Date:         Mon, 11 Mar 2002 12:59:57 -0500
Reply-To:     "Dorfman, Paul" <Paul.Dorfman@BCBSFL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Dorfman, Paul" <Paul.Dorfman@BCBSFL.COM>
Subject:      Re: ERROR: Required operator not found in expression
Comments: To: "Cacialli, Doug" <Doug_Cacialli@URMC.ROCHESTER.EDU>
Content-Type: text/plain; charset=iso-8859-1

Doug,

You are trying to make a Data step function NMISS() return a macro value in
a macro conditional sentence, where it could be tested. To the macro
processor, NMISS is just text, for it does not begin with a macro trigger.
However, this:

%sysfunc(nmiss(&nme_lstcsv))

is a directive to execute NMISS(), convert the result to a character string,
and move it to the macro memory, where it can be tested implicitly using the
%eval() macro function. In a macro expression, you should always use
%sysfunc() if you need a Data step function to return a value. %sysfunc()
functions (but not their arguments!) can be nested. RTFM on it.

Kind regards,
=======================
Paul M. Dorfman
Jacksonville, FL
=======================

> -----Original Message-----
> From: Cacialli, Doug [mailto:Doug_Cacialli@URMC.ROCHESTER.EDU]
> Sent: Monday, March 11, 2002 12:20 PM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: ERROR: Required operator not found in expression
>
>
> Y'all,
>
> Let me start off by saying that I've been using SAS less than
> six months and
> I am really just starting to use macros in a meaningful way.
> I was hoping
> someone could elaborate on this error message SAS keeps
> kicking back to me.
> I've included the macro below.
>
> 592  %macro dolooping;
> 593      %if nmiss(&nme_lstcsv)>0 %then %do;
> 594          data affdx;
> 595              update affdx test_merge;
> 596              by id;
> 597          n_var = n(&nme_lstcsv) + 1;
> 598              array date_cng (*) &nme_lstspc;
> 599                  do i=1 to hbound(date_cng) until (i = n_var);
> 600                      if date_cng(i) = . then date_cng(i) = datea;
> 601                  end;
> 602          drop datea;
> 603      %end;
> 604
> 605      %else %if nmiss(&nme_lstcsv)=0 %then %do;
> 606          proc contents data = affdx noprint out = recompute;
> 607
> 608          proc sql feedback noprint;
> 609              select distinct name into :lst_recompute
> separated ','
> 610              from recompute;
> 611              where name like 'datea%';
> 612          quit;
> 613          %put &lst_recompute;
> 614
> 615          data symputit;
> 616              set affdx;
> 617              by id;
> 618          call symput ("maxvars", (left(n(&lst_recompute)+1)));
> 619
> 620          data affdx;
> 621              update affdx test_merge;
> 622              by id;
> 623          datea&maxvars = datea;
> 624      %end;
> 625  %mend dolooping;
> 626
> 627  %dolooping
> SYMBOLGEN:  Macro variable NME_LSTCSV resolves to
>
> datea01,datea02,datea03,datea04,datea05,datea06,datea07,datea0
> 8,datea09,date
> a10,date
>
> a11,datea12,datea13,datea14,datea15,datea16,datea17,datea18
> ERROR: Required operator not found in expression: nmiss(&nme_lstcsv)>0
> ERROR: The macro DOLOOPING will stop executing.
> 628  run;
>
> I apologize if this is really poorly written.  Thank you for
> any suggestions
> you might have!
>
> Regards,
>
> Doug Cacialli
>
> ----------------------------------------------------------------------
> Doug Cacialli - Information Analyst / Laboratory Coordinator
> Sleep Research Laboratory
> Department of Psychiatry
> URMC - Strong Memorial Hospital
> 300 Crittenden Blvd.
> Rochester, New York 14642
> Phone - (585) 275-2900    Fax - (585) 273-3682
> ----------------------------------------------------------------------
>
>


Blue Cross Blue Shield of Florida, Inc., and its subsidiary and
affiliate companies are not responsible for errors or omissions in this e-mail message. Any personal comments made in this e-mail do not reflect the views of Blue Cross Blue Shield of Florida, Inc.