Re: non-recursive WITH clause support

"David Fetter" <david(at)fetter(dot)org> writes:

> On Mon, Apr 09, 2007 at 06:43:08PM -0400, Tom Lane wrote:
>
>> So this does not really add any new functionality, it's just variant
>> syntax for something you can do about as easily without it, right?
>
> Not totally as easily. For example, you can do some kinds of
> aggregation with a few fewer keystrokes.

I think "fewer keystrokes" is exactly what Tom meant by a variant syntax
without new functionality. That's an accurate description.

I suppose it depends in part on how important we think it is to add variant
syntaxes just because they're blessed by the ANSI standard. If this were a
syntax we were creating just for our user's convenience it would be a pretty
weak justification for an incompatibility. But if there are users who expect
this syntax to work because it's standard then it could be considered an
omission in our standards compliance.

I'm actually not too sure what the answer is. I hadn't heard of it before the
discussion about recursive queries myself.

--
Gregory Stark
EnterpriseDB http://www.enterprisedb.com