FreeTDS Development Group

[Brian Bruns <brian.bruns AT gmail.com>]

Of course, it goes without saying that you'll eventually run into
situtations where you can only get an approximate match of the data
you need due to external factors (merging with external data,
calculations that can't be done by the server), in which case you'll
need to return the smallest set you can containing your data and do as
the loop did in this case.

Brian


On Thu, 10 Mar 2005 13:39:25 -0500, Lowden, James K
<LowdenJK AT bernstein.com> wrote:
> > From: craigs
> > Sent: Thursday, March 10, 2005 7:03 AM
> >
> > i want to exit query when i get what i want,
> > theres over 2.4 millions records in my table so really need to get
> > this right, can someone advise me if the following code would work.
> 
> Bill is helping you get your dbcancel() code correct, and maybe some
> other aspects of your application in the bargain.  I want to emphasize a
> point he made, because it may help eventually (or sooner).  You didn't
> ask for design advice, but even it's no use to you, it might help
> someone else.
> 
> Your table has 2.4 million rows.  Based on some criterion, you want to
> pick out one and discard the others.  That's the genesis of wanting to
> understand dbcancel() etc.
> 
> You're not the first person to have this challenge, nor the first to try
> to approach it this way.  It won't work very well, though.  You'll
> require the server to send a great many rows over the wire that you
> don't need, wasting server, network, and client resources.  In the worst
> case, you'll need row number 2,399,999.  If every row is 100 bytes, the
> server would find and send 240 MB for the 100 you want.  On average, the
> row you want will be midway through the result set, wasting 120 MB per
> query.  It will be roughly 1,000,000 times slower than necessary.
> 
> The right way to do it is to tell the server what you want: write a
> specific query for the row you're looking for.  The server will search
> the data locally (much faster) send you the one fine row.  You'll call
> dbnextrow() exactly twice: once to get the row and once to confirm there
> are NO_MORE_ROWS.  As Bill suggested, a nice refinement is to package
> the SQL in a stored procedure and call that with your parameter(s).
> Send the query as "execute my_proc @my_param" or, better still, use the
> dbrpc functions.
> 
> If you formulate your query to return only what you want, the networking
> and cancellation issues go away, and you can focus on your application.
> 
> HTH.
> 
> --jkl
> 
> -----------------------------------------
> The information contained in this transmission may contain privileged and 
> confidential information and is intended only for the use of the person(s) 
> named above. If you are not the intended recipient, or an employee or agent 
> responsible for delivering this message to the intended recipient, any 
> review, dissemination, distribution or duplication of this communication is 
> strictly prohibited. If you are not the intended recipient, please contact 
> the sender immediately by reply e-mail and destroy all copies of the 
> original message. Please note that we do not accept account orders and/or 
> instructions by e-mail, and therefore will not be responsible for carrying 
> out such orders and/or instructions.
> If you, as the intended recipient of this message, the purpose of which is 
> to inform and update our clients, prospects and consultants of developments 
> relating to our services and products, would not like to receive further 
> e-mail correspondence from the sender, please "reply" to the sender 
> indicating your wishes.  In the U.S.: 1345 Avenue of the Americas, New 
> York, NY 10105.
> 
> _______________________________________________
> FreeTDS mailing list
> FreeTDS AT lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/freetds
>