topic Re: Change dataset with loop count into append statement in SAS Programming

Change dataset with loop count into append statement

imdickson — Mon, 24 Sep 2018 03:34:30 GMT

TOPIC: Change dataset with loop count into append statement

I have a macro that will loop and create a new dataset with a counter behind.

Code like this:

PROC IMPORT OUT=WORK.out&i  DATAFILE= "&dir/&name" 			
		/*excelout*/			
            DBMS=csv REPLACE;					
			delimiter='09'x;		
			getnames=no;		
		RUN;		

data test&i (drop=			
		%do k=1 %to &cnt;			
		&&col&k..			
		%end;			
		);	
			
		length station $10 voltage $10 year 8 month $20 transformer $10			
		Day $20 Date Time MW_Imp MW_Exp MVAR_Imp MVAR_Exp MVA Power_Factor 8;			
		format Time hhmm.;			
		set out&i. end=last;

Currently the script will generate about 4 data sets if i have 4 external files by PROC IMPORT.

What i want is to eliminate the creation of multiple datasets but just append them into the master file. Is there a way to do so?

Re: Change dataset with loop count into append statement

PGStats — Mon, 24 Sep 2018 03:44:17 GMT

Yes, if you read your data as text files. The infile statement can accept multiple file names. The files are simply read in sequence. Check the documentation for the infile statement.

Re: Change dataset with loop count into append statement

Astounding — Mon, 24 Sep 2018 13:09:16 GMT

You can, but you still need both the PROC IMPORT and the DATA step first. For example, in PROC IMPORT, re-use the same data set:

proc import out=temp ......

Then in the DATA step, modify it:

data temp (drop= .... same as before ...,.);

set temp;

..... same as before .....;

run;

Then add it to a master data set:

proc append data=temp base=master;

run;

This change relies on the DATA step logic accurately guaranteeing the same structure for all the data sets.

Re: Change dataset with loop count into append statement

Reeza — Mon, 24 Sep 2018 15:17:21 GMT

If your files are csv with the same structure you can read it in a single step and avoid a macro which is the better approach. PROC IMPORT may truncate records so I wouldn't recommend it either in a situation like this.