.

I'll patch.

-- 
Álvaro Herrera                https://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services