Date:         Tue, 14 Sep 2010 13:22:49 -0700
Reply-To:     Arthur Tabachneck <art297@NETSCAPE.NET>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Arthur Tabachneck <art297@NETSCAPE.NET>
Subject:      Re: Uninitialized as error, and error as abort
Content-Type: text/plain; charset=ISO-8859-1

Andrew,

Take a look at:
http://support.sas.com/forums/thread.jspa?messageID=41900#41900

It describes an undocument option that appears to be exactly what you
are looking for, namely NOTE2ERR.

Art
----------
On Sep 13, 3:42 pm, "Andrew Z." <ahz...@gmail.com> wrote:
> On Sep 13, 12:39 pm, Reeza <fkhurs...@hotmail.com> wrote:
>
> > On Sep 13, 11:07 am, "Andrew Z." <ahz...@gmail.com> wrote:
>
> > > Is there a way to make SAS version 9.x treat uninitialized variables
> > > as errors, and to stop all statements if there is any one error (or
> > > any kind)?
>
> > How do you define uninitialized?
>
> > Does this mean you're never creating any new variables only moving
> > things around?
>
> > It seems to me at some point all variables would be uninitialized, but
> > perhaps I'm not understanding what you're doing.
>
> I mean referencing or reading (compared to assigning) a non-existent
> variable in a data step like this
>
> 12   data zip;
> 13       set sashelp.zipcode;
> 14       if statcode eq 'NY'; /* typo of statecode */
> 15   run;
>
> NOTE: Variable statcode is uninitialized.
>
> ^^^^^^
>
> SAS gives a NOTE, but I would rather have an ERROR instead.
>
> Andrew