Roger Anderson <[log in to unmask]> writes:
 
> I need to find the last used "dash number" of a search key in a MASTER
data
> set given the following:
 
<snip>
 
> The so-nbr-id is of the form 99999-0000
> where 99999 represents a "base" number for a specific customer order.  The
> 0000 is a sequential "dash number" used to indicate the latest order.
 (-0000
> = the first order, -0123 = the 124th order from this customer.)
 
> My problem is this:  I need to know what the last dash number used was by
> looking up the so-nbr-id by the first five characters.  Since this is a
> master data set, it is NOT in sorted order.  The time to sort this file
would
> be not feasible given the nature of the application.
 
> Does anyone out there have a means of doing the above, or perhaps a
> recommendation of a "third party" indexing product?  TIA!!!
 
<snip>
</PLUG ALERT>
My company sells a product call IMSAM which handles this quite nicely.
I would install a sorted key on this field and then do the equivalent of
 
Find sorted-key >=  "NNNNN-9999";
 
Where "NNNNN" is the account number.  This would return the last used order
number for account "NNNNN".  If you used what we call Discrete mode, you'd
be able to get this information from the sorted key itself, without having
actually do any I/O to the data itself.
 
I'm sure that Bradmark's SuperDex level 1 has this same ability.
</END PLUG>
 
If you're really low on a budget, you could always do something with KSAM,
but you'd have to do a find on "NNNNN" +1 and then read backwards because
you can't do a "FIND LESS THAN" with KSAM.
 
Rich
______________________________________________________________________
________ Dynamic Information Systems Corp.   |
\!DISC / Rich Trapp                          | My opinions are my own.
 \    /  OMNIDEX Product Manager - IMAGE/SQL | With any luck they're
  \  /   Phone : 303 444-4000                | correct ;-)
   \/    E-mail: [log in to unmask]                |
 ----------------------------------------------------------------------