topic CAT, CATT, CATX, TRIM in SAS Procedures

CAT, CATT, CATX, TRIM

1800bigk — Thu, 22 Sep 2011 13:33:21 GMT

I have 4 variables that I am trying to combine into one but each thing I try is not giving me the result I want. Here are my entries 20,1,6312468,1 and I want to combine them to get 2016312468 but every time I try something I get 20 space 1 space 6312468 space 1. I have tried cat, catt, catx and trim each one then use !! to combine and each time it returns the spaces. Any thoughts?

CAT, CATT, CATX, TRIM

NN — Thu, 22 Sep 2011 13:41:00 GMT

would compress keeping your numeric values not work?

compress(column,'0123456789','k')

CAT, CATT, CATX, TRIM

art297 — Thu, 22 Sep 2011 13:50:59 GMT

Use strip rather than trim. E.g.,

data have;

w=20;

x=1;

y=6312468;

z=1;

want=cat(strip(w),

strip(x),

strip(y),

strip(z));

run;

CAT, CATT, CATX, TRIM

1800bigk — Thu, 22 Sep 2011 14:03:37 GMT

Both of these work thanks for the fast response!

CAT, CATT, CATX, TRIM

data_null__ — Thu, 22 Sep 2011 14:17:39 GMT

How about CATS?

CAT, CATT, CATX, TRIM

Peter_C — Thu, 22 Sep 2011 16:36:47 GMT

data_null_; is providing the correct answer, because CATS() performs the STRIP() and the concatenation

e.g.

want = cats( of w x y z );

(not even those commas needed when the list is preceeded with OF )

CAT, CATT, CATX, TRIM

art297 — Thu, 22 Sep 2011 18:01:43 GMT

Peter, Of course data_null_ is correct .. as he virtually always is! However, the other responses were also correct. For the fun of it I ran a one trial test on 1 million identical records. I stopped at one test as I happened to like the results: my proposed method came in first (cpu: 1.01 seconds), data_null_'s second (cpu: 1.04 seconds), and your suggestion third (cpu: 1.14 seconds). Of course, on repeated tests, I would guess that any of the methods could come in at any position. Your and datanull's methods are definitely easier to type and I'd gladly award both of you correct answers if I could, but it wasn't my post.