Date:         Sat, 29 Mar 2008 17:32:09 -0400
Reply-To:     Sigurd Hermansen <HERMANS1@WESTAT.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Sigurd Hermansen <HERMANS1@WESTAT.COM>
Subject:      Re: Distinct on proc sql -- Clarification
Comments: To: "Howard Schreier <hs AT dc-sug DOT org>"
          <schreier.junk.mail@GMAIL.COM>
Content-Type: text/plain; charset="iso-8859-1"

Howard:
Isn't product the key in this case?
S

________________________________

From: owner-sas-l@listserv.uga.edu on behalf of Howard Schreier <hs AT dc-sug DOT org>
Sent: Sat 3/29/2008 5:04 PM
To: SAS-L@LISTSERV.UGA.EDU
Subject: Re: Distinct on proc sql -- Clarification



On Fri, 28 Mar 2008 13:43:02 -0400, Sigurd Hermansen <HERMANS1@WESTAT.COM>
wrote:

>Suren:
>You are describing the behaviour of SAS PROC SORT nodupkey ... It
>arbitrarily takes the first instance of the value of a key in the obs
>sequence of a SAS dataset.

The B/3 row is a counterexample to the NODUPKEY model. It's not at all
obvious to me what the rule is for picking an observation when PRODUCT repeats.

>S
>
>-----Original Message-----
>From: owner-sas-l@listserv.uga.edu [mailto:owner-sas-l@listserv.uga.edu]
>On Behalf Of Suren gc
>Sent: Friday, March 28, 2008 1:34 PM
>To: SAS-L@LISTSERV.UGA.EDU
>Subject: Distinct on proc sql -- Clarification
>
>
>I want to select distinct variables from first column in the select
>clause, however, it doesnot matter whether variables of other columns
>are duplicated. Would this possible?
>Data:
>product Num
>A 1
>A 2
>B 2
>B 3
>C 2
>D 3
>A 2
>B 4
>
>I would want to select distinct products and create a dataset like this:
>A 1 B 3 C 2 D 3
>
>Thanks,
>Suren