CS 3723  Programming Languages
Database Design      with Prolog

Excerpt from a database book (.pdf)
		SUP_INFO SNAME SADDR ITEM PRICE Sunshine Produce 16 River St Granola 129 Sunshine Produce 16 River St Lettuce 89 Sunshine Produce 16 River St Walnuts 109 Purity Foodstuffs 180 Industry Rd Whey 70 Purity Foodstuffs 180 Industry Rd Curds 80 Purity Foodstuffs 180 Industry Rd Granola 125 Purity Foodstuffs 180 Industry Rd Flour 65 Tasti Supply Co 17 River St Lettuce 79 Tasti Supply Co 17 River St Whey 79 Tasti Supply Co 17 River St Walnuts 119
		SUPPLIERS SNAME SADDR Sunshine Produce 16 River St Purity Foodstuffs 180 Industry Rd Tasti Supply Co 17 River St
		SUPPLIES SNAME ITEM PRICE Sunshine Produce Granola 129 Sunshine Produce Lettuce 89 Sunshine Produce Walnuts 109 Purity Foodstuffs Whey 70 Purity Foodstuffs Curds 80 Purity Foodstuffs Granola 125 Purity Foodstuffs Flour 65 Tasti Supply Co Lettuce 79 Tasti Supply Co Whey 79 Tasti Supply Co Walnuts 119

The Prolog below reproduces the situation described above. The two sets of Prolog "facts": suppliers and supplies are the same as the "proper" way to do database management in the write-up above. These are the Prolog analog to "normalized" relations in database terms. The code below illustrates that one can easily construct the "bad" relation sup_info from the two that are provided, using the Prolog equivalent of relational join. The Prolog code also illustrates how to print information. (In the code below, Prolog tries to match up the information in a rule. The the final presense of fail in the rule tells Prolog to try to match a different way. Because there are write in these rules, a side-effect of the matching is to print the data.)

In the listing below, boldface black is user input, blue is the Prolog source, and red is the output from the Prolog program. (In the final listing, I added few extra tabs by hand.)

SUPPLIES

elk01:~> cat supply2.pl /* SUPPLIERS (SNAME, SADDR) */ suppliers(sunshine_produce, a_16_river_st). suppliers(purity_foodstuffs, a_180_industry_rd). suppliers(tasti_supply_co, a_17_river_st). /* SUPPLIES (SNAME, ITEM, PRICE) */ supplies(sunshine_produce, granola, 129). supplies(sunshine_produce, lettuce, 89). supplies(sunshine_produce, walnuts, 109). supplies(purity_foodstuffs, whey, 70). supplies(purity_foodstuffs, curds, 80). supplies(purity_foodstuffs, granola, 125). supplies(purity_foodstuffs, flour, 65). supplies(tasti_supply_co, lettuce, 79). supplies(tasti_supply_co, whey, 79). supplies(tasti_supply_co, walnuts, 119). items(ITEM, SNAME, SADDR) :- supplies(SNAME, ITEM, _), suppliers(SNAME, SADDR). items_print(ITEM) :- items(ITEM, SNAME, SADDR), write(ITEM), write(' '), write(SNAME), write(' '), write(SADDR), write(' '), nl, fail. sup_info(SNAME, SADDR, ITEM, PRICE) :- suppliers(SNAME, SADDR), supplies(SNAME, ITEM, PRICE). sup_info_print(_) :- sup_info(SNAME, SADDR, ITEM, PRICE), write(SNAME), write(' \t'), write(SADDR), write(' \t'), write(ITEM), write(' \t'), write(PRICE), nl, fail.

elk01:~> prolog Welcome to SWI-Prolog blah, blah, blah ... ?- consult(supply3). % supply3 compiled 0.00 sec, 3,396 bytes ?- items(granola, SNAME, SADDR). SNAME = sunshine_produce SADDR = a_16_river_st ; SNAME = purity_foodstuffs SADDR = a_180_industry_rd ; No ?- items_print(granola). granola sunshine_produce a_16_river_st granola purity_foodstuffs a_180_industry_rd No ?- sup_info(X, Y, Z, W). X = sunshine_produce Y = a_16_river_st Z = granola W = 129 ; (7 items deleted) X = tasti_supply_co Y = a_17_river_st Z = walnuts W = 119. ?- sup_info_print(_). sunshine_produce a_16_river_st granola 129 sunshine_produce a_16_river_st lettuce 89 sunshine_produce a_16_river_st walnuts 109 purity_foodstuffs a_180_industry_rd whey 70 purity_foodstuffs a_180_industry_rd curds 80 purity_foodstuffs a_180_industry_rd granola 125 purity_foodstuffs a_180_industry_rd flour 65 tasti_supply_co a_17_river_st lettuce 79 tasti_supply_co a_17_river_st whey 79 tasti_supply_co a_17_river_st walnuts 119 false. ?- halt.