Date:         Fri, 7 Jan 2005 08:30:43 +1100
Reply-To:     Scott <usenet739_yahoo_com_au@CRONKITE.CC.UGA.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Scott <usenet739_yahoo_com_au@CRONKITE.CC.UGA.EDU>
Subject:      Re: How to concatenate multiple data sets??

I agree with Toby that, conceptually, I can't see why you're wanting to
concatenate 500-600 datasets together?  Regardless of the dataset size(s),
it seems an incredible waste of disk space and processing time, and well as
redundancy of data.

If you REALLY do need to concatenate the data, I suggest:

1.  Perhaps use a data step view, so that your data is logically
concatenated, rather than physically concatenated.

2.  If your data sets have the same structure, consider using PROC APPEND
instead.  It will be more efficient than opening and reading each record for
500-600 datasets.

The use of SQL is just to read the metadata.  If you really want to type
500-600 data set names by hand, knock yourself out :-).  As already
mentioned, you can also access the metadata via the data step.

As an aside, one of the best procedures I ever learned was PROC SQL.  It
totally changed how I approach certain coding problems.

HTH,
Scott

"toby dunn" <tobydunn@HOTMAIL.COM> wrote in message
news:BAY101-F173CDCC7F9773085B4A227DE930@phx.gbl...
> Saqi,
>
> Can you say metadat files they will become your best friend.  Also, I am
> not
> wholly sure that appending that many files is the best thing to do unless
> you can be posotively sure that you will have enough space to write the
> thing out to the work or a permanent library.  Other wise it will bomb out
> if it becomes to large for the available resources.
>
> Sounds easy enough to write good luck.
>
>
>
> Toby Dunn
>
>
>
>
> From: Saqi <saqi2000@GMAIL.COM>
> Reply-To: Saqi <saqi2000@gmail.com>
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: How to concatenate multiple data sets??
> Date: Thu, 6 Jan 2005 14:02:38 +0000
> MIME-Version: 1.0
> Received: from mc8-f4.hotmail.com ([65.54.253.140]) by mc8-s4.hotmail.com
> with Microsoft SMTPSVC(5.0.2195.6824); Thu, 6 Jan 2005 06:02:40 -0800
> Received: from malibu.cc.uga.edu ([128.192.1.103]) by mc8-f4.hotmail.com
> with Microsoft SMTPSVC(5.0.2195.6713); Thu, 6 Jan 2005 06:02:40 -0800
> Received: from listserv.cc.uga.edu (128.192.1.75) by malibu.cc.uga.edu
> (LSMTP for Windows NT v1.1b) with SMTP id <0.00DE77C6@malibu.cc.uga.edu>;
> Thu, 6 Jan 2005 9:02:38 -0500
> Received: from LISTSERV.UGA.EDU by LISTSERV.UGA.EDU (LISTSERV-TCP/IP
> release
>          1.8d) with spool id 5147 for SAS-L@LISTSERV.UGA.EDU; Thu, 6 Jan
> 2005          09:02:38 -0500
> Received: from wproxy.gmail.com (wproxy.gmail.com [64.233.184.205]) by
>    listserv.cc.uga.edu (8.12.11/8.12.11) with ESMTP id j06E2c25028138
>    for <sas-l@listserv.uga.edu>; Thu, 6 Jan 2005 09:02:38 -0500
> Received: by wproxy.gmail.com with SMTP id 71so1723853wra for
> <sas-l@listserv.uga.edu>; Thu, 06 Jan 2005 06:02:38 -0800 (PST)
> Received: by 10.54.30.58 with SMTP id d58mr443273wrd; Thu, 06 Jan 2005
> 06:02:38          -0800 (PST)
> Received: by 10.54.32.10 with HTTP; Thu, 6 Jan 2005 06:02:38 -0800 (PST)
> X-Message-Info: diIfNwdpkLyNYPCiD34vqhmV4Wtt76yMBljaTXVTx3Y=
> DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com;
>
> h=received:message-id:date:from:reply-to:to:subject:mime-version:content-type:content-transfer-encoding;
>
> b=f2n/+heKXIRNZrNR1zrbZSbafcdM3UzIrnt6JnmfXpyVYq9DQCOojPcRD8rm8nCCFPgc2Un2S5S1KwqhanIe4KFNI8vdfMnLn38xk0fWniW6xDT3vYKKGgKfj+Tia0dnuSiwl5mQuAUqbMBHy9HvGPJkK85/ZLJ8ZNGEeadrxr4=
> Newsgroups:   bit.listserv.sas-l
> Return-Path: owner-sas-l@LISTSERV.UGA.EDU
> X-OriginalArrivalTime: 06 Jan 2005 14:02:40.0321 (UTC)
> FILETIME=[627FFB10:01C4F3F8]
>
> Hi SAS Lers,
>
> I am going to write a SAS macro and this macro will have optional
> parameters.  If parameters are specified then it is easy to extract
> the data but if parameters are not specified then it will appent all
> 500-600 datasets that are held in library products data set name
> starting with P.  The output will be inserted into a temporary data
> set.
>
> Can any one suggest the most efficientand quick way to do this??
>
>  Thanks
>
> Saqi