If you are looking for a list of the unique values identified by a
particilar dictionary, I usually just SORT the file by that value and
BREAK-ON that value.
If your 100000 records have a dictionary called "TYPE" and there are
say... only 20 distinct types: Using ENGLISH :
SORT MY-FILE BY TYPE BREAK-ON TYPE ID-SUPP DET-SUPP
will give you a sorted list of the "distinct" TYPEs. ENGLISH and
dictionaries can do all things. You could also get an "item count" of
how many records contain each specific value for instance. I'm sure
someone here can recomend a good resource for Reality ENGLISH.