Date:         Thu, 19 Apr 2001 11:29:53 -0700
Reply-To:     "Huang, Ya" <ya.huang@AGOURON.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Huang, Ya" <ya.huang@AGOURON.COM>
Subject:      A fundamental data step questions
Content-Type: multipart/alternative;

Hi there,

This question is for those data step gurus.

I always feel that data step programming is the toughest part
of SAS programming. Most time I just give up and turn to those
PROCs, especially PROC SQL. I feel that it is now the time for
me to take a close look at some fundamental (maybe I should
say advanced?) data step techniques, just for the sake of writing
more efficient program.

Following is my sample code and the logs:


1    data xx;
2    do i=1 to 5;
3    output;
4    end;
5

NOTE: The data set WORK.XX has 5 observations and 1 variables.
NOTE: DATA statement used:
      real time           1.30 seconds
      cpu time            0.04 seconds


6    data _null_;
7     set xx;
8     put _all_;
9     set xx;
10    put _all_;
11    stop;
12    run;

i=1 _ERROR_=0 _N_=1
i=1 _ERROR_=0 _N_=1
NOTE: There were 1 observations read from the dataset WORK.XX.
NOTE: There were 1 observations read from the dataset WORK.XX.
NOTE: DATA statement used:
      real time           0.02 seconds
      cpu time            0.02 seconds


13
14   data _null_;
15    do k=1 to 2;
16    set xx;
17    put _all_;
18    end;
19    stop;
20    run;

k=1 i=1 _ERROR_=0 _N_=1
k=2 i=2 _ERROR_=0 _N_=1
NOTE: There were 2 observations read from the dataset WORK.XX.
NOTE: DATA statement used:
      real time           0.02 seconds
      cpu time            0.02 seconds



Now my question is that why the two set statement in the
first data step read the same record from xx, but if I
put it in a do loop as it in the second data step, then
it read different record? For me they look same, do loop
is just a compact writing.


Thanks for explanation.


Ya Huang



[text/html]