It works for me ... View more

@KalaBhairava wrote: Thanks for your replay Sir.   I don't want set  all datasets , it should be separately and macro variable should create in all datasets.   Huh?  This makes no sense.   I promise I'm not trolling.  But if English isn't your native language, do you have a colleague who can help you clearly state what you are trying to do?  I think that will help a lot in us helping you better. ... View more

@KalaBhairava wrote: Hi Every one,   i have 10 data sets (AA,BB,CC,DD,EE,FF,GG,HH) in one library and i want create macro to convert data set name into variable name (New ). for ex: for AA data set i want variable New =AA for BB data set i want variable New =BB;   please give me a solution.   That's only 8 datasets.   Define "variable".  We need more context.  What do you really need to do?     data aa bb cc dd ee ff gg hh ii jj; x=1; run; * delete the _PRODSAVAIL dataset ; proc datasets lib=work kill nowarn nolist; quit; proc sql noprint; select catx('.',libname,memname) into :datasets separated by ' ' from dictionary.tables where libname='WORK'; quit; %put &=datasets; %macro code; %let new=&word; %put &=new; %mend; %loop(&datasets)   Add https://github.com/scottbass/SAS/blob/master/Macro/loop.sas to your macro autocall library.   Change:   select catx('.',libname,memname) into :datasets separated by ' ' to select memname into :datasets separated by ' '     if you just want the member name only. ... View more

@newboy1218 wrote: Hi.. my following macro keeps on running and won't stop.. Can anyone spot any mistake there?   In my macro, I am first checking if the file lib.dataset_&dt2. exists or not. If it does then append the file name to a list. If not, then skip the date dt2, and move on to the next date and next file.   At the end, my macro variable &fileList1 should have all the existing files. For example, %put &fileList1.; (lib.dataset_20190501 lib.dataset_20190524 lib.dataset_20190602 lib.dataset_20190603 .....)   rsubmit; %macro code(d1=, d2=); %let start_date = &d1.; %let end_date = &d2.; %let dt1 = &start_date.; %do %while (%sysevalf("&dt1."d <= "&end_date."d)); %let dt1 = %sysfunc(putn("&dt1."d, date9.)); %let dt2 = &dt1.; %let dt2 = %sysfunc(putn("&dt1."d, yymmddn8.)); %if %sysfunc(exist(lib.dataset_&dt2.)) = 1 %then %do; %let fileList1 = %str(); %let thefile1 = lib.dataset_&dt2.; %let fileList1 = %str(&fileList1 &thefile1); %let dt1 = %sysfunc(intnx(day, "&dt1."d, 1, s), date9.); %end; %end; %put &fileList1.; %mend; endrsubmit; rsubmit; %code(d1 = '01MAY2019', d2 = '02JUL2019'); endrsubmit;   Macros and macro variables are just text.  Yes, we have the %eval and %sysevalf functions, but this line "worries" me.   %do %while (%sysevalf("&dt1."d <= "&end_date."d)); Why make this more complicated than it needs to be?  If it were me, I'd code it this way:   %macro get_dates(start,end); data _null_; length buffer $32767; %* adjust as required, but the max length does not really hurt performance ; ; do date="&start"d to "&end"d; buffer=catx(" ",buffer,put(date,yymmddn8.)); end; call symputx("dates",buffer,"G"); run; %mend; %get_dates(01MAY2019,02JUL2019); %put &=dates; %* this is the block of code you need to repeat for each token in the list ; %macro code; %let dataset=lib.dataset_&word; %put &=dataset; %mend; %loop(&dates); * if you want to wrap this in one uber-macro: ; %macro my_uber_macro(start,end); %* create a list of dates ; data _null_; length buffer $32767; %* adjust as required, but the max length does not really hurt performance ; ; do date="&start"d to "&end"d; buffer=catx(" ",buffer,put(date,yymmddn8.)); end; call symputx("dates",buffer,"G"); run; %* run a block of code over those dates ; %let i=1; %let word=%scan(&dates,&i,%str( )); %do %while (&word ne ); %* this is the bit of code you need to repeat for each token ; %let dataset=lib.dataset_&word; %put &=dataset; %* end of the bit of code you need to repeat ; %let i=%eval(&i+1); %let word=%scan(&dates,&i,%str( )); %end; %mend; %my_uber_macro(01MAY2019,02JUL2019);   See https://github.com/scottbass/SAS/blob/master/Macro/loop.sas for the loop macro.  Save it to your autocall macro library.   If you don't want to use that macro, then just use the concepts in that macro to parse each token in the &dates list, and call a child macro or block of code for each word in the list.  Put the block of code you need to repeat in that child macro.   You'll have to use a different approach if your list of dates won't fit into a macro variable (64K).  Well, actually, the 32K limit of the buffer variable in the data step.  However, I rarely encounter this limit in my day-to-day work.   (if you did encounter that limit, see https://github.com/scottbass/SAS/blob/master/Macro/loop_control.sas)   ... View more

call symputx ... View more

Too much crud to wade through.  I read your post three times and still don't want to touch it.  Perhaps I'm just lazy (isn't that a hallmark of a good programmer 😉 )   1) Re-post or edit your original post.  Post a source/target or "have"/"want" data steps, using datalines, with representative samples of your source data and desired target data.   2) Post your code using the "Insert SAS Code" (running man) icon, so it does not get reformatted by the forum software.   3) Test your code by cut-and-pasting it back into SAS from your post, ensuring it runs as expected and without error.   4) Succinctly state your problem.  I think you want unique subjects, but your post title is "merging", there's not a single merge in your code, you've got a proc transpose in there, and your post is just too much work for me to try to figure out.   Expend as much effort in asking your question as you expect us in answering it. ... View more