How to populate a group of variables with the non missing value for all the group

[sbopster]

Input data:

group flag
1 . 
1 . 
1 1 
1 .
2 . 
2 1 
2 . 
3 .
3 1
3 .
3 1
3 .
3 .
;

I want the following output
so for every group, if there is a flag=1 then all the flag_new should be a 1 for everyone in that group. If missing then all the values should be 0

Output desired:

group flag new_flag
1 . 1
1 . 1
1 1 1
1 . 1
2 . 0
2 . 0
2 . 0
3 . 1
3 1 1
3 . 1
3 1 1
3 . 1
3 . 1

[PaigeMiller]

Why is the output for group 2 a zero when there was one non-missing in the input?

PS: Don't type in a code box if you are not going to show us code.

[sbopster]

if I knew how to do it or had code I would type it in. I mean how else are you supposed to ask a question? thank you for your response.

[PaigeMiller]

You just did it. You typed your comment.

But why does group 2 have a zero in the output but a non-missing in the input?

[mkeintz]

data want;
  merge have   have (keep=group flag where=(flag^=.));
  by group;
  if  flag=. then flag=0;
run;

This works  as expected if (1) data are sorted by group, (2) no group has any variation in non-missing values  of flag.

[Ksharp]

data have;
input group flag;
cards;
1 . 
1 . 
1 1 
1 .
2 . 
2 . 
2 . 
3 .
3 1
3 .
3 1
3 .
3 .
;

proc sql;
create table want as
select *,max(flag)=1 as new_flag
 from have 
  group by group;
quit;