--- In IB-Architect@egroups.com, Helen Borrie <helebor@w...> wrote:

> > >According to my book on SQL-92 standards, neither TOP nor LIMIT

is standard

> > >SQL.
> >

[snip]

> ...but SQL 3 is not a published standard, is it?

I don't follow this list on a regular basis, and actually hate to
distract it from things-far-more-interesting-than-SQL, so I'll just
say:

1. Standard SQL (SQL92, SQL99, whatever) does not have a mechanism for
limiting the number of rows returned by a query. No TOP, no LIMIT, no
SET QUERY LITE, nothing.

2. You will never see anything called "SQL 3" as a published standard
because SQL 3 was just the working name for "the next version" of SQL
standard that would follow (and in fact *replace*) SQL92. (Just as
"SQL 2" was the working name for the thing following SQL87/89 that
eventually was published as SQL92, or more formally, ISO/IEC
9075-<n>:1992.)

And, surprise, surprise, various parts of the
thing-formerly-known-as-SQL3 were indeed published as standards by ISO
at the end of 1999. Informally, the published pieces are called
SQL99; formally, they are ISO/IEC 9075-n:1999. (See MERS list for how
to (or why not to) acquire copies for all your friends.)

>
> "What's in a name? A rose by any other name would smell as sweet."

[Hmmm, "sweet" .... "SQL" ... "sweet" .... "SQL" ... I'm having
trouble with this.]

Anyway, I'm very sorry to interrupt ... I'm outta here!

db