Unexpected behavior (negations?)

[iowaguy]

I could use some help understanding some behavior I'm seeing. Given the facts and rule in the snippet below, I would expect to see an output where any value of x in the output must be returned by relation G, but not relation E.

Concretely, given the data below, I am expecting not to see any v=6 or x=1 in the output, because I would think they would be ruled out by the E facts.

Any explanation on why I'm seeing what I'm seeing would be very helpful!

.output Test
.decl Test(v: number, x: number, y: number, z: number)
Test2(v, x, y, z) :-
   G(x, v, z),
   // expected results to this point: v=2,6; x=1,5; y=; z=4,8;

   D(_, y),
   // expected results to this point: v=2,6; x=1,5; y=2,3,4; z=4,8;

   !E(x, y, v).
   // expected results to this point: v=2; x=5; y=3,4; z=4,8;


.decl D(var: number, var2: number)
D(1, 2).
D(2, 3).
D(3, 4).

.decl E(var: number, var2: number, var3: number)
E(1, 2, 3).
E(4, 5, 6).

.decl G(var: number, var2: number, var3: number)
G(1, 2, 4).
G(5, 6, 8).

The output from Test.csv is

2   1   2   4
2   1   3   4
2   1   4   4
6   5   2   8
6   5   3   8
6   5   4   8

[quentin]

Hi, your comment is a bit misleading here, suggesting that any combination is possible:

G(x, v, z),
   // expected results to this point: v=2,6; x=1,5; y=; z=4,8;

The variable assignments can't be considered independently of each others.
Instead you have to think in tuples:

G(x, v, z)
// expected results to this point: (x, v, z) in {(1, 2, 4), (5, 6, 8)}

I invite you to continue the reasoning in terms of tuples and you will eventually find that there is not tuple in E that contradicts the Test output.

[iowaguy]

It all makes sense now, thank you!