Prolog workshops

Workshop 1 - 5 are Haskell topics.

The code in workshop is guaranteed to be error-free.

Workshop 6

Some files required:

Please put them in same dir as workshop 6 source code

:- ensure_loaded(borders).
:- ensure_loaded(cities).
:- ensure_loaded(countries).
:- ensure_loaded(rivers).

    Q1 borders(X,australia).
    Q2 borders(B, france), borders(B, spain).
    Q3 borders(B, france), borders(B, spain), country(B,_,_,_,_,_,_,_).

/* Q4
    country(C) is True if C is a country 
country(A) :- 

/* Q5
    larger(C1, C2) holds when C1 is larger

        Area (sqmiles),
larger(C1, C2) :- 
    A1 > A2.

    river_country(River, Country) holds if:
    - River is a river
    - Country is a country
    - River flows into or outof Country

    country_region(Country, Region) holds if:
    - Country is a country in Region
river_country(River, Country) :- 
    river(River, L),
    member(Country, L).

country_region(Country, Region) :- 
    country(Country, Region, _,_,_,_,_,_).

Workshop 7

/* Q1
    list_of(Elt, List) holds if:
    - every element if List equals to Elt

    Works in any mode

list_of(E, [E|T]) :- 

/* Q2
    all_same(List) holds if:
    - All elements are same
all_same([E|[E|T]]) :- 

/* Q3
    adjacent(E1, E2, List) holds if:
    E1 appear immediately before E2 in List


adjacent(E1, E2, List) :- 
    append(_, [E1, E2|_], List).

/* Q4
    same predicate as Q3, but implement it recursively

adjacent_r(E1, E2, [E1,E2|_]).
adjacent_r(E1, E2, [_|T]) :-
    adjacent_r(E1, E2, T).

/* Q5
    before(E1,E2,List) holds if: 
    - E1 occurs before E2 in List

    Solve this recursively.
    If List starts with E1, then we need to make sure 
    E2 is in tail. If not, apply this to tail.
before(E1, E2, [E1|T]) :- 
before(E1, E2, [_|T]) :- 
    before(E1, E2, T).

/* Q6
    tree(L, N, R) represents a tree with label N and left
    sub-tree L and right sub-tree R. Atom empty is Leaf.

    intset_member(N, Set) holds if:
    - N is a member of Set
    *Avoid unnecessary recursion by comparing N with label of Set

    intset_insert(N, Set0, Set) where
    Set has N as a member. There should be no duplicated elements
intset_member(N, tree(_,N,_)).
intset_member(N, tree(L,N0,_)) :- 
    N < N0,
    intset_member(N, L).
intset_member(N, tree(_, N0, R)) :- 
    N >= N0,

intset_insert(N, empty, tree(empty, N, empty)).
intset_insert(N, tree(L,N,R), tree(L,N,R)).
intset_insert(N, tree(L1,N1,R1), tree(L2,N1,R1)) :-
    N < N1,
    intset_insert(N, L1, L2).
intset_insert(N, tree(L1,N1,R1), tree(L1,N1,R2)) :- 
    N >= N1,
    intset_insert(N, R1, R2).

Workshop 8

/* Q1
    Convert the code to tail recursive version
    sumList([], 0).
    sumList([N|Ns], Sum) :- 
        sumList(Ns, Sum0),
        Sum is Sum0 + N.

    Solve this by introducing an accumulator A
    When the list is empty, the sum is A
    Otherwise, Add head to A and recursively call

sumListA([N|Ns], Sum, A) :-
    Sum0 is A + N,
    sumListA(Ns, Sum, Sum0).

sumList(L, Sum) :- 
    sumListA(L, Sum, 0).

/* Q2
    a binary tree defined as:
    tree(node(Left, _, Right)) :- 
    tree_list(Tree, List) holds if:
    - List is list of all elem of tree
    - Left to Right order


tree_list(empty, []).
tree_list(node(Left,Elt,Right), List) :-
	tree_list(Left, List1),
	tree_list(Right, List2),
	append(List1, [Elt|List2], List).

/* Q3 
    Same as last question, but not use append
    Make sure the cost is proportional to number of 
    nodes in tree.

    Refer to tail recursion optimization

    It can be solved like this:
    We are basically adding the tree to accumulator to get
    the result.

    Left hand side:
        adding left tree to accumulator A where A is calculated from
        right hand side and the node of current tree

    Right hand side:
        The accumulator is current accumulator passed as argument 

tree_list_a(empty, A, A).
tree_list_a(node(Left, V, Right), List, A) :- 
    tree_list_a(Left, List, List1),
    List1 = [V|List2],
    tree_list_a(Right, List2, A).

/* Q4
    list_tree_b(List, Tree) holds if:
    - List is a list of elements in Tree
    - Tree is balanced
    Work in the mode where List is proper list

    The below solution is not efficient because
    it tries to find the appropriate list combination
    one by one
list_tree_b(List, node(Left, V, Right)) :- 
    append(List_L, List_R_V, List),
    List_R_V = [V| List_R],
    length(List_L, Len_L),
    length(List_R, Len_R),
    Len_L - Len_R =< 1,
    list_tree_b(List_L, Left),
    list_tree_b(List_R, Right).

    A better solution is to specify the length at 
    the beginning


list_tree_b_([], empty).
list_tree_b_([E|List], node(Left,Elt,Right)) :-
	length(List, Len),
	Len2 is Len // 2,
	length(Front, Len2),
	append(Front, [Elt|Back], [E|List]),
	list_tree_b_(Front, Left),
	list_tree_b_(Back, Right).

Workshop 9

/* Q1
    same_element(L1, L2) holds if:
    - Elems in L1 are in L2
    - Vice versa

    Only work in mode where both lists are ground


members([H|T], L) :- 
    member(H, L),

same_elements(L1, L2) :- 
    members(L1, L2),
    members(L2, L1).

/* Q2
    same as previous one, make it Nlog(N)
    in time complexity

    Note that prolog sort removes duplicates

same_elements_log(L1, L2) :- 
    sort(L1, Sorted),
    sort(L2, Sorted).

/* Q3
    times(W,X,Y,Z). holds if:
    - All arguments are integers
    - W*X + Y = Z
    - 0 <= Y < |W|

    Works when W and at least on of X and Z are bound to integer
    Deterministic when 3 of 4 args are ground

/* Q4 
    Solve water container problem
    two containers: 5L and 3L. The goal is to get 
    4L water in 5L container.

    fill(To) where To is the capacity of container to fill

    empty(From) where From is capacity of container to empty

    pour(From, To)


    Solve this by first define all the possible next operations and their effects
    Bring non-deteministic movement by using Prolog Variables

    It eventually reach the goal, and all the actions after that are don't cares


next(fill(3), [_,B], [3, B]).
next(fill(5), [A,_], [A, 5]). 
next(empty(3), [_,B], [0,B]).
next(empty(5), [A,_], [A,0]).

next(pour(3,5), [A,B], [C,D]) :- 
        A + B > 5 -> 
        C is A - (5-B),
        D is 5;
        C is 0,
        D is B + A

next(pour(5,3), [A,B], [C,D]) :- 
        A + B > 3 ->
        C is 3,
        D is B - (3-A);
        C is B + A,
        D is 0

container([_,4], _, _):- 
container(State, History, [Action|Actions]) :- 
    next(Action, State, State_),
    \+ member(State_, History),
    container(State_, [State_|History], Actions).

container(Action) :-
    container([0,0], [], Action).