Re: [HACKERS] DISTINCT and ORDER BY bug?

At 12:12 AM 2/7/00 -0600, Taral wrote:

>The thing here is that random() is not deterministic on its inputs,
>whereas sin() is. Perhaps we should only allow fully deterministic ORDER
>BY? (Ugh, another flag for functions...)

Which, by it's nature is probably a misnomer, because I imagine that
PL/pgSQL functions would always have to be non deterministic whatever
their inputs? Given that unrecognized syntax is just tossed the
query executor. Thus calling any 'ole function without PL/pgSQL
really knowing what's going on?

So you probably end up with a LIST of functions by name that are built-in
and deterministic.

Or ... you simply say that results are really weird if the function has
undeterministic behavior and document it.

Tom's on the right path asking what the standard might say and what
delphic, incomprehensible answer the Oracle might have for us.

(the more I learn about the SQL standard, the more I appreciate the irony
of Oracle's corporate name!)

- Don Baccus, Portland OR <dhogaza(at)pacifier(dot)com>
Nature photos, on-line guides, Pacific Northwest
Rare Bird Alert Service and other goodies at
http://donb.photo.net.