Date:         Fri, 14 Mar 1997 03:55:56 GMT
Reply-To:     lpogoda@AOL.COM
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         lpogoda@AOL.COM
Organization: AOL http://www.aol.com
Subject:      Re: (no subject given)

In article <85256458.00518A43.00@njros1ngw04.metro.prudential.com>, David Heck
<david.heck@PRUDENTIAL.COM> writes:
 
>I have a small question, I  set up a table that was created in a previous
>step. I want to read the next file and select off policies only if  they
>have a matching policy number off the table. It works well , to a point. If
>there are any duplicate policies on the new file I'm reading( and there
>usually are) I miss the second one, because they are sorted by policy
>number. What am I doing wrong and is there  a better way to do this?
>Here is the code I'm trying to use. For each policy number there can be up
>to 5 logmsg fields. I'm writing out a fixed length record for each message
>and moving on to the next one. When I hit the same policy number two times
>(or more) in a row, it will only catch the first one.
 
A real kludge would be to use a dummy key value between each real lookup, that
way you'd never be looking for the same policy number two or more times in a
row.  (That particular "solution" was actually suggested to me by SAS technical
support when I called for advice in a similar situation.)
 
Really though, I think you need to rethink your logic.  You have an
observation, you've looked up your policy number.  Then you have another
observation and the policy number on this second observation is identical to
the one you just looked up.  Why would you want to look it up again?  You
already know what you'll find, or you will if you redo your logic.  Something
along the lines of
 
IF POLNUM ^= LAG(POLNUM) THEN SET NEXTC.IF952C
   KEY = POLNUM;