hi ... neat use of _FILE_ ... another variation (I think this works for any set of variable names)

use a different data set (SASHELP.ZIPCODE) use CALL MISSING to reduce the text when PUT _NUMERIC_ and PUT _CHARACTER_ are used insert the commas using the PUT _NUMERIC_ and PUT _CHARACTER_ statements use SUBSTR to get rid of a leading comma use COMPRESS to keep COMMAS, UNDERSCORES, NUMBERS, LETTERS use the extra PUT to clear _FILE_

filename nosee dummy;

data _null_; file nosee; set sashelp.zipcode (obs=1); call missing(of _all_); put (_numeric_ ) (= ',') @; call symput('nm',compress(substr(_file_,2),",_","kda")); put; put (_character_ ) (= ',') @; call symput('ch',compress(substr(_file_,2),",_","kda")); run;

%put &nm; %put &ch;

-- Mike Zdeb U@Albany School of Public Health One University Place Rensselaer, New York 12144-3456 P/518-402-6479 F/630-604-1475

> On Tue, 11 Nov 2008 19:23:07 -0500, Peter Zhu <peter.peterzhu@GMAIL.COM> wrote: > >>Hi, >> >>I wanted to create two sets of variables from a SAS data set. One >>group is all char variables, the other is numeric. >>Can I do that with Proc Contents and how? Or is there other way to do it? >> >>Thanks, >> >>Peter > > It appears from the later part of the thread that Peter wants > comma-separated lists in macro variables. Here's a solution with no macro > code, no arrays, and no loops: > > filename nosee dummy; > > data _null_; > if 0 then set sashelp.class; > file nosee; > put (_numeric_ ) ( = +(-2) ) +(-3) '|' +2 > (_character_) ( = +(-2) ) +(-3) ' ' @ ; > call symput( 'numlist',tranwrd(scan(_file_,1,'|'),'=',', ') ); > call symput( 'charlist',tranwrd(scan(_file_,2,'|'),'=',', ') ); > stop; > run; > > %put &numlist; > %put &charlist; > >