The missing values seem to be towards the end of your records ??

Try adding a LRECL=500 (or some other large number up to 32K ) to the end of your filename statement . filename lista '/ahome/mstrv/itdvlp/pid_aggr.csv' LRECL=500 ;

Cheers Robin

-----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of rmat Sent: Saturday, 3 November 2007 1:23 a.m. To: SAS-L@LISTSERV.UGA.EDU Subject: filename statement in Proc Import

Hi SASers,

I am facing a very curious issue. I am trying to import a CSV file in UNIX environemnt using two approaches and I am confused by changes in results.

1st Approach -------------------

proc import datafile='/ahome/mstrv/itdvlp/pid_aggr.csv' out=aggr dbms=dlm replace; delimiter=','; getnames = yes; run;

This approach works fine. A section of the print out is shown below

Obs agg64 agg65 agg66 agg67 agg68 agg69 1 183 184 185 186 187 188 2 102 104 106 106 108 108 3 103 105 107 107 109 109

2nd Approach -------------------

filename lista '/ahome/mstrv/itdvlp/pid_aggr.csv';

proc import datafile=lista out=aggr dbms=dlm replace; delimiter=','; getnames = yes; run;

In this case I get certain values of one row missing (shown below). As you can see only change is how I call the external file.

Obs agg64 agg65 agg66 agg67 agg68 agg69 1 183 . . . . . 2 102 104 106 106 108 108 3 103 105 107 107 109 109

Can somebody please explain why these two approaches are giving different results?

Thanks very much, Regi