🚀 go-pugleaf

I have a access dbase with about 8 million records. A lot of the records
have the same data in certain fields. I have used Lookup fields to cut down
on the size. What I am trying to do is make it so the user can build filters
using several fields. Is there a quick way to obtain a non-duplicate list of
the data in certain fields, so i can make a list of multiple choice as the
filter is built.
Example:
A record has fileds for Company,Address,City,St,Zipcode,Phone,Fax,Category
code1,Category code2,Category code3

What I would like to do is create ComboBox controls in succession so that
when the user selects a state, it filters a combobox with zipcodes to only
display the zipcodes in that state, with no duplicates, then they select a
zipcode and it filters a combobox with cities to only display cities in that
zipcode with no duplicates etc, etc etc. Then it builds a filter using " AND
" to filter the Main dataset.

I know I can iterate through and use TDuplicates << dpIgnore, but with 8
million records this seems very time consuming.
It sure seems there has to be a better way, what am i missing?
Thanks in advance,
Albert Fowler