regex example showing one possible way of dealing with "unstructured-like" data within SAS.

regex.sas

data test;
input a $1-200;
cards;
Organisation trades in a high risk country China, [JURIS: CHN] and is
also involed in charity [INDUSTRY: CHARITY]. Customers acquisition
date is [ACQUISITION_DATE: 10JUL1999]
Customer A is a [JOB: SOLE TRADER]
;
run;

data test1;
    set test;
format testing1-testing10 $100.;
drop re start stop position length i;
array testing{10} testing1-testing10;
re = prxparse('/\[([\w\s:-]+)\]/');
start = 1;
    stop = length(a);

call prxnext(re, start, stop, a, position, length);
i = 0;
do while (position > 0);
  i= i+1;
         testing{i} = substr(a, position+1, length-2);
/*shifted by 1 for the first bracket, -1 to undo the shift and -1 to
remove last bracket*/
         *put found= position= length=;
         call prxnext(re, start, stop, a, position, length);
      end;
 output;
run;

proc sort data = test1;
    by a;
run;

proc transpose data = test1 out = test2;
var testing1-testing3;
    by a;
run;

data test3;
    set test2;
re = prxparse("/(.*?):(.*)/");
drop re name_re col1;
name_re = prxmatch(re,COL1);
field_name = prxposn(re,1,COL1);
names = strip(prxposn(re,2,COL1));
run;

/*re-transpose*/

proc transpose data = test3 out = test4 (drop = _NAME_);
   by a;
   id field_name;
   var names;
run;