how to take some variables from raw data using infile

[thanikondharish]

given a data file with 100 observations of data, you only need to read in the first 70 columns 

[PeterClemmensen]

data have;
   do x=1 to 100;
      output;
   end;
run;

data want;
   set have(obs=70);
run;

[thanikondharish]

I want variables not observations

[PeterClemmensen]

Ah ok. Do something like this

data have;
   array SomeArray{100} x1-x100;
run;

proc sql noprint;
   select name into :varnames separated by ' '
   from dictionary.columns
   where libname="WORK" and memname="HAVE" and varnum<=70
   order by varnum;
quit;

%put &varnames.;

data want;
   set have;
   keep &varnames.;
run;

[thanikondharish]

It's good but how do we do using infile statement read only 70variables out
of 100 (is any options used like line,linesize,length

[PeterClemmensen]

What kind of data source do you want to read from with this infile statement? a .txt, .xlsx file? 

An infile statement is used to read data from an external file.

[thanikondharish]

.xlsx file

[andreas_lds]

---

@thanikondharish wrote:
.xlsx file

---

Reading a named range is possible if you setup fits - there are some many papers and posts explaining this, i don't like to repeat it. If you don't have a range, you have to import everything and drop the unwanted variables.

[PeterClemmensen]

You cannot INFILE an XLSX file in DATA step.  You need to use an engine that can read the data structure of an Excel spreadsheet.

You can however import a .csv file with the INFILE Statement.