Re: pg, mysql comparison with "group by" clause

Sheesh...

I work with 2 MIT guys, and man, you guys will never admit you're wrong.
Must be something in the water down there ;)

Hey man, you know what, to each his own
(but Apostol is one of my favorites, so maybe have that in common? :)

You apparently like this shortcut, so be it.

I'll say this tho, Oracle and db2 don't do it even tho
it's in the standard and their optimizer is already doing
the right thing. That's gotta tell ya something, no?

don't look for this feature to be something you can do everywhere.
If for some reason postgres implements it, it will be the only
vendor to do so (though, this seems like a very MySQL-ish thing to do
so maybe not just Postgres)

take care,
Anthony

-----Original Message-----
From: gsstark(at)mit(dot)edu [mailto:gsstark(at)mit(dot)edu]
Sent: Friday, October 14, 2005 3:21 AM
To: Anthony Molinaro
Cc: Greg Stark; Scott Marlowe; Stephan Szabo; Rick Schumeyer;
pgsql-sql(at)postgresql(dot)org
Subject: Re: [SQL] pg, mysql comparison with "group by" clause

"Anthony Molinaro" <amolinaro(at)wgen(dot)net> writes:

> More awkward? What *you're* suggesting is more awkward. You realize
that
> right? How can syntax that is understood and accepted for years be
more
> awkward?

Well gosh, I would say that that's something only a newbie could say
about
SQL of all things...

I had a whole thing written after that but I just deleted it. I grow
tired of
this thread.

I am pretty happy to hear that the SQL standard endorsed the idea having
the
right thing happen if the primary key is present in the grouping list.
That
would be a wonderful feature for Postgres.

--
greg