Re: RULE regression test fragility?

On 2013-10-26 11:27:19 -0400, Tom Lane wrote:
> Andres Freund <andres(at)2ndquadrant(dot)com> writes:
> > [ patch for \a\t mode in rules and sanity_check output ]
>
> Committed with some minor adjustment of the comments.

Thanks.

> >> +1 (but what are those silly parens in pg_seclabels definition?),
>
> > That's because it contain several UNION ALLs and ruleutils makes sure
> > the order is correct.
>
> That looks weird to me too, but it's surely not the fault of this patch.
> Maybe we should take a look at exactly what ruleutils is doing there.

Imo what it does looks sane - it adds parentheses whenever a child of a
set operation is a set operation again to make sure the order in which
the generated set operations are parsed/interpreted stays the same.

Now, we could probably remove that in some more cases (left is SetOp but
doesn't have an ORDER BY/LIMIT/...), but it's hard enough to figure out
when that's safe that I wouldn't bother.

Greetings,

Andres Freund

--
Andres Freund http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services