recode rows based on certain condition

[anandas]

Hello

I have the following dataset;

data have;
input id 1. term $8. ca 1.;
datalines;
1 199602 1
1 199603 0
1 199701 1
1 199702 1
1 199703 0
1 199802 1
2 200501 1
2 200502 1
2 200503 0
2 200601 0
2 200602 0
3 201001 1
3 201002 1
3 201003 0
3 201101 1
3 201102 1
;
run;

I want it recoded to the following;

data want;
input id 1. term $8. ca 1.;
datalines;
1 199602 1
1 199603 1
1 199701 1
1 199702 1
1 199703 1
1 199802 1
2 200501 1
2 200502 1
2 200503 0
2 200601 0
2 200602 0
3 201001 1
3 201002 1
3 201003 1
3 201101 1
3 201102 1
;
run;

Basically, i want to recode id=1 and id=3 because both those ids have 1 in the first and 1 in the last term and therefore, if there are zeros in the in-between terms, i want to recode them into one.

Please help!

Thanks

ananda

[novinosrin · Accepted solution]

data have;
input id 1. term $8. ca 1.;
datalines;
1 199602 1
1 199603 0
1 199701 1
1 199702 1
1 199703 0
1 199802 1
2 200501 1
2 200502 1
2 200503 0
2 200601 0
2 200602 0
3 201001 1
3 201002 1
3 201003 0
3 201101 1
3 201102 1
;
run;

data want;
 do _n_=1 by 1 until(last.id);
  set have;
  by id;
  if first.id then _n=ca;
 end;
 _n=ca & _n;
 do _n_=1 to _n_;
  set have;
  if _n then ca=_n;
  output;
 end;
 drop _:;
run;

[anandas]

Thank you very much for your quick response!