Date:         Wed, 16 Sep 1998 22:43:55 -0500
Reply-To:     Gaylen Fraley <gfraley@EARTHLINK.NET>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Gaylen Fraley <gfraley@EARTHLINK.NET>
Organization: EarthLink Network, Inc.
Subject:      Re: SCL and Macro variables.

One point to remember, though, is that the Code in a submit block, is not
resolved at compile time.  This code is more dynamic and SAS will not
resolve the &vars until runtime.  In his example, the %INCLUDE is not
subject to resolution.  He could change the code in the %INCLUDE as often as
he likes and never have to recompile the code.  If there is not a
corresponding SCL variable then there shouldn't be any problem at all.  This
is more than likely an environmental issue, local or global.
 
Gaylen
 
lpogoda@aol.com wrote in message <6tpnjo$cth$1@nnrp1.dejanews.com>...
>You don't give much information here.  An easy way for macro variables to
>trip you up in SCL is to forget that macro variable values are resolved AT
>COMPILE time in SCL just as they are in regular data step code. For SCL,
>that's when you say "compile", while data steps normally compile just
before
>they are run.  The other point is that the common way of referencing macro
>variables (&varname) doesn't work too well in SCL submit blocks, since SCL
>tries to substitute the value of an SCL program variable for &varname
before
>the code get submitted.
>
>-----== Posted via Deja News, The Leader in Internet Discussion ==-----
>http://www.dejanews.com/rg_mkgrp.xp   Create Your Own Free Member Forum