On 25 Oct 2001, at 11:53, Ann W. Harrison wrote:

> At 11:43 AM 10/25/2001 -0400, Paul Schmidt wrote:
>
> >The basic question is, is there an SQL standard way of doing the same
> >thing? If there is then it should be adopted...
>
> Yes there is and that's what we're doing.

Okay, so that standard gives you two options, standard SQL table
names, i.e. JOE_BLOUGH = "joe_blough" or "JOE BLOUGH" <>
"joe blough", and that is as far as the standard goes. Since I
personally use the first form, it doesn't matter.

> > Otherwise,
> >sure why not "Joe Blough" <> "JOE BLOUGH" but [Joe Blough] =
> >[JOE BLOUGH]
>
> Arrrg! Adding one type of delimited identifiers broke damn
> near everything. Do we really need another one?

Probably not, but then I don't port non-standard stuff, Woody does.
If it's that hard to fix, then maybe we should forget it.

> >However it should happen for dialect 5 (they went
> >from 1 to 3, so I assume the next one is 5).
>
> Dialect 2 does exists - it accepts dialect 1 and returns
> errors if anything in it conflicts with dialect 3.
>

Okay, so D2 is used for engine debugging, maybe D4 will go the
same way....

Paul
Paul Schmidt
Tricat Technologies
paul@...
www.tricattechnologies.com