Date: Fri, 1 Sep 2006 20:45:48 -0700
Reply-To: toby989@HOTPOP.COM
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: toby989@HOTPOP.COM
Subject: Re: exist function question
In-Reply-To: <BAY123-F72DB11D89BE625F480FB2DE3C0@phx.gbl>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Hi Toby
The c dataset looks like:
_name_ f1 f2 f3
q01 1 0 0
q02 1 0 0
q07 1 0 0
q03 0 1 0
q04 0 1 0
q06 0 1 0
q05 0 0 1
and the q's are variables in the ps2004ms dataset.
And my macro that now works only partially looks like:
%macro efamac(n=,m=,p=,r=,i=);
%do ix=1 %to &n;
proc sql;
select trim(left(_name_)) into :lista separated by ' ' from c where f&ix=1;
quit;
proc corr data=psdata.ps2004ms(where=(id<200 and id not in (49, 109) and
half=0)) alpha nocorr nosimple outp=outcal(where=(_type_='STDALPHA'));
var &lista ;
run;
proc transpose data=outcal(drop=_type_ _name_) out=outcal(drop=_name_); run;
proc sql;
insert into outc set _type_='STAT', _name_='STDALPHA', &m&n&p&r=(select distinct
col1 from outcal);
quit;
%end;
data y;
%sysfunc(ifc(&i=1, set outc, merge y outc));
attrib _all_ label='';
run;
%mend;
data _null_;
i=1;
do n=' 3';
do m='a';
do p='a';
do r='n', 'q', 'v';
call execute('%efamac(n='||n||',m='||m||',p='||p||',r='||r||',i='||i||')');
i=i+1;
end;
end;
end;
end;
run;
Some of the log file: The issue is where my macro variable suddenly becomes
listed twice, and three times, and 4 times if I my 'to' value of the loop would
be 4 and so on.
trim(left(_name_)) into :lista separated by ' ' from c where f3=1;
17 +
quit;
NOTE: PROCEDURE SQL used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds
17 +
proc corr data=psdata.ps2004ms(where=(id<200 and id not in (49, 109) and half=0))
18 + alpha nocorr nosimple outp=outcal(where=(_type_='STDALPHA')); var &lista
&lista &lista ;
run; proc transpose data=outcal(drop=_type_ _name_) out=outcal(drop=_name_);
run; proc sql; insert
into outc set _type_='STAT', _name_='STDALPHA', a3an=(select
WARNING: Variable q10 already exists on file WORK.OUTCAL.
WARNING: Variable q18 already exists on file WORK.OUTCAL.
WARNING: Variable q20 already exists on file WORK.OUTCAL.
WARNING: Variable q23 already exists on file WORK.OUTCAL.
WARNING: Variable q26 already exists on file WORK.OUTCAL.
WARNING: Variable q31 already exists on file WORK.OUTCAL.
WARNING: Variable q10 already exists on file WORK.OUTCAL.
WARNING: Variable q18 already exists on file WORK.OUTCAL.
WARNING: Variable q20 already exists on file WORK.OUTCAL.
WARNING: Variable q23 already exists on file WORK.OUTCAL.
WARNING: Variable q26 already exists on file WORK.OUTCAL.
WARNING: Variable q31 already exists on file WORK.OUTCAL.
WARNING: Cronbach's Alpha has been computed with missing values present. Using
the NOMISS option is
recommended.
NOTE: The data set WORK.OUTCAL has 1 observations and 8 variables.
NOTE: PROCEDURE CORR used (Total process time):
real time 0.15 seconds
cpu time 0.15 seconds
NOTE: There were 1 observations read from the data set WORK.OUTCAL.
NOTE: The data set WORK.OUTCAL has 6 observations and 1 variables.
NOTE: PROCEDURE TRANSPOSE used (Total process time):
real time 0.03 seconds
cpu time 0.03 seconds
19 + distinct col1 from outcal);
NOTE: 1 row was inserted into WORK.OUTC.
19 + quit;
NOTE: PROCEDURE SQL used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds
19 + data y; set outc; attrib _all_ label='';
run;
NOTE: There were 37 observations read from the data set WORK.OUTC.
NOTE: The data set WORK.Y has 37 observations and 3 variables.
NOTE: DATA statement used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds
Toby
toby dunn wrote:
> Toby ,
>
> I would look at your macro code. I tested the code out in two different
> macro set ups and it always seems to work just fine for me.
>
>
>
> Toby Dunn
>
> When everything is coming at you all at once, your in the wrong lane.
>
> A truly happy person is someone who can smile and enjoy the scenery on a
> detour.
>
>
>
>
>
> From: toby989@HOTPOP.COM
> Reply-To: toby989@HOTPOP.COM
> To: SAS-L@LISTSERV.UGA.EDU
> Subject: Re: exist function question
> Date: Thu, 31 Aug 2006 19:47:57 -0700
>
> Toby
>
> The error, I know, just for testing purposes, to see what the
> sysfunc(exist(y))
> returns.
>
> Your example works on my side too.
>
> ....
>
> Can it be, since my thing is part of a macro, that when I compile the macro
> the
> sysfuncs are evaluated and the datastep is created/set in stone. The
> datastep is
> not dynamically recreated everytime I call the macro......tested it, seems
> not
> to be the case.
>
> Toby
>
>
>
> toby dunn wrote:
>
>>Toby ,
>>
>>data y ;
>>%sysfunc(exist(y)) ;
>>run ;
>>
>>
>>Is always gonna give you an error, why well consider what the resulting
>>data
>>step code will look like:
>>
>>If Data Set Y exists:
>>
>>Data Y ;
>>1
>>Run ;
>>
>>And If it doesnt exists:
>>
>>Data Y ;
>>0
>>Run ;
>>
>>Either way you just have a number sitting out there all alone in the night
>>with no context to help SAS figure out what you are doing.
>>
>>
>>
>>data y;
>>%sysfunc(ifc(%sysfunc(exist(y)),merge y outcfa,set outcfa));
>>run;
>>
>>Works just as Howard advertised it too work just read your log:
>>
>>NOTE: There were 34 observations read from the data set WORK.Y.
>>NOTE: There were 34 observations read from the data set WORK.OUTCFA.
>>NOTE: The data set WORK.Y has 34 observations and 3 variables.
>>NOTE: DATA statement used (Total process time):
>> real time 0.01 seconds
>> cpu time 0.01 seconds
>>
>>Notice that it found data set Y and reads observation from both and merges
>>them however you have no by statement. Which may or may not cause you
>>problems.
>>
>>
>>Data Y ;
>>X = 'A' ;
>>Y = '1' ;
>>Output ;
>>
>>X = 'B' ;
>>Y = '2' ;
>>Output ;
>>
>>X = 'C' ;
>>Y = '3' ;
>>Output ;
>>
>>Run ;
>>
>>
>>Data X ;
>>X = 'A' ;
>>Z = '999' ;
>>Output ;
>>
>>X = 'C' ;
>>Z = '999' ;
>>Output ;
>>
>>Run ;
>>
>>
>>Data Y ;
>>%SysFunc( IfC( %SysFunc( Exist( Y ) ) , Merge Y X %Str( ; ) By X %Str( ; )
>>,
>>Set X %Str( ; ) ) ) ;
>>Run ;
>>
>>Proc Print
>>Data = Y ;
>>Run ;
>>
>>
>>
>>Toby Dunn
>>
>>When everything is coming at you all at once, your in the wrong lane.
>>
>>A truly happy person is someone who can smile and enjoy the scenery on a
>>detour.
>>
>>
>>
>>
>>
>>From: toby989@HOTPOP.COM
>>Reply-To: toby989@HOTPOP.COM
>>To: SAS-L@LISTSERV.UGA.EDU
>>Subject: Re: exist function question
>>Date: Thu, 31 Aug 2006 16:11:13 -0700
>>
>>Howard Schreier <hs AT dc-sug DOT org> wrote:
>>
>>
>>>On Wed, 30 Aug 2006 14:40:13 -0700, toby989@HOTPOP.COM wrote:
>>>
>>>
>>>
>>>
>>>>do I miss somehting here?
>>>
>>>
>>>Yes, if you were expecting a different outcome.
>>>
>>>Remember that the DATA step examines the headers of all input data sets
>>>(those named on SET, MERGE, UPDATE, and MODIFY statements) at compile
>>>time.
>>>
>>>Try this:
>>>
>>> data y;
>>> %sysfunc(ifc(%sysfunc(exist(y)),merge y outcfa,set outcfa));
>>> run;
>>>
>>
>>Thank you. This is very helpful. the %sysfunc(exist(y)) returns '1' as can
>>be
>>seen below. But the %sysfunc(ifc(...)) does not take it somehow.
>>
>>76 data y;
>>77 %sysfunc(exist(y));
>>NOTE: Line generated by the macro function "SYSFUNC".
>>1 1
>> -
>> 180
>>ERROR 180-322: Statement is not valid or it is used out of proper order.
>>
>>78 run;
>>
>>NOTE: The SAS System stopped processing this step because of errors.
>>WARNING: The data set WORK.Y may be incomplete. When this step was stopped
>>there were 0
>> observations and 0 variables.
>>WARNING: Data set WORK.Y was not replaced because this step was stopped.
>>NOTE: DATA statement used (Total process time):
>> real time 0.01 seconds
>> cpu time 0.01 seconds
>>
>>
>>
>>
>>It only performs the 'set' command.
>>
>>79 data y;
>>80 %sysfunc(ifc(%sysfunc(exist(y)),merge y outcfa,set outcfa));
>>81 run;
>>
>>NOTE: There were 34 observations read from the data set WORK.Y.
>>NOTE: There were 34 observations read from the data set WORK.OUTCFA.
>>NOTE: The data set WORK.Y has 34 observations and 3 variables.
>>NOTE: DATA statement used (Total process time):
>> real time 0.01 seconds
>> cpu time 0.01 seconds
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>>>data y; if (exist('y')) then merge y outcfa; else set outcfa; run;
>>>>ERROR: File WORK.Y.DATA does not exist.
>>>>
>>>>NOTE: The SAS System stopped processing this step because of errors.
>>>>WARNING: The data set WORK.Y may be incomplete. When this step was
>>>>stopped
>>>>there were 0
>>>> observations and 3 variables.
>>>>NOTE: DATA statement used (Total process time):
>>>> real time 0.01 seconds
>>>> cpu time 0.01 seconds
| 
