I don't know what you mean by "script". Normally you call a file with SAS code in it a program. How easy it is to convert to working on different directories depends a lot on what exactly it does.
Let's just create a trivial example of a program. One that just runs a LIBNAME statement.
libname mydata 'C:\mydata';
To prepare to make it dynamic change the hard coded directory specification into a macro variable reference.
Remember that macro triggers ( & % ) are not processed inside single quotes.
%let mydir=C:\mydata;
libname mydata "&mydir.";
The next step it so convert the logic that processes one directory into a macro. You can change your macro variable into a parameter to the macro instead. And then pass in the actual directory path when you call the macro.
%macro mymacro(mydir);
libname mydata "&mydir.";
%mend mymacro;
%mymacro(C:\mydata)
Now you could just replicate that macro call 50 times and type in the other directory names.
Or if you have the list of directories in a dataset then use that to generate the macro calls.
So for example if you had a dataset named MYLIST with a variable named MYDIR you could use this data step to generate one macro call for each observation.
data _null_;
set mylist;
call execute(cats('%nrstr(mymacro)(',mydir,')'));
run;
The %NRSTR() will prevent SAS from trying the expand the macro when CALL EXECUTE pushes the code onto the stack to run after the data step. Instead the macro will be called when SAS pulls the stored lines of code back off of the stack generated by the CALL EXECUTE() statements running.
The next level of automation is to generate the list of directories automatically. Usually that is most easily done if you can access operating system commands. For example on a PC you might read the results of a DIR commmand. Or on Unix you might read the results of a ls or find command.
data mylist;
infile 'dir /b/ad c:\mydata\*' pipe truncover;
input mydir $256. ;
run;
