topic Re: log in SAS Programming

log

ravindra2 — Wed, 31 Oct 2018 06:43:36 GMT

i have one dataset i want log as there is a duplicate record tn dataset and that duplicate dataset should be printed in log window.

data lb;
input usubjid lbdtc $ 16. lbtptnum rbc $ wbc $ visit;
cards;
101 01-10-2018T10:15 1 1 1 1
101 01-10-2018T10:20 2 2 2 1.01
101 01-10-2018T10:20 2 2 2 1.01
;

thank you

Re: log

PeterClemmensen — Wed, 31 Oct 2018 06:45:04 GMT

I'm not following. You have a duplicate record in your data set. What do you want to be printed in the log? The data set name? The duplicate record?

Re: log

ravindra2 — Wed, 31 Oct 2018 06:50:16 GMT

there should be a warning that 'lb dataset having duplicate record'

next task is

that duplicate record which is there in lb dataset should also have to print in log

Re: log

Kurt_Bremser — Wed, 31 Oct 2018 09:44:38 GMT

@ravindra2 wrote:

there should be a warning that 'lb dataset having duplicate record'

Since that is not a native problem condition for SAS, you have to write code that detects this, eg

proc sort
  data=have
  out=sort
  dupout=test
  nodupkey
;
by key;
run;

data _null_;
set test;
put 'Duplicate observation detected!';
stop;
run;

If no duplicates are there, the data step will stop on its own when the set statement encounters the EOF condition.

next task is

that duplicate record which is there in lb dataset should also have to print in log

Don't do that. If you apply that to a large dataset, your log file will grow out of proportion, which takes space and time. If you need documentation, use a reference to the dupout dataset (and write that to a permanent library).

Edit: added the nodupkey option in the proc sort statement

Re: log

singhsahab — Wed, 31 Oct 2018 16:47:37 GMT

data lb;
input usubjid lbdtc $ 16. lbtptnum rbc $ wbc $ visit;
cards;
101 01-10-2018T10:15 1 1 1 1
101 01-10-2018T10:20 2 2 2 1.01
101 01-10-2018T10:20 2 2 2 1.01
;
run;


proc sort data=lb dupout=want nodupkey;
by _all_;
run;

data _null_;
set want;
put _all_;
run;