Index scan question

[Kenneth Brody]

Quoting Don Coleman (Fri, 25 Jul 2008 10:44:45 -0400):

[...]
> Built an index on a YESNO field where for a *report export process where
> I only want to select 11 (YESNO) eq "" (blank)
>
> There are 465,000 records in this file.   This export is to run every 3
> minutes so I would expect a maximum of 2000-3000 records (peak periods) to
> be selected for each 3 minute interval scan.  The number exported could drop
> to several hundred records or less during non-peak times.  However when I
> run the *report export process the app. still reads through 25,000-30,000
> records.  While that's not the entire file and some time has been saved, I
> don't understand why it is not only reading the exact records to be
> selected.  There is no select process involved and no "ne" statement in my
> selection set.  The selection set is simply:
>
> Group	Field		Description		REL	Value
> 	11		Exported (Y/N)	eq
[...]

That doesn't make sense.  If index scan is turned off, it would read all
465,000 records in the file.  If index scan is being used, filePro would
only read the records where field 11 is blank.

Is there any chance that there are really only about 30,000 used records
in the file, and the rest are deleted?

-- 
Read the truth behind the movie "Expelled" at <http://www.ExpelledExposed.com>
--
KenBrody at BestWeb dot net        spamtrap: <g8ymh8uf001 at sneakemail.com>
http://www.hvcomputer.com
http://www.fileProPlus.com